diff options
Diffstat (limited to 'tools/perf')
404 files changed, 82369 insertions, 0 deletions
diff --git a/tools/perf/CREDITS b/tools/perf/CREDITS new file mode 100644 index 00000000..c2ddcb3a --- /dev/null +++ b/tools/perf/CREDITS @@ -0,0 +1,30 @@ +Most of the infrastructure that 'perf' uses here has been reused +from the Git project, as of version: + + 66996ec: Sync with 1.6.2.4 + +Here is an (incomplete!) list of main contributors to those files +in util/* and elsewhere: + + Alex Riesen + Christian Couder + Dmitry Potapov + Jeff King + Johannes Schindelin + Johannes Sixt + Junio C Hamano + Linus Torvalds + Matthias Kestenholz + Michal Ostrowski + Miklos Vajna + Petr Baudis + Pierre Habouzit + René Scharfe + Samuel Tardieu + Shawn O. Pearce + Steffen Prohaska + Steve Haslam + +Thanks guys! + +The full history of the files can be found in the upstream Git commits. diff --git a/tools/perf/Documentation/Makefile b/tools/perf/Documentation/Makefile new file mode 100644 index 00000000..eb30044a --- /dev/null +++ b/tools/perf/Documentation/Makefile @@ -0,0 +1,352 @@ +include ../config/utilities.mak + +OUTPUT := ./ +ifeq ("$(origin O)", "command line") + ifneq ($(O),) + OUTPUT := $(O)/ + endif +endif + +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,$(MAN_TXT)) + +MAN_XML=$(addprefix $(OUTPUT),$(_MAN_XML)) +MAN_HTML=$(addprefix $(OUTPUT),$(_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 = $(_MAN_HTML) +_DOC_HTML+=$(patsubst %,%.html,$(ARTICLES) $(SP_ARTICLES)) +DOC_HTML=$(addprefix $(OUTPUT),$(_DOC_HTML)) + +_DOC_MAN1=$(patsubst %.txt,%.1,$(MAN1_TXT)) +_DOC_MAN5=$(patsubst %.txt,%.5,$(MAN5_TXT)) +_DOC_MAN7=$(patsubst %.txt,%.7,$(MAN7_TXT)) + +DOC_MAN1=$(addprefix $(OUTPUT),$(_DOC_MAN1)) +DOC_MAN5=$(addprefix $(OUTPUT),$(_DOC_MAN5)) +DOC_MAN7=$(addprefix $(OUTPUT),$(_DOC_MAN7)) + +# Make the path relative to DESTDIR, not prefix +ifndef DESTDIR +prefix?=$(HOME) +endif +bindir?=$(prefix)/bin +htmldir?=$(prefix)/share/doc/perf-doc +pdfdir?=$(prefix)/share/doc/perf-doc +mandir?=$(prefix)/share/man +man1dir=$(mandir)/man1 +man5dir=$(mandir)/man5 +man7dir=$(mandir)/man7 + +ASCIIDOC=asciidoc +ASCIIDOC_EXTRA = --unsafe +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 +XMLTO=xmlto +ifndef PERL_PATH + PERL_PATH = /usr/bin/perl +endif + +-include ../config.mak.autogen +-include ../config.mak + +_tmp_tool_path := $(call get-executable,$(ASCIIDOC)) +ifeq ($(_tmp_tool_path),) + missing_tools = $(ASCIIDOC) +endif + +_tmp_tool_path := $(call get-executable,$(XMLTO)) +ifeq ($(_tmp_tool_path),) + missing_tools += $(XMLTO) +endif + +# +# 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: $(OUTPUT)perf.info $(OUTPUT)perfman.info + +pdf: $(OUTPUT)user-manual.pdf + +install: install-man + +check-man-tools: +ifdef missing_tools + $(error "You need to install $(missing_tools) for man pages") +endif + +do-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-man: check-man-tools man + +try-install-man: +ifdef missing_tools + $(warning Please install $(missing_tools) to have the man pages installed) +else + $(MAKE) do-install-man +endif + +install-info: info + $(INSTALL) -d -m 755 $(DESTDIR)$(infodir) + $(INSTALL) -m 644 $(OUTPUT)perf.info $(OUTPUT)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 $(OUTPUT)user-manual.pdf $(DESTDIR)$(pdfdir) + +#install-html: html +# '$(SHELL_PATH_SQ)' ./install-webdoc.sh $(DESTDIR)$(htmldir) + +ifneq ($(MAKECMDGOALS),clean) +ifneq ($(MAKECMDGOALS),tags) +$(OUTPUT)PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE + $(QUIET_SUBDIR0)../ $(QUIET_SUBDIR1) $(OUTPUT)PERF-VERSION-FILE + +-include $(OUTPUT)PERF-VERSION-FILE +endif +endif + +# +# Determine "include::" file references in asciidoc files. +# +$(OUTPUT)doc.dep : $(wildcard *.txt) build-docdep.perl + $(QUIET_GEN)$(RM) $@+ $@ && \ + $(PERL_PATH) ./build-docdep.perl >$@+ $(QUIET_STDERR) && \ + mv $@+ $@ + +-include $(OUPTUT)doc.dep + +_cmds_txt = cmds-ancillaryinterrogators.txt \ + cmds-ancillarymanipulators.txt \ + cmds-mainporcelain.txt \ + cmds-plumbinginterrogators.txt \ + cmds-plumbingmanipulators.txt \ + cmds-synchingrepositories.txt \ + cmds-synchelpers.txt \ + cmds-purehelpers.txt \ + cmds-foreignscminterface.txt +cmds_txt=$(addprefix $(OUTPUT),$(_cmds_txt)) + +$(cmds_txt): $(OUTPUT)cmd-list.made + +$(OUTPUT)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) $(MAN_XML) $(addsuffix +,$(MAN_XML)) + $(RM) $(MAN_HTML) $(addsuffix +,$(MAN_HTML)) + $(RM) $(DOC_HTML) $(DOC_MAN1) $(DOC_MAN5) $(DOC_MAN7) + $(RM) $(OUTPUT)*.texi $(OUTPUT)*.texi+ $(OUTPUT)*.texi++ + $(RM) $(OUTPUT)perf.info $(OUTPUT)perfman.info + $(RM) $(OUTPUT)howto-index.txt $(OUTPUT)howto/*.html $(OUTPUT)doc.dep + $(RM) $(OUTPUT)technical/api-*.html $(OUTPUT)technical/api-index.txt + $(RM) $(cmds_txt) $(OUTPUT)*.made + +$(MAN_HTML): $(OUTPUT)%.html : %.txt + $(QUIET_ASCIIDOC)$(RM) $@+ $@ && \ + $(ASCIIDOC) -b xhtml11 -d manpage -f asciidoc.conf \ + $(ASCIIDOC_EXTRA) -aperf_version=$(PERF_VERSION) -o $@+ $< && \ + mv $@+ $@ + +$(OUTPUT)%.1 $(OUTPUT)%.5 $(OUTPUT)%.7 : $(OUTPUT)%.xml + $(QUIET_XMLTO)$(RM) $@ && \ + $(XMLTO) -o $(OUTPUT) -m $(MANPAGE_XSL) $(XMLTO_EXTRA) man $< + +$(OUTPUT)%.xml : %.txt + $(QUIET_ASCIIDOC)$(RM) $@+ $@ && \ + $(ASCIIDOC) -b docbook -d manpage -f asciidoc.conf \ + $(ASCIIDOC_EXTRA) -aperf_version=$(PERF_VERSION) -o $@+ $< && \ + mv $@+ $@ + +XSLT = docbook.xsl +XSLTOPTS = --xinclude --stringparam html.stylesheet docbook-xsl.css + +$(OUTPUT)user-manual.html: $(OUTPUT)user-manual.xml + $(QUIET_XSLTPROC)xsltproc $(XSLTOPTS) -o $@ $(XSLT) $< + +$(OUTPUT)perf.info: $(OUTPUT)user-manual.texi + $(QUIET_MAKEINFO)$(MAKEINFO) --no-split -o $@ $(OUTPUT)user-manual.texi + +$(OUTPUT)user-manual.texi: $(OUTPUT)user-manual.xml + $(QUIET_DB2TEXI)$(RM) $@+ $@ && \ + $(DOCBOOK2X_TEXI) $(OUTPUT)user-manual.xml --encoding=UTF-8 --to-stdout >$@++ && \ + $(PERL_PATH) fix-texi.perl <$@++ >$@+ && \ + rm $@++ && \ + mv $@+ $@ + +$(OUTPUT)user-manual.pdf: $(OUTPUT)user-manual.xml + $(QUIET_DBLATEX)$(RM) $@+ $@ && \ + $(DBLATEX) -o $@+ -p /etc/asciidoc/dblatex/asciidoc-dblatex.xsl -s /etc/asciidoc/dblatex/asciidoc-dblatex.sty $< && \ + mv $@+ $@ + +$(OUTPUT)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 $@+ $@ + +$(OUTPUT)perfman.info: $(OUTPUT)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 $@+ $@ + +# UNIMPLEMENTED +#install-webdoc : html +# '$(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/android.txt b/tools/perf/Documentation/android.txt new file mode 100644 index 00000000..8484c3a0 --- /dev/null +++ b/tools/perf/Documentation/android.txt @@ -0,0 +1,78 @@ +How to compile perf for Android +========================================= + +I. Set the Android NDK environment +------------------------------------------------ + +(a). Use the Android NDK +------------------------------------------------ +1. You need to download and install the Android Native Development Kit (NDK). +Set the NDK variable to point to the path where you installed the NDK: + export NDK=/path/to/android-ndk + +2. Set cross-compiling environment variables for NDK toolchain and sysroot. +For arm: + export NDK_TOOLCHAIN=${NDK}/toolchains/arm-linux-androideabi-4.6/prebuilt/linux-x86/bin/arm-linux-androideabi- + export NDK_SYSROOT=${NDK}/platforms/android-9/arch-arm +For x86: + export NDK_TOOLCHAIN=${NDK}/toolchains/x86-4.6/prebuilt/linux-x86/bin/i686-linux-android- + export NDK_SYSROOT=${NDK}/platforms/android-9/arch-x86 + +This method is not working for Android NDK versions up to Revision 8b. +perf uses some bionic enhancements that are not included in these NDK versions. +You can use method (b) described below instead. + +(b). Use the Android source tree +----------------------------------------------- +1. Download the master branch of the Android source tree. +Set the environment for the target you want using: + source build/envsetup.sh + lunch + +2. Build your own NDK sysroot to contain latest bionic changes and set the +NDK sysroot environment variable. + cd ${ANDROID_BUILD_TOP}/ndk +For arm: + ./build/tools/build-ndk-sysroot.sh --abi=arm + export NDK_SYSROOT=${ANDROID_BUILD_TOP}/ndk/build/platforms/android-3/arch-arm +For x86: + ./build/tools/build-ndk-sysroot.sh --abi=x86 + export NDK_SYSROOT=${ANDROID_BUILD_TOP}/ndk/build/platforms/android-3/arch-x86 + +3. Set the NDK toolchain environment variable. +For arm: + export NDK_TOOLCHAIN=${ANDROID_TOOLCHAIN}/arm-linux-androideabi- +For x86: + export NDK_TOOLCHAIN=${ANDROID_TOOLCHAIN}/i686-linux-android- + +II. Compile perf for Android +------------------------------------------------ +You need to run make with the NDK toolchain and sysroot defined above: +For arm: + make ARCH=arm CROSS_COMPILE=${NDK_TOOLCHAIN} CFLAGS="--sysroot=${NDK_SYSROOT}" +For x86: + make ARCH=x86 CROSS_COMPILE=${NDK_TOOLCHAIN} CFLAGS="--sysroot=${NDK_SYSROOT}" + +III. Install perf +----------------------------------------------- +You need to connect to your Android device/emulator using adb. +Install perf using: + adb push perf /data/perf + +If you also want to use perf-archive you need busybox tools for Android. +For installing perf-archive, you first need to replace #!/bin/bash with #!/system/bin/sh: + sed 's/#!\/bin\/bash/#!\/system\/bin\/sh/g' perf-archive >> /tmp/perf-archive + chmod +x /tmp/perf-archive + adb push /tmp/perf-archive /data/perf-archive + +IV. Environment settings for running perf +------------------------------------------------ +Some perf features need environment variables to run properly. +You need to set these before running perf on the target: + adb shell + # PERF_PAGER=cat + +IV. Run perf +------------------------------------------------ +Run perf on your device/emulator to which you previously connected using adb: + # ./data/perf diff --git a/tools/perf/Documentation/asciidoc.conf b/tools/perf/Documentation/asciidoc.conf new file mode 100644 index 00000000..356b23a4 --- /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=* +plus=+ +caret=^ +startsb=[ +endsb=] +tilde=~ + +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[] + .ft C +endif::doctype-manpage[] +| +ifdef::doctype-manpage[] + .ft +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/examples.txt b/tools/perf/Documentation/examples.txt new file mode 100644 index 00000000..77f95276 --- /dev/null +++ b/tools/perf/Documentation/examples.txt @@ -0,0 +1,225 @@ + + ------------------------------ + ****** perf by examples ****** + ------------------------------ + +[ From an e-mail by Ingo Molnar, http://lkml.org/lkml/2009/8/4/346 ] + + +First, discovery/enumeration of available counters can be done via +'perf list': + +titan:~> perf list + [...] + kmem:kmalloc [Tracepoint event] + kmem:kmem_cache_alloc [Tracepoint event] + kmem:kmalloc_node [Tracepoint event] + kmem:kmem_cache_alloc_node [Tracepoint event] + kmem:kfree [Tracepoint event] + kmem:kmem_cache_free [Tracepoint event] + kmem:mm_page_free [Tracepoint event] + kmem:mm_page_free_batched [Tracepoint event] + kmem:mm_page_alloc [Tracepoint event] + kmem:mm_page_alloc_zone_locked [Tracepoint event] + kmem:mm_page_pcpu_drain [Tracepoint event] + kmem:mm_page_alloc_extfrag [Tracepoint event] + +Then any (or all) of the above event sources can be activated and +measured. For example the page alloc/free properties of a 'hackbench +run' are: + + titan:~> perf stat -e kmem:mm_page_pcpu_drain -e kmem:mm_page_alloc + -e kmem:mm_page_free_batched -e kmem:mm_page_free ./hackbench 10 + Time: 0.575 + + Performance counter stats for './hackbench 10': + + 13857 kmem:mm_page_pcpu_drain + 27576 kmem:mm_page_alloc + 6025 kmem:mm_page_free_batched + 20934 kmem:mm_page_free + + 0.613972165 seconds time elapsed + +You can observe the statistical properties as well, by using the +'repeat the workload N times' feature of perf stat: + + titan:~> perf stat --repeat 5 -e kmem:mm_page_pcpu_drain -e + kmem:mm_page_alloc -e kmem:mm_page_free_batched -e + kmem:mm_page_free ./hackbench 10 + Time: 0.627 + Time: 0.644 + Time: 0.564 + Time: 0.559 + Time: 0.626 + + Performance counter stats for './hackbench 10' (5 runs): + + 12920 kmem:mm_page_pcpu_drain ( +- 3.359% ) + 25035 kmem:mm_page_alloc ( +- 3.783% ) + 6104 kmem:mm_page_free_batched ( +- 0.934% ) + 18376 kmem:mm_page_free ( +- 4.941% ) + + 0.643954516 seconds time elapsed ( +- 2.363% ) + +Furthermore, these tracepoints can be used to sample the workload as +well. For example the page allocations done by a 'git gc' can be +captured the following way: + + titan:~/git> perf record -f -e kmem:mm_page_alloc -c 1 ./git gc + Counting objects: 1148, done. + Delta compression using up to 2 threads. + Compressing objects: 100% (450/450), done. + Writing objects: 100% (1148/1148), done. + Total 1148 (delta 690), reused 1148 (delta 690) + [ perf record: Captured and wrote 0.267 MB perf.data (~11679 samples) ] + +To check which functions generated page allocations: + + titan:~/git> perf report + # Samples: 10646 + # + # Overhead Command Shared Object + # ........ ............... .......................... + # + 23.57% git-repack /lib64/libc-2.5.so + 21.81% git /lib64/libc-2.5.so + 14.59% git ./git + 11.79% git-repack ./git + 7.12% git /lib64/ld-2.5.so + 3.16% git-repack /lib64/libpthread-2.5.so + 2.09% git-repack /bin/bash + 1.97% rm /lib64/libc-2.5.so + 1.39% mv /lib64/ld-2.5.so + 1.37% mv /lib64/libc-2.5.so + 1.12% git-repack /lib64/ld-2.5.so + 0.95% rm /lib64/ld-2.5.so + 0.90% git-update-serv /lib64/libc-2.5.so + 0.73% git-update-serv /lib64/ld-2.5.so + 0.68% perf /lib64/libpthread-2.5.so + 0.64% git-repack /usr/lib64/libz.so.1.2.3 + +Or to see it on a more finegrained level: + +titan:~/git> perf report --sort comm,dso,symbol +# Samples: 10646 +# +# Overhead Command Shared Object Symbol +# ........ ............... .......................... ...... +# + 9.35% git-repack ./git [.] insert_obj_hash + 9.12% git ./git [.] insert_obj_hash + 7.31% git /lib64/libc-2.5.so [.] memcpy + 6.34% git-repack /lib64/libc-2.5.so [.] _int_malloc + 6.24% git-repack /lib64/libc-2.5.so [.] memcpy + 5.82% git-repack /lib64/libc-2.5.so [.] __GI___fork + 5.47% git /lib64/libc-2.5.so [.] _int_malloc + 2.99% git /lib64/libc-2.5.so [.] memset + +Furthermore, call-graph sampling can be done too, of page +allocations - to see precisely what kind of page allocations there +are: + + titan:~/git> perf record -f -g -e kmem:mm_page_alloc -c 1 ./git gc + Counting objects: 1148, done. + Delta compression using up to 2 threads. + Compressing objects: 100% (450/450), done. + Writing objects: 100% (1148/1148), done. + Total 1148 (delta 690), reused 1148 (delta 690) + [ perf record: Captured and wrote 0.963 MB perf.data (~42069 samples) ] + + titan:~/git> perf report -g + # Samples: 10686 + # + # Overhead Command Shared Object + # ........ ............... .......................... + # + 23.25% git-repack /lib64/libc-2.5.so + | + |--50.00%-- _int_free + | + |--37.50%-- __GI___fork + | make_child + | + |--12.50%-- ptmalloc_unlock_all2 + | make_child + | + --6.25%-- __GI_strcpy + 21.61% git /lib64/libc-2.5.so + | + |--30.00%-- __GI_read + | | + | --83.33%-- git_config_from_file + | git_config + | | + [...] + +Or you can observe the whole system's page allocations for 10 +seconds: + +titan:~/git> perf stat -a -e kmem:mm_page_pcpu_drain -e +kmem:mm_page_alloc -e kmem:mm_page_free_batched -e +kmem:mm_page_free sleep 10 + + Performance counter stats for 'sleep 10': + + 171585 kmem:mm_page_pcpu_drain + 322114 kmem:mm_page_alloc + 73623 kmem:mm_page_free_batched + 254115 kmem:mm_page_free + + 10.000591410 seconds time elapsed + +Or observe how fluctuating the page allocations are, via statistical +analysis done over ten 1-second intervals: + + titan:~/git> perf stat --repeat 10 -a -e kmem:mm_page_pcpu_drain -e + kmem:mm_page_alloc -e kmem:mm_page_free_batched -e + kmem:mm_page_free sleep 1 + + Performance counter stats for 'sleep 1' (10 runs): + + 17254 kmem:mm_page_pcpu_drain ( +- 3.709% ) + 34394 kmem:mm_page_alloc ( +- 4.617% ) + 7509 kmem:mm_page_free_batched ( +- 4.820% ) + 25653 kmem:mm_page_free ( +- 3.672% ) + + 1.058135029 seconds time elapsed ( +- 3.089% ) + +Or you can annotate the recorded 'git gc' run on a per symbol basis +and check which instructions/source-code generated page allocations: + + titan:~/git> perf annotate __GI___fork + ------------------------------------------------ + Percent | Source code & Disassembly of libc-2.5.so + ------------------------------------------------ + : + : + : Disassembly of section .plt: + : Disassembly of section .text: + : + : 00000031a2e95560 <__fork>: + [...] + 0.00 : 31a2e95602: b8 38 00 00 00 mov $0x38,%eax + 0.00 : 31a2e95607: 0f 05 syscall + 83.42 : 31a2e95609: 48 3d 00 f0 ff ff cmp $0xfffffffffffff000,%rax + 0.00 : 31a2e9560f: 0f 87 4d 01 00 00 ja 31a2e95762 <__fork+0x202> + 0.00 : 31a2e95615: 85 c0 test %eax,%eax + +( this shows that 83.42% of __GI___fork's page allocations come from + the 0x38 system call it performs. ) + +etc. etc. - a lot more is possible. I could list a dozen of +other different usecases straight away - neither of which is +possible via /proc/vmstat. + +/proc/vmstat is not in the same league really, in terms of +expressive power of system analysis and performance +analysis. + +All that the above results needed were those new tracepoints +in include/tracing/events/kmem.h. + + Ingo + + diff --git a/tools/perf/Documentation/jit-interface.txt b/tools/perf/Documentation/jit-interface.txt new file mode 100644 index 00000000..a8656f56 --- /dev/null +++ b/tools/perf/Documentation/jit-interface.txt @@ -0,0 +1,15 @@ +perf supports a simple JIT interface to resolve symbols for dynamic code generated +by a JIT. + +The JIT has to write a /tmp/perf-%d.map (%d = pid of process) file + +This is a text file. + +Each line has the following format, fields separated with spaces: + +START SIZE symbolname + +START and SIZE are hex numbers without 0x. +symbolname is the rest of the line, so it could contain special characters. + +The ownership of the file has to match the process. diff --git a/tools/perf/Documentation/manpage-1.72.xsl b/tools/perf/Documentation/manpage-1.72.xsl new file mode 100644 index 00000000..b4d315cb --- /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">▓</xsl:param> +<xsl:param name="git.docbook.dot" >⌂</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 00000000..a264fa61 --- /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 </xsl:text> + <xsl:apply-templates/> + <xsl:text> </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="$git.docbook.dot"/> + <xsl:text>br </xsl:text> +</xsl:template> + +</xsl:stylesheet> diff --git a/tools/perf/Documentation/manpage-bold-literal.xsl b/tools/perf/Documentation/manpage-bold-literal.xsl new file mode 100644 index 00000000..608eb5df --- /dev/null +++ b/tools/perf/Documentation/manpage-bold-literal.xsl @@ -0,0 +1,17 @@ +<!-- manpage-bold-literal.xsl: + special formatting for manpages rendered from asciidoc+docbook --> +<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" + version="1.0"> + +<!-- render literal text as bold (instead of plain or monospace); + this makes literal text easier to distinguish in manpages + viewed on a tty --> +<xsl:template match="literal"> + <xsl:value-of select="$git.docbook.backslash"/> + <xsl:text>fB</xsl:text> + <xsl:apply-templates/> + <xsl:value-of select="$git.docbook.backslash"/> + <xsl:text>fR</xsl:text> +</xsl:template> + +</xsl:stylesheet> diff --git a/tools/perf/Documentation/manpage-normal.xsl b/tools/perf/Documentation/manpage-normal.xsl new file mode 100644 index 00000000..a48f5b11 --- /dev/null +++ b/tools/perf/Documentation/manpage-normal.xsl @@ -0,0 +1,13 @@ +<!-- manpage-normal.xsl: + special settings for manpages rendered from asciidoc+docbook + handles anything we want to keep away from docbook-xsl 1.72.0 --> +<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" + version="1.0"> + +<xsl:import href="manpage-base.xsl"/> + +<!-- these are the normal values for the roff control characters --> +<xsl:param name="git.docbook.backslash">\</xsl:param> +<xsl:param name="git.docbook.dot" >.</xsl:param> + +</xsl:stylesheet> diff --git a/tools/perf/Documentation/manpage-suppress-sp.xsl b/tools/perf/Documentation/manpage-suppress-sp.xsl new file mode 100644 index 00000000..a63c7632 --- /dev/null +++ b/tools/perf/Documentation/manpage-suppress-sp.xsl @@ -0,0 +1,21 @@ +<!-- manpage-suppress-sp.xsl: + special settings for manpages rendered from asciidoc+docbook + handles erroneous, inline .sp in manpage output of some + versions of docbook-xsl --> +<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" + version="1.0"> + +<!-- attempt to work around spurious .sp at the tail of the line + that some versions of docbook stylesheets seem to add --> +<xsl:template match="simpara"> + <xsl:variable name="content"> + <xsl:apply-templates/> + </xsl:variable> + <xsl:value-of select="normalize-space($content)"/> + <xsl:if test="not(ancestor::authorblurb) and + not(ancestor::personblurb)"> + <xsl:text> </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 00000000..5ad07ef4 --- /dev/null +++ b/tools/perf/Documentation/perf-annotate.txt @@ -0,0 +1,98 @@ +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 unless stdin is a fifo) + +-d:: +--dsos=<dso[,dso...]>:: + Only consider symbols in these dsos. +-s:: +--symbol=<symbol>:: + Symbol to annotate. + +-f:: +--force:: + Don't complain, do it. + +-v:: +--verbose:: + Be more verbose. (Show symbol address, etc) + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + +-k:: +--vmlinux=<file>:: + vmlinux pathname. + +-m:: +--modules:: + Load module symbols. WARNING: use only with -k and LIVE kernel. + +-l:: +--print-line:: + Print matching source lines (may be slow). + +-P:: +--full-paths:: + Don't shorten the displayed pathnames. + +--stdio:: Use the stdio interface. + +--tui:: Use the TUI interface. Use of --tui requires a tty, if one is not + present, as when piping to other commands, the stdio interface is + used. This interfaces starts by centering on the line with more + samples, TAB/UNTAB cycles through the lines with more samples. + +--gtk:: Use the GTK interface. + +-C:: +--cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can + be provided as a comma-separated list with no space: 0,1. Ranges of + CPUs are specified with -: 0-2. Default is to report samples on all + CPUs. + +--asm-raw:: + Show raw instruction encoding of assembly instructions. + +--source:: + Interleave source code with assembly code. Enabled by default, + disable with --no-source. + +--symfs=<directory>:: + Look for files with symbols relative to this directory. + +-M:: +--disassembler-style=:: Set disassembler style for objdump. + +--objdump=<path>:: + Path to objdump binary. + +--skip-missing:: + Skip symbols that cannot be annotated. + +SEE ALSO +-------- +linkperf:perf-record[1], linkperf:perf-report[1] diff --git a/tools/perf/Documentation/perf-archive.txt b/tools/perf/Documentation/perf-archive.txt new file mode 100644 index 00000000..fae174dc --- /dev/null +++ b/tools/perf/Documentation/perf-archive.txt @@ -0,0 +1,22 @@ +perf-archive(1) +=============== + +NAME +---- +perf-archive - Create archive with object files with build-ids found in perf.data file + +SYNOPSIS +-------- +[verse] +'perf archive' [file] + +DESCRIPTION +----------- +This command runs runs perf-buildid-list --with-hits, and collects the files +with the buildids found so that analisys of perf.data contents can be possible +on another machine. + + +SEE ALSO +-------- +linkperf:perf-record[1], linkperf:perf-buildid-list[1], linkperf:perf-report[1] diff --git a/tools/perf/Documentation/perf-bench.txt b/tools/perf/Documentation/perf-bench.txt new file mode 100644 index 00000000..7065cd6f --- /dev/null +++ b/tools/perf/Documentation/perf-bench.txt @@ -0,0 +1,192 @@ +perf-bench(1) +============= + +NAME +---- +perf-bench - General framework for benchmark suites + +SYNOPSIS +-------- +[verse] +'perf bench' [<common options>] <subsystem> <suite> [<options>] + +DESCRIPTION +----------- +This 'perf bench' command is a general framework for benchmark suites. + +COMMON OPTIONS +-------------- +-f:: +--format=:: +Specify format style. +Current available format styles are: + +'default':: +Default style. This is mainly for human reading. +--------------------- +% perf bench sched pipe # with no style specified +(executing 1000000 pipe operations between two tasks) + Total time:5.855 sec + 5.855061 usecs/op + 170792 ops/sec +--------------------- + +'simple':: +This simple style is friendly for automated +processing by scripts. +--------------------- +% perf bench --format=simple sched pipe # specified simple +5.988 +--------------------- + +SUBSYSTEM +--------- + +'sched':: + Scheduler and IPC mechanisms. + +'mem':: + Memory access performance. + +'all':: + All benchmark subsystems. + +SUITES FOR 'sched' +~~~~~~~~~~~~~~~~~~ +*messaging*:: +Suite for evaluating performance of scheduler and IPC mechanisms. +Based on hackbench by Rusty Russell. + +Options of *messaging* +^^^^^^^^^^^^^^^^^^^^^^ +-p:: +--pipe:: +Use pipe() instead of socketpair() + +-t:: +--thread:: +Be multi thread instead of multi process + +-g:: +--group=:: +Specify number of groups + +-l:: +--loop=:: +Specify number of loops + +Example of *messaging* +^^^^^^^^^^^^^^^^^^^^^^ + +--------------------- +% perf bench sched messaging # run with default +options (20 sender and receiver processes per group) +(10 groups == 400 processes run) + + Total time:0.308 sec + +% perf bench sched messaging -t -g 20 # be multi-thread, with 20 groups +(20 sender and receiver threads per group) +(20 groups == 800 threads run) + + Total time:0.582 sec +--------------------- + +*pipe*:: +Suite for pipe() system call. +Based on pipe-test-1m.c by Ingo Molnar. + +Options of *pipe* +^^^^^^^^^^^^^^^^^ +-l:: +--loop=:: +Specify number of loops. + +Example of *pipe* +^^^^^^^^^^^^^^^^^ + +--------------------- +% perf bench sched pipe +(executing 1000000 pipe operations between two tasks) + + Total time:8.091 sec + 8.091833 usecs/op + 123581 ops/sec + +% perf bench sched pipe -l 1000 # loop 1000 +(executing 1000 pipe operations between two tasks) + + Total time:0.016 sec + 16.948000 usecs/op + 59004 ops/sec +--------------------- + +SUITES FOR 'mem' +~~~~~~~~~~~~~~~~ +*memcpy*:: +Suite for evaluating performance of simple memory copy in various ways. + +Options of *memcpy* +^^^^^^^^^^^^^^^^^^^ +-l:: +--length:: +Specify length of memory to copy (default: 1MB). +Available units are B, KB, MB, GB and TB (case insensitive). + +-r:: +--routine:: +Specify routine to copy (default: default). +Available routines are depend on the architecture. +On x86-64, x86-64-unrolled, x86-64-movsq and x86-64-movsb are supported. + +-i:: +--iterations:: +Repeat memcpy invocation this number of times. + +-c:: +--cycle:: +Use perf's cpu-cycles event instead of gettimeofday syscall. + +-o:: +--only-prefault:: +Show only the result with page faults before memcpy. + +-n:: +--no-prefault:: +Show only the result without page faults before memcpy. + +*memset*:: +Suite for evaluating performance of simple memory set in various ways. + +Options of *memset* +^^^^^^^^^^^^^^^^^^^ +-l:: +--length:: +Specify length of memory to set (default: 1MB). +Available units are B, KB, MB, GB and TB (case insensitive). + +-r:: +--routine:: +Specify routine to set (default: default). +Available routines are depend on the architecture. +On x86-64, x86-64-unrolled, x86-64-stosq and x86-64-stosb are supported. + +-i:: +--iterations:: +Repeat memset invocation this number of times. + +-c:: +--cycle:: +Use perf's cpu-cycles event instead of gettimeofday syscall. + +-o:: +--only-prefault:: +Show only the result with page faults before memset. + +-n:: +--no-prefault:: +Show only the result without page faults before memset. + +SEE ALSO +-------- +linkperf:perf[1] diff --git a/tools/perf/Documentation/perf-buildid-cache.txt b/tools/perf/Documentation/perf-buildid-cache.txt new file mode 100644 index 00000000..e9a8349a --- /dev/null +++ b/tools/perf/Documentation/perf-buildid-cache.txt @@ -0,0 +1,40 @@ +perf-buildid-cache(1) +===================== + +NAME +---- +perf-buildid-cache - Manage build-id cache. + +SYNOPSIS +-------- +[verse] +'perf buildid-cache <options>' + +DESCRIPTION +----------- +This command manages the build-id cache. It can add and remove files to/from +the cache. In the future it should as well purge older entries, set upper +limits for the space used by the cache, etc. + +OPTIONS +------- +-a:: +--add=:: + Add specified file to the cache. +-r:: +--remove=:: + Remove specified file from the cache. +-M:: +--missing=:: + List missing build ids in the cache for the specified file. +-u:: +--update:: + Update specified file of the cache. It can be used to update kallsyms + kernel dso to vmlinux in order to support annotation. +-v:: +--verbose:: + Be more verbose. + +SEE ALSO +-------- +linkperf:perf-record[1], linkperf:perf-report[1], linkperf:perf-buildid-list[1] diff --git a/tools/perf/Documentation/perf-buildid-list.txt b/tools/perf/Documentation/perf-buildid-list.txt new file mode 100644 index 00000000..25c52efc --- /dev/null +++ b/tools/perf/Documentation/perf-buildid-list.txt @@ -0,0 +1,43 @@ +perf-buildid-list(1) +==================== + +NAME +---- +perf-buildid-list - List the buildids in a perf.data file + +SYNOPSIS +-------- +[verse] +'perf buildid-list <options>' + +DESCRIPTION +----------- +This command displays the buildids found in a perf.data file, so that other +tools can be used to fetch packages with matching symbol tables for use by +perf report. + +It can also be used to show the build id of the running kernel or in an ELF +file using -i/--input. + +OPTIONS +------- +-H:: +--with-hits:: + Show only DSOs with hits. +-i:: +--input=:: + Input file name. (default: perf.data unless stdin is a fifo) +-f:: +--force:: + Don't do ownership validation. +-k:: +--kernel:: + Show running kernel build id. +-v:: +--verbose:: + Be more verbose. + +SEE ALSO +-------- +linkperf:perf-record[1], linkperf:perf-top[1], +linkperf:perf-report[1] diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt new file mode 100644 index 00000000..5b3123d5 --- /dev/null +++ b/tools/perf/Documentation/perf-diff.txt @@ -0,0 +1,133 @@ +perf-diff(1) +============ + +NAME +---- +perf-diff - Read two perf.data files and display the differential profile + +SYNOPSIS +-------- +[verse] +'perf diff' [oldfile] [newfile] + +DESCRIPTION +----------- +This command displays the performance difference amongst two perf.data files +captured via perf record. + +If no parameters are passed it will assume perf.data.old and perf.data. + +The differential profile is displayed only for events matching both +specified perf.data files. + +OPTIONS +------- +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + +-m:: +--modules:: + Load module symbols. WARNING: use only with -k and LIVE kernel + +-d:: +--dsos=:: + Only consider symbols in these dsos. CSV that understands + file://filename entries. + +-C:: +--comms=:: + Only consider symbols in these comms. CSV that understands + file://filename entries. + +-S:: +--symbols=:: + Only consider these symbols. CSV that understands + file://filename entries. + +-s:: +--sort=:: + Sort by key(s): pid, comm, dso, symbol. + +-t:: +--field-separator=:: + + Use a special separator character and don't pad with spaces, replacing + all occurrences of this separator in symbol names (and other output) + with a '.' character, that thus it's the only non valid separator. + +-v:: +--verbose:: + Be verbose, for instance, show the raw counts in addition to the + diff. + +-f:: +--force:: + Don't complain, do it. + +--symfs=<directory>:: + Look for files with symbols relative to this directory. + +-b:: +--baseline-only:: + Show only items with match in baseline. + +-c:: +--compute:: + Differential computation selection - delta,ratio,wdiff (default is delta). + If '+' is specified as a first character, the output is sorted based + on the computation results. + See COMPARISON METHODS section for more info. + +-p:: +--period:: + Show period values for both compared hist entries. + +-F:: +--formula:: + Show formula for given computation. + +COMPARISON METHODS +------------------ +delta +~~~~~ +If specified the 'Delta' column is displayed with value 'd' computed as: + + d = A->period_percent - B->period_percent + +with: + - A/B being matching hist entry from first/second file specified + (or perf.data/perf.data.old) respectively. + + - period_percent being the % of the hist entry period value within + single data file + +ratio +~~~~~ +If specified the 'Ratio' column is displayed with value 'r' computed as: + + r = A->period / B->period + +with: + - A/B being matching hist entry from first/second file specified + (or perf.data/perf.data.old) respectively. + + - period being the hist entry period value + +wdiff +~~~~~ +If specified the 'Weighted diff' column is displayed with value 'd' computed as: + + d = B->period * WEIGHT-A - A->period * WEIGHT-B + + - A/B being matching hist entry from first/second file specified + (or perf.data/perf.data.old) respectively. + + - period being the hist entry period value + + - WEIGHT-A/WEIGHT-B being user suplied weights in the the '-c' option + behind ':' separator like '-c wdiff:1,2'. + +SEE ALSO +-------- +linkperf:perf-record[1] diff --git a/tools/perf/Documentation/perf-evlist.txt b/tools/perf/Documentation/perf-evlist.txt new file mode 100644 index 00000000..1ceb3700 --- /dev/null +++ b/tools/perf/Documentation/perf-evlist.txt @@ -0,0 +1,38 @@ +perf-evlist(1) +============== + +NAME +---- +perf-evlist - List the event names in a perf.data file + +SYNOPSIS +-------- +[verse] +'perf evlist <options>' + +DESCRIPTION +----------- +This command displays the names of events sampled in a perf.data file. + +OPTIONS +------- +-i:: +--input=:: + Input file name. (default: perf.data unless stdin is a fifo) + +-F:: +--freq=:: + Show just the sample frequency used for each event. + +-v:: +--verbose=:: + Show all fields. + +-g:: +--group:: + Show event group information. + +SEE ALSO +-------- +linkperf:perf-record[1], linkperf:perf-list[1], +linkperf:perf-report[1] diff --git a/tools/perf/Documentation/perf-help.txt b/tools/perf/Documentation/perf-help.txt new file mode 100644 index 00000000..51439181 --- /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-inject.txt b/tools/perf/Documentation/perf-inject.txt new file mode 100644 index 00000000..a00a3427 --- /dev/null +++ b/tools/perf/Documentation/perf-inject.txt @@ -0,0 +1,46 @@ +perf-inject(1) +============== + +NAME +---- +perf-inject - Filter to augment the events stream with additional information + +SYNOPSIS +-------- +[verse] +'perf inject <options>' + +DESCRIPTION +----------- +perf-inject reads a perf-record event stream and repipes it to stdout. At any +point the processing code can inject other events into the event stream - in +this case build-ids (-b option) are read and injected as needed into the event +stream. + +Build-ids are just the first user of perf-inject - potentially anything that +needs userspace processing to augment the events stream with additional +information could make use of this facility. + +OPTIONS +------- +-b:: +--build-ids=:: + Inject build-ids into the output stream +-v:: +--verbose:: + Be more verbose. +-i:: +--input=:: + Input file name. (default: stdin) +-o:: +--output=:: + Output file name. (default: stdout) +-s:: +--sched-stat:: + Merge sched_stat and sched_switch for getting events where and how long + tasks slept. sched_switch contains a callchain where a task slept and + sched_stat contains a timeslice how long a task slept. + +SEE ALSO +-------- +linkperf:perf-record[1], linkperf:perf-report[1], linkperf:perf-archive[1] diff --git a/tools/perf/Documentation/perf-kmem.txt b/tools/perf/Documentation/perf-kmem.txt new file mode 100644 index 00000000..7c8fbbf3 --- /dev/null +++ b/tools/perf/Documentation/perf-kmem.txt @@ -0,0 +1,47 @@ +perf-kmem(1) +============ + +NAME +---- +perf-kmem - Tool to trace/measure kernel memory(slab) properties + +SYNOPSIS +-------- +[verse] +'perf kmem' {record|stat} [<options>] + +DESCRIPTION +----------- +There are two variants of perf kmem: + + 'perf kmem record <command>' to record the kmem events + of an arbitrary workload. + + 'perf kmem stat' to report kernel memory statistics. + +OPTIONS +------- +-i <file>:: +--input=<file>:: + Select the input file (default: perf.data unless stdin is a fifo) + +--caller:: + Show per-callsite statistics + +--alloc:: + Show per-allocation statistics + +-s <key[,key2...]>:: +--sort=<key[,key2...]>:: + Sort the output (default: frag,hit,bytes) + +-l <num>:: +--line=<num>:: + Print n lines only + +--raw-ip:: + Print raw ip instead of symbol + +SEE ALSO +-------- +linkperf:perf-record[1] diff --git a/tools/perf/Documentation/perf-kvm.txt b/tools/perf/Documentation/perf-kvm.txt new file mode 100644 index 00000000..326f2cb3 --- /dev/null +++ b/tools/perf/Documentation/perf-kvm.txt @@ -0,0 +1,100 @@ +perf-kvm(1) +=========== + +NAME +---- +perf-kvm - Tool to trace/measure kvm guest os + +SYNOPSIS +-------- +[verse] +'perf kvm' [--host] [--guest] [--guestmount=<path> + [--guestkallsyms=<path> --guestmodules=<path> | --guestvmlinux=<path>]] + {top|record|report|diff|buildid-list} +'perf kvm' [--host] [--guest] [--guestkallsyms=<path> --guestmodules=<path> + | --guestvmlinux=<path>] {top|record|report|diff|buildid-list|stat} + +DESCRIPTION +----------- +There are a couple of variants of perf kvm: + + 'perf kvm [options] top <command>' to generates and displays + a performance counter profile of guest os in realtime + of an arbitrary workload. + + 'perf kvm record <command>' to record the performance counter profile + of an arbitrary workload and save it into a perf data file. If both + --host and --guest are input, the perf data file name is perf.data.kvm. + If there is no --host but --guest, the file name is perf.data.guest. + If there is no --guest but --host, the file name is perf.data.host. + + 'perf kvm report' to display the performance counter profile information + recorded via perf kvm record. + + 'perf kvm diff' to displays the performance difference amongst two perf.data + files captured via perf record. + + 'perf kvm buildid-list' to display the buildids found in a perf data file, + so that other tools can be used to fetch packages with matching symbol tables + for use by perf report. + + 'perf kvm stat <command>' to run a command and gather performance counter + statistics. + Especially, perf 'kvm stat record/report' generates a statistical analysis + of KVM events. Currently, vmexit, mmio and ioport events are supported. + 'perf kvm stat record <command>' records kvm events and the events between + start and end <command>. + And this command produces a file which contains tracing results of kvm + events. + + 'perf kvm stat report' reports statistical data which includes events + handled time, samples, and so on. + +OPTIONS +------- +-i:: +--input=:: + Input file name. +-o:: +--output:: + Output file name. +--host=:: + Collect host side performance profile. +--guest=:: + Collect guest side performance profile. +--guestmount=<path>:: + Guest os root file system mount directory. Users mounts guest os + root directories under <path> by a specific filesystem access method, + typically, sshfs. For example, start 2 guest os. The one's pid is 8888 + and the other's is 9999. + #mkdir ~/guestmount; cd ~/guestmount + #sshfs -o allow_other,direct_io -p 5551 localhost:/ 8888/ + #sshfs -o allow_other,direct_io -p 5552 localhost:/ 9999/ + #perf kvm --host --guest --guestmount=~/guestmount top +--guestkallsyms=<path>:: + Guest os /proc/kallsyms file copy. 'perf' kvm' reads it to get guest + kernel symbols. Users copy it out from guest os. +--guestmodules=<path>:: + Guest os /proc/modules file copy. 'perf' kvm' reads it to get guest + kernel module information. Users copy it out from guest os. +--guestvmlinux=<path>:: + Guest os kernel vmlinux. + +STAT REPORT OPTIONS +------------------- +--vcpu=<value>:: + analyze events which occures on this vcpu. (default: all vcpus) + +--events=<value>:: + events to be analyzed. Possible values: vmexit, mmio, ioport. + (default: vmexit) +-k:: +--key=<value>:: + Sorting key. Possible values: sample (default, sort by samples + number), time (sort by average time). + +SEE ALSO +-------- +linkperf:perf-top[1], linkperf:perf-record[1], linkperf:perf-report[1], +linkperf:perf-diff[1], linkperf:perf-buildid-list[1], +linkperf:perf-stat[1] diff --git a/tools/perf/Documentation/perf-list.txt b/tools/perf/Documentation/perf-list.txt new file mode 100644 index 00000000..d1e39dc8 --- /dev/null +++ b/tools/perf/Documentation/perf-list.txt @@ -0,0 +1,118 @@ +perf-list(1) +============ + +NAME +---- +perf-list - List all symbolic event types + +SYNOPSIS +-------- +[verse] +'perf list' [hw|sw|cache|tracepoint|event_glob] + +DESCRIPTION +----------- +This command displays the symbolic event types which can be selected in the +various perf commands with the -e option. + +[[EVENT_MODIFIERS]] +EVENT MODIFIERS +--------------- + +Events can optionally have a modifer by appending a colon and one or +more modifiers. Modifiers allow the user to restrict the events to be +counted. The following modifiers exist: + + u - user-space counting + k - kernel counting + h - hypervisor counting + G - guest counting (in KVM guests) + H - host counting (not in KVM guests) + p - precise level + +The 'p' modifier can be used for specifying how precise the instruction +address should be. The 'p' modifier can be specified multiple times: + + 0 - SAMPLE_IP can have arbitrary skid + 1 - SAMPLE_IP must have constant skid + 2 - SAMPLE_IP requested to have 0 skid + 3 - SAMPLE_IP must have 0 skid + +For Intel systems precise event sampling is implemented with PEBS +which supports up to precise-level 2. + +On AMD systems it is implemented using IBS (up to precise-level 2). +The precise modifier works with event types 0x76 (cpu-cycles, CPU +clocks not halted) and 0xC1 (micro-ops retired). Both events map to +IBS execution sampling (IBS op) with the IBS Op Counter Control bit +(IbsOpCntCtl) set respectively (see AMD64 Architecture Programmer’s +Manual Volume 2: System Programming, 13.3 Instruction-Based +Sampling). Examples to use IBS: + + perf record -a -e cpu-cycles:p ... # use ibs op counting cycles + perf record -a -e r076:p ... # same as -e cpu-cycles:p + perf record -a -e r0C1:p ... # use ibs op counting micro-ops + +RAW HARDWARE EVENT DESCRIPTOR +----------------------------- +Even when an event is not available in a symbolic form within perf right now, +it can be encoded in a per processor specific way. + +For instance For x86 CPUs NNN represents the raw register encoding with the +layout of IA32_PERFEVTSELx MSRs (see [Intel® 64 and IA-32 Architectures Software Developer's Manual Volume 3B: System Programming Guide] Figure 30-1 Layout +of IA32_PERFEVTSELx MSRs) or AMD's PerfEvtSeln (see [AMD64 Architecture Programmer’s Manual Volume 2: System Programming], Page 344, +Figure 13-7 Performance Event-Select Register (PerfEvtSeln)). + +Note: Only the following bit fields can be set in x86 counter +registers: event, umask, edge, inv, cmask. Esp. guest/host only and +OS/user mode flags must be setup using <<EVENT_MODIFIERS, EVENT +MODIFIERS>>. + +Example: + +If the Intel docs for a QM720 Core i7 describe an event as: + + Event Umask Event Mask + Num. Value Mnemonic Description Comment + + A8H 01H LSD.UOPS Counts the number of micro-ops Use cmask=1 and + delivered by loop stream detector invert to count + cycles + +raw encoding of 0x1A8 can be used: + + perf stat -e r1a8 -a sleep 1 + perf record -e r1a8 ... + +You should refer to the processor specific documentation for getting these +details. Some of them are referenced in the SEE ALSO section below. + +OPTIONS +------- + +Without options all known events will be listed. + +To limit the list use: + +. 'hw' or 'hardware' to list hardware events such as cache-misses, etc. + +. 'sw' or 'software' to list software events such as context switches, etc. + +. 'cache' or 'hwcache' to list hardware cache events such as L1-dcache-loads, etc. + +. 'tracepoint' to list all tracepoint events, alternatively use + 'subsys_glob:event_glob' to filter by tracepoint subsystems such as sched, + block, etc. + +. If none of the above is matched, it will apply the supplied glob to all + events, printing the ones that match. + +One or more types can be used at the same time, listing the events for the +types specified. + +SEE ALSO +-------- +linkperf:perf-stat[1], linkperf:perf-top[1], +linkperf:perf-record[1], +http://www.intel.com/Assets/PDF/manual/253669.pdf[Intel® 64 and IA-32 Architectures Software Developer's Manual Volume 3B: System Programming Guide], +http://support.amd.com/us/Processor_TechDocs/24593_APM_v2.pdf[AMD64 Architecture Programmer’s Manual Volume 2: System Programming] diff --git a/tools/perf/Documentation/perf-lock.txt b/tools/perf/Documentation/perf-lock.txt new file mode 100644 index 00000000..c7f5f556 --- /dev/null +++ b/tools/perf/Documentation/perf-lock.txt @@ -0,0 +1,66 @@ +perf-lock(1) +============ + +NAME +---- +perf-lock - Analyze lock events + +SYNOPSIS +-------- +[verse] +'perf lock' {record|report|script|info} + +DESCRIPTION +----------- +You can analyze various lock behaviours +and statistics with this 'perf lock' command. + + 'perf lock record <command>' records lock events + between start and end <command>. And this command + produces the file "perf.data" which contains tracing + results of lock events. + + 'perf lock report' reports statistical data. + + 'perf lock script' shows raw lock events. + + 'perf lock info' shows metadata like threads or addresses + of lock instances. + +COMMON OPTIONS +-------------- + +-i:: +--input=<file>:: + Input file name. (default: perf.data unless stdin is a fifo) + +-v:: +--verbose:: + Be more verbose (show symbol address, etc). + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + +REPORT OPTIONS +-------------- + +-k:: +--key=<value>:: + Sorting key. Possible values: acquired (default), contended, + wait_total, wait_max, wait_min. + +INFO OPTIONS +------------ + +-t:: +--threads:: + dump thread list in perf.data + +-m:: +--map:: + dump map of lock instances (address:name table) + +SEE ALSO +-------- +linkperf:perf[1] diff --git a/tools/perf/Documentation/perf-probe.txt b/tools/perf/Documentation/perf-probe.txt new file mode 100644 index 00000000..b715cb71 --- /dev/null +++ b/tools/perf/Documentation/perf-probe.txt @@ -0,0 +1,205 @@ +perf-probe(1) +============= + +NAME +---- +perf-probe - Define new dynamic tracepoints + +SYNOPSIS +-------- +[verse] +'perf probe' [options] --add='PROBE' [...] +or +'perf probe' [options] PROBE +or +'perf probe' [options] --del='[GROUP:]EVENT' [...] +or +'perf probe' --list +or +'perf probe' [options] --line='LINE' +or +'perf probe' [options] --vars='PROBEPOINT' + +DESCRIPTION +----------- +This command defines dynamic tracepoint events, by symbol and registers +without debuginfo, or by C expressions (C line numbers, C function names, +and C local variables) with debuginfo. + + +OPTIONS +------- +-k:: +--vmlinux=PATH:: + Specify vmlinux path which has debuginfo (Dwarf binary). + +-m:: +--module=MODNAME|PATH:: + Specify module name in which perf-probe searches probe points + or lines. If a path of module file is passed, perf-probe + treat it as an offline module (this means you can add a probe on + a module which has not been loaded yet). + +-s:: +--source=PATH:: + Specify path to kernel source. + +-v:: +--verbose:: + Be more verbose (show parsed arguments, etc). + +-a:: +--add=:: + Define a probe event (see PROBE SYNTAX for detail). + +-d:: +--del=:: + Delete probe events. This accepts glob wildcards('*', '?') and character + classes(e.g. [a-z], [!A-Z]). + +-l:: +--list:: + List up current probe events. + +-L:: +--line=:: + Show source code lines which can be probed. This needs an argument + which specifies a range of the source code. (see LINE SYNTAX for detail) + +-V:: +--vars=:: + Show available local variables at given probe point. The argument + syntax is same as PROBE SYNTAX, but NO ARGs. + +--externs:: + (Only for --vars) Show external defined variables in addition to local + variables. + +-F:: +--funcs:: + Show available functions in given module or kernel. With -x/--exec, + can also list functions in a user space executable / shared library. + +--filter=FILTER:: + (Only for --vars and --funcs) Set filter. FILTER is a combination of glob + pattern, see FILTER PATTERN for detail. + Default FILTER is "!__k???tab_* & !__crc_*" for --vars, and "!_*" + for --funcs. + If several filters are specified, only the last filter is used. + +-f:: +--force:: + Forcibly add events with existing name. + +-n:: +--dry-run:: + Dry run. With this option, --add and --del doesn't execute actual + adding and removal operations. + +--max-probes:: + Set the maximum number of probe points for an event. Default is 128. + +-x:: +--exec=PATH:: + Specify path to the executable or shared library file for user + space tracing. Can also be used with --funcs option. + +In absence of -m/-x options, perf probe checks if the first argument after +the options is an absolute path name. If its an absolute path, perf probe +uses it as a target module/target user space binary to probe. + +PROBE SYNTAX +------------ +Probe points are defined by following syntax. + + 1) Define event based on function name + [EVENT=]FUNC[@SRC][:RLN|+OFFS|%return|;PTN] [ARG ...] + + 2) Define event based on source file with line number + [EVENT=]SRC:ALN [ARG ...] + + 3) Define event based on source file with lazy pattern + [EVENT=]SRC;PTN [ARG ...] + + +'EVENT' specifies the name of new event, if omitted, it will be set the name of the probed function. Currently, event group name is set as 'probe'. +'FUNC' specifies a probed function name, and it may have one of the following options; '+OFFS' is the offset from function entry address in bytes, ':RLN' is the relative-line number from function entry line, and '%return' means that it probes function return. And ';PTN' means lazy matching pattern (see LAZY MATCHING). Note that ';PTN' must be the end of the probe point definition. In addition, '@SRC' specifies a source file which has that function. +It is also possible to specify a probe point by the source line number or lazy matching by using 'SRC:ALN' or 'SRC;PTN' syntax, where 'SRC' is the source file path, ':ALN' is the line number and ';PTN' is the lazy matching pattern. +'ARG' specifies the arguments of this probe point, (see PROBE ARGUMENT). + +PROBE ARGUMENT +-------------- +Each probe argument follows below syntax. + + [NAME=]LOCALVAR|$retval|%REG|@SYMBOL[:TYPE] + +'NAME' specifies the name of this argument (optional). You can use the name of local variable, local data structure member (e.g. var->field, var.field2), local array with fixed index (e.g. array[1], var->array[0], var->pointer[2]), or kprobe-tracer argument format (e.g. $retval, %ax, etc). Note that the name of this argument will be set as the last member name if you specify a local data structure member (e.g. field2 for 'var->field1.field2'.) +'TYPE' casts the type of this argument (optional). If omitted, perf probe automatically set the type based on debuginfo. You can specify 'string' type only for the local variable or structure member which is an array of or a pointer to 'char' or 'unsigned char' type. + +LINE SYNTAX +----------- +Line range is described by following syntax. + + "FUNC[@SRC][:RLN[+NUM|-RLN2]]|SRC[:ALN[+NUM|-ALN2]]" + +FUNC specifies the function name of showing lines. 'RLN' is the start line +number from function entry line, and 'RLN2' is the end line number. As same as +probe syntax, 'SRC' means the source file path, 'ALN' is start line number, +and 'ALN2' is end line number in the file. It is also possible to specify how +many lines to show by using 'NUM'. Moreover, 'FUNC@SRC' combination is good +for searching a specific function when several functions share same name. +So, "source.c:100-120" shows lines between 100th to l20th in source.c file. And "func:10+20" shows 20 lines from 10th line of func function. + +LAZY MATCHING +------------- + The lazy line matching is similar to glob matching but ignoring spaces in both of pattern and target. So this accepts wildcards('*', '?') and character classes(e.g. [a-z], [!A-Z]). + +e.g. + 'a=*' can matches 'a=b', 'a = b', 'a == b' and so on. + +This provides some sort of flexibility and robustness to probe point definitions against minor code changes. For example, actual 10th line of schedule() can be moved easily by modifying schedule(), but the same line matching 'rq=cpu_rq*' may still exist in the function.) + +FILTER PATTERN +-------------- + The filter pattern is a glob matching pattern(s) to filter variables. + In addition, you can use "!" for specifying filter-out rule. You also can give several rules combined with "&" or "|", and fold those rules as one rule by using "(" ")". + +e.g. + With --filter "foo* | bar*", perf probe -V shows variables which start with "foo" or "bar". + With --filter "!foo* & *bar", perf probe -V shows variables which don't start with "foo" and end with "bar", like "fizzbar". But "foobar" is filtered out. + +EXAMPLES +-------- +Display which lines in schedule() can be probed: + + ./perf probe --line schedule + +Add a probe on schedule() function 12th line with recording cpu local variable: + + ./perf probe schedule:12 cpu + or + ./perf probe --add='schedule:12 cpu' + + this will add one or more probes which has the name start with "schedule". + + Add probes on lines in schedule() function which calls update_rq_clock(). + + ./perf probe 'schedule;update_rq_clock*' + or + ./perf probe --add='schedule;update_rq_clock*' + +Delete all probes on schedule(). + + ./perf probe --del='schedule*' + +Add probes at zfree() function on /bin/zsh + + ./perf probe -x /bin/zsh zfree or ./perf probe /bin/zsh zfree + +Add probes at malloc() function on libc + + ./perf probe -x /lib/libc.so.6 malloc or ./perf probe /lib/libc.so.6 malloc + +SEE ALSO +-------- +linkperf:perf-trace[1], linkperf:perf-record[1] diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt new file mode 100644 index 00000000..938e8904 --- /dev/null +++ b/tools/perf/Documentation/perf-record.txt @@ -0,0 +1,187 @@ +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) + + - a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a + hexadecimal event descriptor. + + - a hardware breakpoint event in the form of '\mem:addr[:access]' + where addr is the address in memory you want to break in. + Access is the memory access type (read, write, execute) it can + be passed as follows: '\mem:addr[:[r][w][x]]'. + If you want to profile read-write accesses in 0x1000, just set + 'mem:0x1000:rw'. + +--filter=<filter>:: + Event filter. + +-a:: +--all-cpus:: + System-wide collection from all CPUs. + +-l:: + Scale counter values. + +-p:: +--pid=:: + Record events on existing process ID (comma separated list). + +-t:: +--tid=:: + Record events on existing thread ID (comma separated list). + +-u:: +--uid=:: + Record events in threads owned by uid. Name or number. + +-r:: +--realtime=:: + Collect data with this RT SCHED_FIFO priority. +-D:: +--no-delay:: + Collect data without buffering. +-A:: +--append:: + Append to the output file to do incremental profiling. + +-f:: +--force:: + Overwrite existing data file. (deprecated) + +-c:: +--count=:: + Event period to sample. + +-o:: +--output=:: + Output file name. + +-i:: +--no-inherit:: + Child tasks do not inherit counters. +-F:: +--freq=:: + Profile at this frequency. + +-m:: +--mmap-pages=:: + Number of mmap data pages. Must be a power of two. + +-g:: +--call-graph:: + Do call-graph (stack chain/backtrace) recording. + +-q:: +--quiet:: + Don't print any message, useful for scripting. + +-v:: +--verbose:: + Be more verbose (show counter open errors, etc). + +-s:: +--stat:: + Per thread counts. + +-d:: +--data:: + Sample addresses. + +-T:: +--timestamp:: + Sample timestamps. Use it with 'perf report -D' to see the timestamps, + for instance. + +-n:: +--no-samples:: + Don't sample. + +-R:: +--raw-samples:: +Collect raw sample records from all opened counters (default for tracepoint counters). + +-C:: +--cpu:: +Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +In per-thread mode with inheritance mode on (default), samples are captured only when +the thread executes on the designated CPUs. Default is to monitor all CPUs. + +-N:: +--no-buildid-cache:: +Do not update the builid cache. This saves some overhead in situations +where the information in the perf.data file (which includes buildids) +is sufficient. + +-G name,...:: +--cgroup name,...:: +monitor only in the container (cgroup) called "name". This option is available only +in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to +container "name" are monitored when they run on the monitored CPUs. Multiple cgroups +can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup +to first event, second cgroup to second event and so on. It is possible to provide +an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have +corresponding events, i.e., they always refer to events defined earlier on the command +line. + +-b:: +--branch-any:: +Enable taken branch stack sampling. Any type of taken branch may be sampled. +This is a shortcut for --branch-filter any. See --branch-filter for more infos. + +-j:: +--branch-filter:: +Enable taken branch stack sampling. Each sample captures a series of consecutive +taken branches. The number of branches captured with each sample depends on the +underlying hardware, the type of branches of interest, and the executed code. +It is possible to select the types of branches captured by enabling filters. The +following filters are defined: + + - any: any type of branches + - any_call: any function call or system call + - any_ret: any function return or system call return + - ind_call: any indirect branch + - u: only when the branch target is at the user level + - k: only when the branch target is in the kernel + - hv: only when the target is at the hypervisor level + ++ +The option requires at least one branch type among any, any_call, any_ret, ind_call. +The privilege levels may be omitted, in which case, the privilege levels of the associated +event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege +levels are subject to permissions. When sampling on multiple events, branch stack sampling +is enabled for all the sampling events. The sampled branch type is the same for all events. +The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k +Note that this feature may not be available on all processors. + +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 00000000..02284a00 --- /dev/null +++ b/tools/perf/Documentation/perf-report.txt @@ -0,0 +1,211 @@ +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 record. + +OPTIONS +------- +-i:: +--input=:: + Input file name. (default: perf.data unless stdin is a fifo) + +-v:: +--verbose:: + Be more verbose. (show symbol address, etc) + +-d:: +--dsos=:: + Only consider symbols in these dsos. CSV that understands + file://filename entries. +-n:: +--show-nr-samples:: + Show the number of samples for each symbol + +--showcpuutilization:: + Show sample percentage for different cpu modes. + +-T:: +--threads:: + Show per-thread event counters +-c:: +--comms=:: + Only consider symbols in these comms. CSV that understands + file://filename entries. +-S:: +--symbols=:: + Only consider these symbols. CSV that understands + file://filename entries. + +--symbol-filter=:: + Only show symbols that match (partially) with this filter. + +-U:: +--hide-unresolved:: + Only display entries resolved to a symbol. + +-s:: +--sort=:: + Sort histogram entries by given key(s) - multiple keys can be specified + in CSV format. Following sort keys are available: + pid, comm, dso, symbol, parent, cpu, srcline. + + Each key has following meaning: + + - comm: command (name) of the task which can be read via /proc/<pid>/comm + - pid: command and tid of the task + - dso: name of library or module executed at the time of sample + - symbol: name of function executed at the time of sample + - parent: name of function matched to the parent regex filter. Unmatched + entries are displayed as "[other]". + - cpu: cpu number the task ran at the time of sample + - srcline: filename and line number executed at the time of sample. The + DWARF debuggin info must be provided. + + By default, comm, dso and symbol keys are used. + (i.e. --sort comm,dso,symbol) + + If --branch-stack option is used, following sort keys are also + available: + dso_from, dso_to, symbol_from, symbol_to, mispredict. + + - dso_from: name of library or module branched from + - dso_to: name of library or module branched to + - symbol_from: name of function branched from + - symbol_to: name of function branched to + - mispredict: "N" for predicted branch, "Y" for mispredicted branch + + And default sort keys are changed to comm, dso_from, symbol_from, dso_to + and symbol_to, see '--branch-stack'. + +-p:: +--parent=<regex>:: + A regex filter to identify parent. The parent is a caller of this + function and searched through the callchain, thus it requires callchain + information recorded. The pattern is in the exteneded regex format and + defaults to "\^sys_|^do_page_fault", see '--sort parent'. + +-x:: +--exclude-other:: + Only display entries with parent-match. + +-w:: +--column-widths=<width[,width...]>:: + Force each column width to the provided list, for large terminal + readability. + +-t:: +--field-separator=:: + Use a special separator character and don't pad with spaces, replacing + all occurrences of this separator in symbol names (and other output) + with a '.' character, that thus it's the only non valid separator. + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + +-g [type,min[,limit],order]:: +--call-graph:: + Display call chains using type, min percent threshold, optional print + limit and order. + type can be either: + - flat: single column, linear exposure of call chains. + - graph: use a graph tree, displaying absolute overhead rates. + - fractal: like graph, but displays relative rates. Each branch of + the tree is considered as a new profiled object. + + + order can be either: + - callee: callee based call graph. + - caller: inverted caller based call graph. + + Default: fractal,0.5,callee. + +-G:: +--inverted:: + alias for inverted caller based call graph. + +--pretty=<key>:: + Pretty printing style. key: normal, raw + +--stdio:: Use the stdio interface. + +--tui:: Use the TUI interface, that is integrated with annotate and allows + zooming into DSOs or threads, among other features. Use of --tui + requires a tty, if one is not present, as when piping to other + commands, the stdio interface is used. + +--gtk:: Use the GTK2 interface. + +-k:: +--vmlinux=<file>:: + vmlinux pathname + +--kallsyms=<file>:: + kallsyms pathname + +-m:: +--modules:: + Load module symbols. WARNING: This should only be used with -k and + a LIVE kernel. + +-f:: +--force:: + Don't complain, do it. + +--symfs=<directory>:: + Look for files with symbols relative to this directory. + +-C:: +--cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can + be provided as a comma-separated list with no space: 0,1. Ranges of + CPUs are specified with -: 0-2. Default is to report samples on all + CPUs. + +-M:: +--disassembler-style=:: Set disassembler style for objdump. + +--source:: + Interleave source code with assembly code. Enabled by default, + disable with --no-source. + +--asm-raw:: + Show raw instruction encoding of assembly instructions. + +--show-total-period:: Show a column with the sum of periods. + +-I:: +--show-info:: + Display extended information about the perf.data file. This adds + information which may be very large and thus may clutter the display. + It currently includes: cpu and numa topology of the host system. + +-b:: +--branch-stack:: + Use the addresses of sampled taken branches instead of the instruction + address to build the histograms. To generate meaningful output, the + perf.data file must have been obtained using perf record -b or + perf record --branch-filter xxx where xxx is a branch filter option. + perf report is able to auto-detect whether a perf.data file contains + branch stacks and it will automatically switch to the branch view mode, + unless --no-branch-stack is used. + +--objdump=<path>:: + Path to objdump binary. + +--group:: + Show event group information together. + +SEE ALSO +-------- +linkperf:perf-stat[1], linkperf:perf-annotate[1] diff --git a/tools/perf/Documentation/perf-sched.txt b/tools/perf/Documentation/perf-sched.txt new file mode 100644 index 00000000..8ff4df95 --- /dev/null +++ b/tools/perf/Documentation/perf-sched.txt @@ -0,0 +1,55 @@ +perf-sched(1) +============== + +NAME +---- +perf-sched - Tool to trace/measure scheduler properties (latencies) + +SYNOPSIS +-------- +[verse] +'perf sched' {record|latency|map|replay|script} + +DESCRIPTION +----------- +There are five variants of perf sched: + + 'perf sched record <command>' to record the scheduling events + of an arbitrary workload. + + 'perf sched latency' to report the per task scheduling latencies + and other scheduling properties of the workload. + + 'perf sched script' to see a detailed trace of the workload that + was recorded (aliased to 'perf script' for now). + + 'perf sched replay' to simulate the workload that was recorded + via perf sched record. (this is done by starting up mockup threads + that mimic the workload based on the events in the trace. These + threads can then replay the timings (CPU runtime and sleep patterns) + of the workload as it occurred when it was recorded - and can repeat + it a number of times, measuring its performance.) + + 'perf sched map' to print a textual context-switching outline of + workload captured via perf sched record. Columns stand for + individual CPUs, and the two-letter shortcuts stand for tasks that + are running on a CPU. A '*' denotes the CPU that had the event, and + a dot signals an idle CPU. + +OPTIONS +------- +-i:: +--input=<file>:: + Input file name. (default: perf.data unless stdin is a fifo) + +-v:: +--verbose:: + Be more verbose. (show symbol address, etc) + +-D:: +--dump-raw-trace=:: + Display verbose dump of the sched data. + +SEE ALSO +-------- +linkperf:perf-record[1] diff --git a/tools/perf/Documentation/perf-script-perl.txt b/tools/perf/Documentation/perf-script-perl.txt new file mode 100644 index 00000000..d00bef23 --- /dev/null +++ b/tools/perf/Documentation/perf-script-perl.txt @@ -0,0 +1,216 @@ +perf-script-perl(1) +================== + +NAME +---- +perf-script-perl - Process trace data with a Perl script + +SYNOPSIS +-------- +[verse] +'perf script' [-s [Perl]:script[.pl] ] + +DESCRIPTION +----------- + +This perf script option is used to process perf script data using perf's +built-in Perl interpreter. It reads and processes the input file and +displays the results of the trace analysis implemented in the given +Perl script, if any. + +STARTER SCRIPTS +--------------- + +You can avoid reading the rest of this document by running 'perf script +-g perl' in the same directory as an existing perf.data trace file. +That will generate a starter script containing a handler for each of +the event types in the trace file; it simply prints every available +field for each event in the trace file. + +You can also look at the existing scripts in +~/libexec/perf-core/scripts/perl for typical examples showing how to +do basic things like aggregate event data, print results, etc. Also, +the check-perf-script.pl script, while not interesting for its results, +attempts to exercise all of the main scripting features. + +EVENT HANDLERS +-------------- + +When perf script is invoked using a trace script, a user-defined +'handler function' is called for each event in the trace. If there's +no handler function defined for a given event type, the event is +ignored (or passed to a 'trace_handled' function, see below) and the +next event is processed. + +Most of the event's field values are passed as arguments to the +handler function; some of the less common ones aren't - those are +available as calls back into the perf executable (see below). + +As an example, the following perf record command can be used to record +all sched_wakeup events in the system: + + # perf record -a -e sched:sched_wakeup + +Traces meant to be processed using a script should be recorded with +the above option: -a to enable system-wide collection. + +The format file for the sched_wakep event defines the following fields +(see /sys/kernel/debug/tracing/events/sched/sched_wakeup/format): + +---- + format: + field:unsigned short common_type; + field:unsigned char common_flags; + field:unsigned char common_preempt_count; + field:int common_pid; + + field:char comm[TASK_COMM_LEN]; + field:pid_t pid; + field:int prio; + field:int success; + field:int target_cpu; +---- + +The handler function for this event would be defined as: + +---- +sub sched::sched_wakeup +{ + my ($event_name, $context, $common_cpu, $common_secs, + $common_nsecs, $common_pid, $common_comm, + $comm, $pid, $prio, $success, $target_cpu) = @_; +} +---- + +The handler function takes the form subsystem::event_name. + +The $common_* arguments in the handler's argument list are the set of +arguments passed to all event handlers; some of the fields correspond +to the common_* fields in the format file, but some are synthesized, +and some of the common_* fields aren't common enough to to be passed +to every event as arguments but are available as library functions. + +Here's a brief description of each of the invariant event args: + + $event_name the name of the event as text + $context an opaque 'cookie' used in calls back into perf + $common_cpu the cpu the event occurred on + $common_secs the secs portion of the event timestamp + $common_nsecs the nsecs portion of the event timestamp + $common_pid the pid of the current task + $common_comm the name of the current process + +All of the remaining fields in the event's format file have +counterparts as handler function arguments of the same name, as can be +seen in the example above. + +The above provides the basics needed to directly access every field of +every event in a trace, which covers 90% of what you need to know to +write a useful trace script. The sections below cover the rest. + +SCRIPT LAYOUT +------------- + +Every perf script Perl script should start by setting up a Perl module +search path and 'use'ing a few support modules (see module +descriptions below): + +---- + use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib"; + use lib "./Perf-Trace-Util/lib"; + use Perf::Trace::Core; + use Perf::Trace::Context; + use Perf::Trace::Util; +---- + +The rest of the script can contain handler functions and support +functions in any order. + +Aside from the event handler functions discussed above, every script +can implement a set of optional functions: + +*trace_begin*, if defined, is called before any event is processed and +gives scripts a chance to do setup tasks: + +---- + sub trace_begin + { + } +---- + +*trace_end*, if defined, is called after all events have been + processed and gives scripts a chance to do end-of-script tasks, such + as display results: + +---- +sub trace_end +{ +} +---- + +*trace_unhandled*, if defined, is called after for any event that + doesn't have a handler explicitly defined for it. The standard set + of common arguments are passed into it: + +---- +sub trace_unhandled +{ + my ($event_name, $context, $common_cpu, $common_secs, + $common_nsecs, $common_pid, $common_comm) = @_; +} +---- + +The remaining sections provide descriptions of each of the available +built-in perf script Perl modules and their associated functions. + +AVAILABLE MODULES AND FUNCTIONS +------------------------------- + +The following sections describe the functions and variables available +via the various Perf::Trace::* Perl modules. To use the functions and +variables from the given module, add the corresponding 'use +Perf::Trace::XXX' line to your perf script script. + +Perf::Trace::Core Module +~~~~~~~~~~~~~~~~~~~~~~~~ + +These functions provide some essential functions to user scripts. + +The *flag_str* and *symbol_str* functions provide human-readable +strings for flag and symbolic fields. These correspond to the strings +and values parsed from the 'print fmt' fields of the event format +files: + + flag_str($event_name, $field_name, $field_value) - returns the string represention corresponding to $field_value for the flag field $field_name of event $event_name + symbol_str($event_name, $field_name, $field_value) - returns the string represention corresponding to $field_value for the symbolic field $field_name of event $event_name + +Perf::Trace::Context Module +~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +Some of the 'common' fields in the event format file aren't all that +common, but need to be made accessible to user scripts nonetheless. + +Perf::Trace::Context defines a set of functions that can be used to +access this data in the context of the current event. Each of these +functions expects a $context variable, which is the same as the +$context variable passed into every event handler as the second +argument. + + common_pc($context) - returns common_preempt count for the current event + common_flags($context) - returns common_flags for the current event + common_lock_depth($context) - returns common_lock_depth for the current event + +Perf::Trace::Util Module +~~~~~~~~~~~~~~~~~~~~~~~~ + +Various utility functions for use with perf script: + + nsecs($secs, $nsecs) - returns total nsecs given secs/nsecs pair + nsecs_secs($nsecs) - returns whole secs portion given nsecs + nsecs_nsecs($nsecs) - returns nsecs remainder given nsecs + nsecs_str($nsecs) - returns printable string in the form secs.nsecs + avg($total, $n) - returns average given a sum and a total number of values + +SEE ALSO +-------- +linkperf:perf-script[1] diff --git a/tools/perf/Documentation/perf-script-python.txt b/tools/perf/Documentation/perf-script-python.txt new file mode 100644 index 00000000..9f1f054b --- /dev/null +++ b/tools/perf/Documentation/perf-script-python.txt @@ -0,0 +1,620 @@ +perf-script-python(1) +==================== + +NAME +---- +perf-script-python - Process trace data with a Python script + +SYNOPSIS +-------- +[verse] +'perf script' [-s [Python]:script[.py] ] + +DESCRIPTION +----------- + +This perf script option is used to process perf script data using perf's +built-in Python interpreter. It reads and processes the input file and +displays the results of the trace analysis implemented in the given +Python script, if any. + +A QUICK EXAMPLE +--------------- + +This section shows the process, start to finish, of creating a working +Python script that aggregates and extracts useful information from a +raw perf script stream. You can avoid reading the rest of this +document if an example is enough for you; the rest of the document +provides more details on each step and lists the library functions +available to script writers. + +This example actually details the steps that were used to create the +'syscall-counts' script you see when you list the available perf script +scripts via 'perf script -l'. As such, this script also shows how to +integrate your script into the list of general-purpose 'perf script' +scripts listed by that command. + +The syscall-counts script is a simple script, but demonstrates all the +basic ideas necessary to create a useful script. Here's an example +of its output (syscall names are not yet supported, they will appear +as numbers): + +---- +syscall events: + +event count +---------------------------------------- ----------- +sys_write 455067 +sys_getdents 4072 +sys_close 3037 +sys_swapoff 1769 +sys_read 923 +sys_sched_setparam 826 +sys_open 331 +sys_newfstat 326 +sys_mmap 217 +sys_munmap 216 +sys_futex 141 +sys_select 102 +sys_poll 84 +sys_setitimer 12 +sys_writev 8 +15 8 +sys_lseek 7 +sys_rt_sigprocmask 6 +sys_wait4 3 +sys_ioctl 3 +sys_set_robust_list 1 +sys_exit 1 +56 1 +sys_access 1 +---- + +Basically our task is to keep a per-syscall tally that gets updated +every time a system call occurs in the system. Our script will do +that, but first we need to record the data that will be processed by +that script. Theoretically, there are a couple of ways we could do +that: + +- we could enable every event under the tracing/events/syscalls + directory, but this is over 600 syscalls, well beyond the number + allowable by perf. These individual syscall events will however be + useful if we want to later use the guidance we get from the + general-purpose scripts to drill down and get more detail about + individual syscalls of interest. + +- we can enable the sys_enter and/or sys_exit syscalls found under + tracing/events/raw_syscalls. These are called for all syscalls; the + 'id' field can be used to distinguish between individual syscall + numbers. + +For this script, we only need to know that a syscall was entered; we +don't care how it exited, so we'll use 'perf record' to record only +the sys_enter events: + +---- +# perf record -a -e raw_syscalls:sys_enter + +^C[ perf record: Woken up 1 times to write data ] +[ perf record: Captured and wrote 56.545 MB perf.data (~2470503 samples) ] +---- + +The options basically say to collect data for every syscall event +system-wide and multiplex the per-cpu output into a single stream. +That single stream will be recorded in a file in the current directory +called perf.data. + +Once we have a perf.data file containing our data, we can use the -g +'perf script' option to generate a Python script that will contain a +callback handler for each event type found in the perf.data trace +stream (for more details, see the STARTER SCRIPTS section). + +---- +# perf script -g python +generated Python script: perf-script.py + +The output file created also in the current directory is named +perf-script.py. Here's the file in its entirety: + +# perf script event handlers, generated by perf script -g python +# Licensed under the terms of the GNU GPL License version 2 + +# The common_* event handler fields are the most useful fields common to +# all events. They don't necessarily correspond to the 'common_*' fields +# in the format files. Those fields not available as handler params can +# be retrieved using Python functions of the form common_*(context). +# See the perf-script-python Documentation for the list of available functions. + +import os +import sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * + +def trace_begin(): + print "in trace_begin" + +def trace_end(): + print "in trace_end" + +def raw_syscalls__sys_enter(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + id, args): + print_header(event_name, common_cpu, common_secs, common_nsecs, + common_pid, common_comm) + + print "id=%d, args=%s\n" % \ + (id, args), + +def trace_unhandled(event_name, context, common_cpu, common_secs, common_nsecs, + common_pid, common_comm): + print_header(event_name, common_cpu, common_secs, common_nsecs, + common_pid, common_comm) + +def print_header(event_name, cpu, secs, nsecs, pid, comm): + print "%-20s %5u %05u.%09u %8u %-20s " % \ + (event_name, cpu, secs, nsecs, pid, comm), +---- + +At the top is a comment block followed by some import statements and a +path append which every perf script script should include. + +Following that are a couple generated functions, trace_begin() and +trace_end(), which are called at the beginning and the end of the +script respectively (for more details, see the SCRIPT_LAYOUT section +below). + +Following those are the 'event handler' functions generated one for +every event in the 'perf record' output. The handler functions take +the form subsystem__event_name, and contain named parameters, one for +each field in the event; in this case, there's only one event, +raw_syscalls__sys_enter(). (see the EVENT HANDLERS section below for +more info on event handlers). + +The final couple of functions are, like the begin and end functions, +generated for every script. The first, trace_unhandled(), is called +every time the script finds an event in the perf.data file that +doesn't correspond to any event handler in the script. This could +mean either that the record step recorded event types that it wasn't +really interested in, or the script was run against a trace file that +doesn't correspond to the script. + +The script generated by -g option simply prints a line for each +event found in the trace stream i.e. it basically just dumps the event +and its parameter values to stdout. The print_header() function is +simply a utility function used for that purpose. Let's rename the +script and run it to see the default output: + +---- +# mv perf-script.py syscall-counts.py +# perf script -s syscall-counts.py + +raw_syscalls__sys_enter 1 00840.847582083 7506 perf id=1, args= +raw_syscalls__sys_enter 1 00840.847595764 7506 perf id=1, args= +raw_syscalls__sys_enter 1 00840.847620860 7506 perf id=1, args= +raw_syscalls__sys_enter 1 00840.847710478 6533 npviewer.bin id=78, args= +raw_syscalls__sys_enter 1 00840.847719204 6533 npviewer.bin id=142, args= +raw_syscalls__sys_enter 1 00840.847755445 6533 npviewer.bin id=3, args= +raw_syscalls__sys_enter 1 00840.847775601 6533 npviewer.bin id=3, args= +raw_syscalls__sys_enter 1 00840.847781820 6533 npviewer.bin id=3, args= +. +. +. +---- + +Of course, for this script, we're not interested in printing every +trace event, but rather aggregating it in a useful way. So we'll get +rid of everything to do with printing as well as the trace_begin() and +trace_unhandled() functions, which we won't be using. That leaves us +with this minimalistic skeleton: + +---- +import os +import sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * + +def trace_end(): + print "in trace_end" + +def raw_syscalls__sys_enter(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + id, args): +---- + +In trace_end(), we'll simply print the results, but first we need to +generate some results to print. To do that we need to have our +sys_enter() handler do the necessary tallying until all events have +been counted. A hash table indexed by syscall id is a good way to +store that information; every time the sys_enter() handler is called, +we simply increment a count associated with that hash entry indexed by +that syscall id: + +---- + syscalls = autodict() + + try: + syscalls[id] += 1 + except TypeError: + syscalls[id] = 1 +---- + +The syscalls 'autodict' object is a special kind of Python dictionary +(implemented in Core.py) that implements Perl's 'autovivifying' hashes +in Python i.e. with autovivifying hashes, you can assign nested hash +values without having to go to the trouble of creating intermediate +levels if they don't exist e.g syscalls[comm][pid][id] = 1 will create +the intermediate hash levels and finally assign the value 1 to the +hash entry for 'id' (because the value being assigned isn't a hash +object itself, the initial value is assigned in the TypeError +exception. Well, there may be a better way to do this in Python but +that's what works for now). + +Putting that code into the raw_syscalls__sys_enter() handler, we +effectively end up with a single-level dictionary keyed on syscall id +and having the counts we've tallied as values. + +The print_syscall_totals() function iterates over the entries in the +dictionary and displays a line for each entry containing the syscall +name (the dictonary keys contain the syscall ids, which are passed to +the Util function syscall_name(), which translates the raw syscall +numbers to the corresponding syscall name strings). The output is +displayed after all the events in the trace have been processed, by +calling the print_syscall_totals() function from the trace_end() +handler called at the end of script processing. + +The final script producing the output shown above is shown in its +entirety below (syscall_name() helper is not yet available, you can +only deal with id's for now): + +---- +import os +import sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * +from Util import * + +syscalls = autodict() + +def trace_end(): + print_syscall_totals() + +def raw_syscalls__sys_enter(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + id, args): + try: + syscalls[id] += 1 + except TypeError: + syscalls[id] = 1 + +def print_syscall_totals(): + if for_comm is not None: + print "\nsyscall events for %s:\n\n" % (for_comm), + else: + print "\nsyscall events:\n\n", + + print "%-40s %10s\n" % ("event", "count"), + print "%-40s %10s\n" % ("----------------------------------------", \ + "-----------"), + + for id, val in sorted(syscalls.iteritems(), key = lambda(k, v): (v, k), \ + reverse = True): + print "%-40s %10d\n" % (syscall_name(id), val), +---- + +The script can be run just as before: + + # perf script -s syscall-counts.py + +So those are the essential steps in writing and running a script. The +process can be generalized to any tracepoint or set of tracepoints +you're interested in - basically find the tracepoint(s) you're +interested in by looking at the list of available events shown by +'perf list' and/or look in /sys/kernel/debug/tracing events for +detailed event and field info, record the corresponding trace data +using 'perf record', passing it the list of interesting events, +generate a skeleton script using 'perf script -g python' and modify the +code to aggregate and display it for your particular needs. + +After you've done that you may end up with a general-purpose script +that you want to keep around and have available for future use. By +writing a couple of very simple shell scripts and putting them in the +right place, you can have your script listed alongside the other +scripts listed by the 'perf script -l' command e.g.: + +---- +root@tropicana:~# perf script -l +List of available trace scripts: + wakeup-latency system-wide min/max/avg wakeup latency + rw-by-file <comm> r/w activity for a program, by file + rw-by-pid system-wide r/w activity +---- + +A nice side effect of doing this is that you also then capture the +probably lengthy 'perf record' command needed to record the events for +the script. + +To have the script appear as a 'built-in' script, you write two simple +scripts, one for recording and one for 'reporting'. + +The 'record' script is a shell script with the same base name as your +script, but with -record appended. The shell script should be put +into the perf/scripts/python/bin directory in the kernel source tree. +In that script, you write the 'perf record' command-line needed for +your script: + +---- +# cat kernel-source/tools/perf/scripts/python/bin/syscall-counts-record + +#!/bin/bash +perf record -a -e raw_syscalls:sys_enter +---- + +The 'report' script is also a shell script with the same base name as +your script, but with -report appended. It should also be located in +the perf/scripts/python/bin directory. In that script, you write the +'perf script -s' command-line needed for running your script: + +---- +# cat kernel-source/tools/perf/scripts/python/bin/syscall-counts-report + +#!/bin/bash +# description: system-wide syscall counts +perf script -s ~/libexec/perf-core/scripts/python/syscall-counts.py +---- + +Note that the location of the Python script given in the shell script +is in the libexec/perf-core/scripts/python directory - this is where +the script will be copied by 'make install' when you install perf. +For the installation to install your script there, your script needs +to be located in the perf/scripts/python directory in the kernel +source tree: + +---- +# ls -al kernel-source/tools/perf/scripts/python + +root@tropicana:/home/trz/src/tip# ls -al tools/perf/scripts/python +total 32 +drwxr-xr-x 4 trz trz 4096 2010-01-26 22:30 . +drwxr-xr-x 4 trz trz 4096 2010-01-26 22:29 .. +drwxr-xr-x 2 trz trz 4096 2010-01-26 22:29 bin +-rw-r--r-- 1 trz trz 2548 2010-01-26 22:29 check-perf-script.py +drwxr-xr-x 3 trz trz 4096 2010-01-26 22:49 Perf-Trace-Util +-rw-r--r-- 1 trz trz 1462 2010-01-26 22:30 syscall-counts.py +---- + +Once you've done that (don't forget to do a new 'make install', +otherwise your script won't show up at run-time), 'perf script -l' +should show a new entry for your script: + +---- +root@tropicana:~# perf script -l +List of available trace scripts: + wakeup-latency system-wide min/max/avg wakeup latency + rw-by-file <comm> r/w activity for a program, by file + rw-by-pid system-wide r/w activity + syscall-counts system-wide syscall counts +---- + +You can now perform the record step via 'perf script record': + + # perf script record syscall-counts + +and display the output using 'perf script report': + + # perf script report syscall-counts + +STARTER SCRIPTS +--------------- + +You can quickly get started writing a script for a particular set of +trace data by generating a skeleton script using 'perf script -g +python' in the same directory as an existing perf.data trace file. +That will generate a starter script containing a handler for each of +the event types in the trace file; it simply prints every available +field for each event in the trace file. + +You can also look at the existing scripts in +~/libexec/perf-core/scripts/python for typical examples showing how to +do basic things like aggregate event data, print results, etc. Also, +the check-perf-script.py script, while not interesting for its results, +attempts to exercise all of the main scripting features. + +EVENT HANDLERS +-------------- + +When perf script is invoked using a trace script, a user-defined +'handler function' is called for each event in the trace. If there's +no handler function defined for a given event type, the event is +ignored (or passed to a 'trace_handled' function, see below) and the +next event is processed. + +Most of the event's field values are passed as arguments to the +handler function; some of the less common ones aren't - those are +available as calls back into the perf executable (see below). + +As an example, the following perf record command can be used to record +all sched_wakeup events in the system: + + # perf record -a -e sched:sched_wakeup + +Traces meant to be processed using a script should be recorded with +the above option: -a to enable system-wide collection. + +The format file for the sched_wakep event defines the following fields +(see /sys/kernel/debug/tracing/events/sched/sched_wakeup/format): + +---- + format: + field:unsigned short common_type; + field:unsigned char common_flags; + field:unsigned char common_preempt_count; + field:int common_pid; + + field:char comm[TASK_COMM_LEN]; + field:pid_t pid; + field:int prio; + field:int success; + field:int target_cpu; +---- + +The handler function for this event would be defined as: + +---- +def sched__sched_wakeup(event_name, context, common_cpu, common_secs, + common_nsecs, common_pid, common_comm, + comm, pid, prio, success, target_cpu): + pass +---- + +The handler function takes the form subsystem__event_name. + +The common_* arguments in the handler's argument list are the set of +arguments passed to all event handlers; some of the fields correspond +to the common_* fields in the format file, but some are synthesized, +and some of the common_* fields aren't common enough to to be passed +to every event as arguments but are available as library functions. + +Here's a brief description of each of the invariant event args: + + event_name the name of the event as text + context an opaque 'cookie' used in calls back into perf + common_cpu the cpu the event occurred on + common_secs the secs portion of the event timestamp + common_nsecs the nsecs portion of the event timestamp + common_pid the pid of the current task + common_comm the name of the current process + +All of the remaining fields in the event's format file have +counterparts as handler function arguments of the same name, as can be +seen in the example above. + +The above provides the basics needed to directly access every field of +every event in a trace, which covers 90% of what you need to know to +write a useful trace script. The sections below cover the rest. + +SCRIPT LAYOUT +------------- + +Every perf script Python script should start by setting up a Python +module search path and 'import'ing a few support modules (see module +descriptions below): + +---- + import os + import sys + + sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + + from perf_trace_context import * + from Core import * +---- + +The rest of the script can contain handler functions and support +functions in any order. + +Aside from the event handler functions discussed above, every script +can implement a set of optional functions: + +*trace_begin*, if defined, is called before any event is processed and +gives scripts a chance to do setup tasks: + +---- +def trace_begin: + pass +---- + +*trace_end*, if defined, is called after all events have been + processed and gives scripts a chance to do end-of-script tasks, such + as display results: + +---- +def trace_end: + pass +---- + +*trace_unhandled*, if defined, is called after for any event that + doesn't have a handler explicitly defined for it. The standard set + of common arguments are passed into it: + +---- +def trace_unhandled(event_name, context, common_cpu, common_secs, + common_nsecs, common_pid, common_comm): + pass +---- + +The remaining sections provide descriptions of each of the available +built-in perf script Python modules and their associated functions. + +AVAILABLE MODULES AND FUNCTIONS +------------------------------- + +The following sections describe the functions and variables available +via the various perf script Python modules. To use the functions and +variables from the given module, add the corresponding 'from XXXX +import' line to your perf script script. + +Core.py Module +~~~~~~~~~~~~~~ + +These functions provide some essential functions to user scripts. + +The *flag_str* and *symbol_str* functions provide human-readable +strings for flag and symbolic fields. These correspond to the strings +and values parsed from the 'print fmt' fields of the event format +files: + + flag_str(event_name, field_name, field_value) - returns the string represention corresponding to field_value for the flag field field_name of event event_name + symbol_str(event_name, field_name, field_value) - returns the string represention corresponding to field_value for the symbolic field field_name of event event_name + +The *autodict* function returns a special kind of Python +dictionary that implements Perl's 'autovivifying' hashes in Python +i.e. with autovivifying hashes, you can assign nested hash values +without having to go to the trouble of creating intermediate levels if +they don't exist. + + autodict() - returns an autovivifying dictionary instance + + +perf_trace_context Module +~~~~~~~~~~~~~~~~~~~~~~~~~ + +Some of the 'common' fields in the event format file aren't all that +common, but need to be made accessible to user scripts nonetheless. + +perf_trace_context defines a set of functions that can be used to +access this data in the context of the current event. Each of these +functions expects a context variable, which is the same as the +context variable passed into every event handler as the second +argument. + + common_pc(context) - returns common_preempt count for the current event + common_flags(context) - returns common_flags for the current event + common_lock_depth(context) - returns common_lock_depth for the current event + +Util.py Module +~~~~~~~~~~~~~~ + +Various utility functions for use with perf script: + + nsecs(secs, nsecs) - returns total nsecs given secs/nsecs pair + nsecs_secs(nsecs) - returns whole secs portion given nsecs + nsecs_nsecs(nsecs) - returns nsecs remainder given nsecs + nsecs_str(nsecs) - returns printable string in the form secs.nsecs + avg(total, n) - returns average given a sum and a total number of values + +SEE ALSO +-------- +linkperf:perf-script[1] diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt new file mode 100644 index 00000000..e9cbfcdd --- /dev/null +++ b/tools/perf/Documentation/perf-script.txt @@ -0,0 +1,209 @@ +perf-script(1) +============= + +NAME +---- +perf-script - Read perf.data (created by perf record) and display trace output + +SYNOPSIS +-------- +[verse] +'perf script' [<options>] +'perf script' [<options>] record <script> [<record-options>] <command> +'perf script' [<options>] report <script> [script-args] +'perf script' [<options>] <script> <required-script-args> [<record-options>] <command> +'perf script' [<options>] <top-script> [script-args] + +DESCRIPTION +----------- +This command reads the input file and displays the trace recorded. + +There are several variants of perf script: + + 'perf script' to see a detailed trace of the workload that was + recorded. + + You can also run a set of pre-canned scripts that aggregate and + summarize the raw trace data in various ways (the list of scripts is + available via 'perf script -l'). The following variants allow you to + record and run those scripts: + + 'perf script record <script> <command>' to record the events required + for 'perf script report'. <script> is the name displayed in the + output of 'perf script --list' i.e. the actual script name minus any + language extension. If <command> is not specified, the events are + recorded using the -a (system-wide) 'perf record' option. + + 'perf script report <script> [args]' to run and display the results + of <script>. <script> is the name displayed in the output of 'perf + trace --list' i.e. the actual script name minus any language + extension. The perf.data output from a previous run of 'perf script + record <script>' is used and should be present for this command to + succeed. [args] refers to the (mainly optional) args expected by + the script. + + 'perf script <script> <required-script-args> <command>' to both + record the events required for <script> and to run the <script> + using 'live-mode' i.e. without writing anything to disk. <script> + is the name displayed in the output of 'perf script --list' i.e. the + actual script name minus any language extension. If <command> is + not specified, the events are recorded using the -a (system-wide) + 'perf record' option. If <script> has any required args, they + should be specified before <command>. This mode doesn't allow for + optional script args to be specified; if optional script args are + desired, they can be specified using separate 'perf script record' + and 'perf script report' commands, with the stdout of the record step + piped to the stdin of the report script, using the '-o -' and '-i -' + options of the corresponding commands. + + 'perf script <top-script>' to both record the events required for + <top-script> and to run the <top-script> using 'live-mode' + i.e. without writing anything to disk. <top-script> is the name + displayed in the output of 'perf script --list' i.e. the actual + script name minus any language extension; a <top-script> is defined + as any script name ending with the string 'top'. + + [<record-options>] can be passed to the record steps of 'perf script + record' and 'live-mode' variants; this isn't possible however for + <top-script> 'live-mode' or 'perf script report' variants. + + See the 'SEE ALSO' section for links to language-specific + information on how to write and run your own trace scripts. + +OPTIONS +------- +<command>...:: + Any command you can specify in a shell. + +-D:: +--dump-raw-script=:: + Display verbose dump of the trace data. + +-L:: +--Latency=:: + Show latency attributes (irqs/preemption disabled, etc). + +-l:: +--list=:: + Display a list of available trace scripts. + +-s ['lang']:: +--script=:: + Process trace data with the given script ([lang]:script[.ext]). + If the string 'lang' is specified in place of a script name, a + list of supported languages will be displayed instead. + +-g:: +--gen-script=:: + Generate perf-script.[ext] starter script for given language, + using current perf.data. + +-a:: + Force system-wide collection. Scripts run without a <command> + normally use -a by default, while scripts run with a <command> + normally don't - this option allows the latter to be run in + system-wide mode. + +-i:: +--input=:: + Input file name. (default: perf.data unless stdin is a fifo) + +-d:: +--debug-mode:: + Do various checks like samples ordering and lost events. + +-f:: +--fields:: + Comma separated list of fields to print. Options are: + comm, tid, pid, time, cpu, event, trace, ip, sym, dso, addr, symoff. + Field list can be prepended with the type, trace, sw or hw, + to indicate to which event type the field list applies. + e.g., -f sw:comm,tid,time,ip,sym and -f trace:time,cpu,trace + + perf script -f <fields> + + is equivalent to: + + perf script -f trace:<fields> -f sw:<fields> -f hw:<fields> + + i.e., the specified fields apply to all event types if the type string + is not given. + + The arguments are processed in the order received. A later usage can + reset a prior request. e.g.: + + -f trace: -f comm,tid,time,ip,sym + + The first -f suppresses trace events (field list is ""), but then the + second invocation sets the fields to comm,tid,time,ip,sym. In this case a + warning is given to the user: + + "Overriding previous field request for all events." + + Alternativey, consider the order: + + -f comm,tid,time,ip,sym -f trace: + + The first -f sets the fields for all events and the second -f + suppresses trace events. The user is given a warning message about + the override, and the result of the above is that only S/W and H/W + events are displayed with the given fields. + + For the 'wildcard' option if a user selected field is invalid for an + event type, a message is displayed to the user that the option is + ignored for that type. For example: + + $ perf script -f comm,tid,trace + 'trace' not valid for hardware events. Ignoring. + 'trace' not valid for software events. Ignoring. + + Alternatively, if the type is given an invalid field is specified it + is an error. For example: + + perf script -v -f sw:comm,tid,trace + 'trace' not valid for software events. + + At this point usage is displayed, and perf-script exits. + + Finally, a user may not set fields to none for all event types. + i.e., -f "" is not allowed. + +-k:: +--vmlinux=<file>:: + vmlinux pathname + +--kallsyms=<file>:: + kallsyms pathname + +--symfs=<directory>:: + Look for files with symbols relative to this directory. + +-G:: +--hide-call-graph:: + When printing symbols do not display call chain. + +-C:: +--cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can + be provided as a comma-separated list with no space: 0,1. Ranges of + CPUs are specified with -: 0-2. Default is to report samples on all + CPUs. + +-c:: +--comms=:: + Only display events for these comms. CSV that understands + file://filename entries. + +-I:: +--show-info:: + Display extended information about the perf.data file. This adds + information which may be very large and thus may clutter the display. + It currently includes: cpu and numa topology of the host system. + It can only be used with the perf script report mode. + +--show-kernel-path:: + Try to resolve the path of [kernel.kallsyms] + +SEE ALSO +-------- +linkperf:perf-record[1], linkperf:perf-script-perl[1], +linkperf:perf-script-python[1] diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt new file mode 100644 index 00000000..faf4f4fe --- /dev/null +++ b/tools/perf/Documentation/perf-stat.txt @@ -0,0 +1,149 @@ +perf-stat(1) +============ + +NAME +---- +perf-stat - Run a command and gather performance counter statistics + +SYNOPSIS +-------- +[verse] +'perf stat' [-e <EVENT> | --event=EVENT] [-a] <command> +'perf stat' [-e <EVENT> | --event=EVENT] [-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:: +--no-inherit:: + child tasks do not inherit counters +-p:: +--pid=<pid>:: + stat events on existing process id (comma separated list) + +-t:: +--tid=<tid>:: + stat events on existing thread id (comma separated list) + + +-a:: +--all-cpus:: + system-wide collection from all CPUs + +-c:: +--scale:: + scale/normalize counter values + +-r:: +--repeat=<n>:: + repeat command and print average + stddev (max: 100) + +-B:: +--big-num:: + print large numbers with thousands' separators according to locale + +-C:: +--cpu=:: +Count only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +In per-thread mode, this option is ignored. The -a option is still necessary +to activate system-wide monitoring. Default is to count on all CPUs. + +-A:: +--no-aggr:: +Do not aggregate counts across all monitored CPUs in system-wide mode (-a). +This option is only valid in system-wide mode. + +-n:: +--null:: + null run - don't start any counters + +-v:: +--verbose:: + be more verbose (show counter open errors, etc) + +-x SEP:: +--field-separator SEP:: +print counts using a CSV-style output to make it easy to import directly into +spreadsheets. Columns are separated by the string specified in SEP. + +-G name:: +--cgroup name:: +monitor only in the container (cgroup) called "name". This option is available only +in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to +container "name" are monitored when they run on the monitored CPUs. Multiple cgroups +can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup +to first event, second cgroup to second event and so on. It is possible to provide +an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have +corresponding events, i.e., they always refer to events defined earlier on the command +line. + +-o file:: +--output file:: +Print the output into the designated file. + +--append:: +Append to the output file designated with the -o option. Ignored if -o is not specified. + +--log-fd:: + +Log output to fd, instead of stderr. Complementary to --output, and mutually exclusive +with it. --append may be used here. Examples: + 3>results perf stat --log-fd 3 -- $cmd + 3>>results perf stat --log-fd 3 --append -- $cmd + +--pre:: +--post:: + Pre and post measurement hooks, e.g.: + +perf stat --repeat 10 --null --sync --pre 'make -s O=defconfig-build/clean' -- make -s -j64 O=defconfig-build/ bzImage + +-I msecs:: +--interval-print msecs:: + Print count deltas every N milliseconds (minimum: 100ms) + example: perf stat -I 1000 -e cycles -a sleep 5 + +--aggr-socket:: +Aggregate counts per processor socket for system-wide mode measurements. This +is a useful mode to detect imbalance between sockets. To enable this mode, +use --aggr-socket in addition to -a. (system-wide). The output includes the +socket number and the number of online processors on that socket. This is +useful to gauge the amount of aggregation. + +EXAMPLES +-------- + +$ 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-test.txt b/tools/perf/Documentation/perf-test.txt new file mode 100644 index 00000000..d1d3e512 --- /dev/null +++ b/tools/perf/Documentation/perf-test.txt @@ -0,0 +1,32 @@ +perf-test(1) +============ + +NAME +---- +perf-test - Runs sanity tests. + +SYNOPSIS +-------- +[verse] +'perf test [<options>] [{list <test-name-fragment>|[<test-name-fragments>|<test-numbers>]}]' + +DESCRIPTION +----------- +This command does assorted sanity tests, initially through linked routines but +also will look for a directory with more tests in the form of scripts. + +To get a list of available tests use 'perf test list', specifying a test name +fragment will show all tests that have it. + +To run just specific tests, inform test name fragments or the numbers obtained +from 'perf test list'. + +OPTIONS +------- +-s:: +--skip:: + Tests to skip (comma separater numeric list). + +-v:: +--verbose:: + Be more verbose. diff --git a/tools/perf/Documentation/perf-timechart.txt b/tools/perf/Documentation/perf-timechart.txt new file mode 100644 index 00000000..1632b0ef --- /dev/null +++ b/tools/perf/Documentation/perf-timechart.txt @@ -0,0 +1,46 @@ +perf-timechart(1) +================= + +NAME +---- +perf-timechart - Tool to visualize total system behavior during a workload + +SYNOPSIS +-------- +[verse] +'perf timechart' {record} + +DESCRIPTION +----------- +There are two variants of perf timechart: + + 'perf timechart record <command>' to record the system level events + of an arbitrary workload. + + 'perf timechart' to turn a trace into a Scalable Vector Graphics file, + that can be viewed with popular SVG viewers such as 'Inkscape'. + +OPTIONS +------- +-o:: +--output=:: + Select the output file (default: output.svg) +-i:: +--input=:: + Select the input file (default: perf.data unless stdin is a fifo) +-w:: +--width=:: + Select the width of the SVG file (default: 1000) +-P:: +--power-only:: + Only output the CPU power section of the diagram +-p:: +--process:: + Select the processes to display, by name or PID + +--symfs=<directory>:: + Look for files with symbols relative to this directory. + +SEE ALSO +-------- +linkperf:perf-record[1] diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt new file mode 100644 index 00000000..a414bc95 --- /dev/null +++ b/tools/perf/Documentation/perf-top.txt @@ -0,0 +1,193 @@ +perf-top(1) +=========== + +NAME +---- +perf-top - System profiling tool. + +SYNOPSIS +-------- +[verse] +'perf top' [-e <EVENT> | --event=EVENT] [<options>] + +DESCRIPTION +----------- +This command generates and displays a performance counter profile in real time. + + +OPTIONS +------- +-a:: +--all-cpus:: + System-wide collection. (default) + +-c <count>:: +--count=<count>:: + Event period to sample. + +-C <cpu-list>:: +--cpu=<cpu>:: +Monitor only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +Default is to monitor all CPUS. + +-d <seconds>:: +--delay=<seconds>:: + Number of seconds to delay between refreshes. + +-e <event>:: +--event=<event>:: + Select the PMU event. Selection can be a symbolic event name + (use 'perf list' to list all events) or a raw PMU + event (eventsel+umask) in the form of rNNN where NNN is a + hexadecimal event descriptor. + +-E <entries>:: +--entries=<entries>:: + Display this many functions. + +-f <count>:: +--count-filter=<count>:: + Only display functions with more events than this. + +-g:: +--group:: + Put the counters into a counter group. + +-F <freq>:: +--freq=<freq>:: + Profile at this frequency. + +-i:: +--inherit:: + Child tasks do not inherit counters. + +-k <path>:: +--vmlinux=<path>:: + Path to vmlinux. Required for annotation functionality. + +-m <pages>:: +--mmap-pages=<pages>:: + Number of mmapped data pages. + +-p <pid>:: +--pid=<pid>:: + Profile events on existing Process ID (comma separated list). + +-t <tid>:: +--tid=<tid>:: + Profile events on existing thread ID (comma separated list). + +-u:: +--uid=:: + Record events in threads owned by uid. Name or number. + +-r <priority>:: +--realtime=<priority>:: + Collect data with this RT SCHED_FIFO priority. + +-s <symbol>:: +--sym-annotate=<symbol>:: + Annotate this symbol. + +-K:: +--hide_kernel_symbols:: + Hide kernel symbols. + +-U:: +--hide_user_symbols:: + Hide user symbols. + +-D:: +--dump-symtab:: + Dump the symbol table used for profiling. + +-v:: +--verbose:: + Be more verbose (show counter open errors, etc). + +-z:: +--zero:: + Zero history across display updates. + +-s:: +--sort:: + Sort by key(s): pid, comm, dso, symbol, parent, srcline. + +-n:: +--show-nr-samples:: + Show a column with the number of samples. + +--show-total-period:: + Show a column with the sum of periods. + +--dsos:: + Only consider symbols in these dsos. + +--comms:: + Only consider symbols in these comms. + +--symbols:: + Only consider these symbols. + +-M:: +--disassembler-style=:: Set disassembler style for objdump. + +--source:: + Interleave source code with assembly code. Enabled by default, + disable with --no-source. + +--asm-raw:: + Show raw instruction encoding of assembly instructions. + +-G [type,min,order]:: +--call-graph:: + Display call chains using type, min percent threshold and order. + type can be either: + - flat: single column, linear exposure of call chains. + - graph: use a graph tree, displaying absolute overhead rates. + - fractal: like graph, but displays relative rates. Each branch of + the tree is considered as a new profiled object. + + order can be either: + - callee: callee based call graph. + - caller: inverted caller based call graph. + + Default: fractal,0.5,callee. + +INTERACTIVE PROMPTING KEYS +-------------------------- + +[d]:: + Display refresh delay. + +[e]:: + Number of entries to display. + +[E]:: + Event to display when multiple counters are active. + +[f]:: + Profile display filter (>= hit count). + +[F]:: + Annotation display filter (>= % of total). + +[s]:: + Annotate symbol. + +[S]:: + Stop annotation, return to full profile display. + +[z]:: + Toggle event count zeroing across display updates. + +[qQ]:: + Quit. + +Pressing any unmapped key displays a menu, and prompts for input. + + +SEE ALSO +-------- +linkperf:perf-stat[1], linkperf:perf-list[1] diff --git a/tools/perf/Documentation/perf-trace.txt b/tools/perf/Documentation/perf-trace.txt new file mode 100644 index 00000000..68718ccd --- /dev/null +++ b/tools/perf/Documentation/perf-trace.txt @@ -0,0 +1,59 @@ +perf-trace(1) +============= + +NAME +---- +perf-trace - strace inspired tool + +SYNOPSIS +-------- +[verse] +'perf trace' + +DESCRIPTION +----------- +This command will show the events associated with the target, initially +syscalls, but other system events like pagefaults, task lifetime events, +scheduling events, etc. + +Initially this is a live mode only tool, but eventually will work with +perf.data files like the other tools, allowing a detached 'record' from +analysis phases. + +OPTIONS +------- + +--all-cpus:: + System-wide collection from all CPUs. + +-p:: +--pid=:: + Record events on existing process ID (comma separated list). + +--tid=:: + Record events on existing thread ID (comma separated list). + +--uid=:: + Record events in threads owned by uid. Name or number. + +--no-inherit:: + Child tasks do not inherit counters. + +--mmap-pages=:: + Number of mmap data pages. Must be a power of two. + +--cpu:: +Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +In per-thread mode with inheritance mode on (default), Events are captured only when +the thread executes on the designated CPUs. Default is to monitor all CPUs. + +--duration: + Show only events that had a duration greater than N.M ms. + +--sched: + Accrue thread runtime and provide a summary at the end of the session. + +SEE ALSO +-------- +linkperf:perf-record[1], linkperf:perf-script[1] diff --git a/tools/perf/Documentation/perf.txt b/tools/perf/Documentation/perf.txt new file mode 100644 index 00000000..0eeb247d --- /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 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/Documentation/perfconfig.example b/tools/perf/Documentation/perfconfig.example new file mode 100644 index 00000000..767ea243 --- /dev/null +++ b/tools/perf/Documentation/perfconfig.example @@ -0,0 +1,29 @@ +[colors] + + # These were the old defaults + top = red, lightgray + medium = green, lightgray + normal = black, lightgray + selected = lightgray, magenta + code = blue, lightgray + addr = magenta, lightgray + +[tui] + + # Defaults if linked with libslang + report = on + annotate = on + top = on + +[buildid] + + # Default, disable using /dev/null + dir = /root/.debug + +[annotate] + + # Defaults + hide_src_code = false + use_offset = true + jump_arrows = true + show_nr_jumps = false diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST new file mode 100644 index 00000000..39d41068 --- /dev/null +++ b/tools/perf/MANIFEST @@ -0,0 +1,31 @@ +tools/perf +tools/scripts +tools/lib/traceevent +include/linux/const.h +include/linux/perf_event.h +include/linux/rbtree.h +include/linux/list.h +include/linux/hash.h +include/linux/stringify.h +lib/rbtree.c +include/linux/swab.h +arch/*/include/asm/unistd*.h +arch/*/include/asm/perf_regs.h +arch/*/include/uapi/asm/unistd*.h +arch/*/include/uapi/asm/perf_regs.h +arch/*/lib/memcpy*.S +arch/*/lib/memset*.S +include/linux/poison.h +include/linux/magic.h +include/linux/hw_breakpoint.h +include/linux/rbtree_augmented.h +include/uapi/linux/perf_event.h +include/uapi/linux/const.h +include/uapi/linux/swab.h +include/uapi/linux/hw_breakpoint.h +arch/x86/include/asm/svm.h +arch/x86/include/asm/vmx.h +arch/x86/include/asm/kvm_host.h +arch/x86/include/uapi/asm/svm.h +arch/x86/include/uapi/asm/vmx.h +arch/x86/include/uapi/asm/kvm.h diff --git a/tools/perf/Makefile b/tools/perf/Makefile new file mode 100644 index 00000000..fb5291d4 --- /dev/null +++ b/tools/perf/Makefile @@ -0,0 +1,1182 @@ +include ../scripts/Makefile.include + +# The default target of this Makefile is... +all: + +include config/utilities.mak + +# Define V to have a more verbose compile. +# +# Define O to save output files in a separate directory. +# +# Define ARCH as name of target architecture if you want cross-builds. +# +# Define CROSS_COMPILE as prefix name of compiler if you want cross-builds. +# +# Define NO_LIBPERL to disable perl script extension. +# +# Define NO_LIBPYTHON to disable python script extension. +# +# Define PYTHON to point to the python binary if the default +# `python' is not correct; for example: PYTHON=python2 +# +# Define PYTHON_CONFIG to point to the python-config binary if +# the default `$(PYTHON)-config' is not correct. +# +# Define ASCIIDOC8 if you want to format documentation with AsciiDoc 8 +# +# Define DOCBOOK_XSL_172 if you want to format man pages with DocBook XSL v1.72. +# +# Define LDFLAGS=-static to build a static binary. +# +# Define EXTRA_CFLAGS=-m64 or EXTRA_CFLAGS=-m32 as appropriate for cross-builds. +# +# Define NO_DWARF if you do not want debug-info analysis feature at all. +# +# Define WERROR=0 to disable treating any warnings as errors. +# +# Define NO_NEWT if you do not want TUI support. +# +# Define NO_GTK2 if you do not want GTK+ GUI support. +# +# Define NO_DEMANGLE if you do not want C++ symbol demangling. +# +# Define NO_LIBELF if you do not want libelf dependency (e.g. cross-builds) +# +# Define NO_LIBUNWIND if you do not want libunwind dependency for dwarf +# backtrace post unwind. +# +# Define NO_BACKTRACE if you do not want stack backtrace debug feature +# +# Define NO_LIBNUMA if you do not want numa perf benchmark + +$(OUTPUT)PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE + @$(SHELL_PATH) util/PERF-VERSION-GEN $(OUTPUT) + +uname_M := $(shell uname -m 2>/dev/null || echo not) + +ARCH ?= $(shell echo $(uname_M) | sed -e s/i.86/i386/ -e s/sun4u/sparc64/ \ + -e s/arm.*/arm/ -e s/sa110/arm/ \ + -e s/s390x/s390/ -e s/parisc64/parisc/ \ + -e s/ppc.*/powerpc/ -e s/mips.*/mips/ \ + -e s/sh[234].*/sh/ -e s/aarch64.*/arm64/ ) +NO_PERF_REGS := 1 + +CC = $(CROSS_COMPILE)gcc +AR = $(CROSS_COMPILE)ar + +# Additional ARCH settings for x86 +ifeq ($(ARCH),i386) + override ARCH := x86 + NO_PERF_REGS := 0 + LIBUNWIND_LIBS = -lunwind -lunwind-x86 +endif +ifeq ($(ARCH),x86_64) + override ARCH := x86 + IS_X86_64 := 0 + ifeq (, $(findstring m32,$(EXTRA_CFLAGS))) + IS_X86_64 := $(shell echo __x86_64__ | ${CC} -E -x c - | tail -n 1) + endif + ifeq (${IS_X86_64}, 1) + RAW_ARCH := x86_64 + ARCH_CFLAGS := -DARCH_X86_64 + ARCH_INCLUDE = ../../arch/x86/lib/memcpy_64.S ../../arch/x86/lib/memset_64.S + endif + NO_PERF_REGS := 0 + LIBUNWIND_LIBS = -lunwind -lunwind-x86_64 +endif + +# Treat warnings as errors unless directed not to +ifneq ($(WERROR),0) + CFLAGS_WERROR := -Werror +endif + +ifeq ("$(origin DEBUG)", "command line") + PERF_DEBUG = $(DEBUG) +endif +ifndef PERF_DEBUG + CFLAGS_OPTIMIZE = -O6 -D_FORTIFY_SOURCE=2 +endif + +ifdef PARSER_DEBUG + PARSER_DEBUG_BISON := -t + PARSER_DEBUG_FLEX := -d + PARSER_DEBUG_CFLAGS := -DPARSER_DEBUG +endif + +CFLAGS = $(ANDROID_CFLAGS) -fno-omit-frame-pointer -ggdb3 -funwind-tables -Wall -Wextra -std=gnu99 $(CFLAGS_WERROR) $(CFLAGS_OPTIMIZE) $(EXTRA_WARNINGS) $(EXTRA_CFLAGS) $(PARSER_DEBUG_CFLAGS) +EXTLIBS = -lpthread -lrt -lelf -lm +ALL_CFLAGS = $(CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE +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. + +# Make the path relative to DESTDIR, not to prefix +ifndef DESTDIR +prefix = $(HOME) +endif +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 = $(prefix)/etc +ETC_PERFCONFIG = etc/perfconfig +endif +lib = lib + +export prefix bindir sharedir sysconfdir + +RM = rm -f +MKDIR = mkdir +FIND = find +INSTALL = install +FLEX = flex +BISON= bison + +# sparse is architecture-neutral, which means that we need to tell it +# explicitly what architecture to check for. Fix this up for yours.. +SPARSE_FLAGS = -D__BIG_ENDIAN__ -D__powerpc__ + +ifneq ($(MAKECMDGOALS),clean) +ifneq ($(MAKECMDGOALS),tags) +-include config/feature-tests.mak + +ifeq ($(call get-executable,$(FLEX)),) + dummy := $(error Error: $(FLEX) is missing on this system, please install it) +endif + +ifeq ($(call get-executable,$(BISON)),) + dummy := $(error Error: $(BISON) is missing on this system, please install it) +endif + +ifeq ($(call try-cc,$(SOURCE_HELLO),$(CFLAGS) -Werror -fstack-protector-all,-fstack-protector-all),y) + CFLAGS := $(CFLAGS) -fstack-protector-all +endif + +ifeq ($(call try-cc,$(SOURCE_HELLO),$(CFLAGS) -Werror -Wstack-protector,-Wstack-protector),y) + CFLAGS := $(CFLAGS) -Wstack-protector +endif + +ifeq ($(call try-cc,$(SOURCE_HELLO),$(CFLAGS) -Werror -Wvolatile-register-var,-Wvolatile-register-var),y) + CFLAGS := $(CFLAGS) -Wvolatile-register-var +endif + +### --- END CONFIGURATION SECTION --- + +ifeq ($(srctree),) +srctree := $(patsubst %/,%,$(dir $(shell pwd))) +srctree := $(patsubst %/,%,$(dir $(srctree))) +#$(info Determined 'srctree' to be $(srctree)) +endif + +ifneq ($(objtree),) +#$(info Determined 'objtree' to be $(objtree)) +endif + +ifneq ($(OUTPUT),) +#$(info Determined 'OUTPUT' to be $(OUTPUT)) +endif + +BASIC_CFLAGS = \ + -Iutil/include \ + -Iarch/$(ARCH)/include \ + $(if $(objtree),-I$(objtree)/arch/$(ARCH)/include/generated/uapi) \ + -I$(srctree)/arch/$(ARCH)/include/uapi \ + -I$(srctree)/arch/$(ARCH)/include \ + $(if $(objtree),-I$(objtree)/include/generated/uapi) \ + -I$(srctree)/include/uapi \ + -I$(srctree)/include \ + -I$(OUTPUT)util \ + -Iutil \ + -I. \ + -I$(TRACE_EVENT_DIR) \ + -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE + +BASIC_LDFLAGS = + +ifeq ($(call try-cc,$(SOURCE_BIONIC),$(CFLAGS),bionic),y) + BIONIC := 1 + EXTLIBS := $(filter-out -lrt,$(EXTLIBS)) + EXTLIBS := $(filter-out -lpthread,$(EXTLIBS)) + ANDROID_CFLAGS += -I. -DANDROID -DHAVE_STRLCPY -include $(CURDIR)/compat-android.h + CFLAGS += -I. -DANDROID -DHAVE_STRLCPY -include $(CURDIR)/compat-android.h +endif +endif # MAKECMDGOALS != tags +endif # MAKECMDGOALS != clean + +# Guard against environment variables +BUILTIN_OBJS = +LIB_H = +LIB_OBJS = +PYRF_OBJS = +SCRIPT_SH = + +SCRIPT_SH += perf-archive.sh + +grep-libs = $(filter -l%,$(1)) +strip-libs = $(filter-out -l%,$(1)) + +TRACE_EVENT_DIR = ../lib/traceevent/ + +ifneq ($(OUTPUT),) + TE_PATH=$(OUTPUT) +else + TE_PATH=$(TRACE_EVENT_DIR) +endif + +LIBTRACEEVENT = $(TE_PATH)libtraceevent.a +TE_LIB := -L$(TE_PATH) -ltraceevent + +export LIBTRACEEVENT + +# python extension build directories +PYTHON_EXTBUILD := $(OUTPUT)python_ext_build/ +PYTHON_EXTBUILD_LIB := $(PYTHON_EXTBUILD)lib/ +PYTHON_EXTBUILD_TMP := $(PYTHON_EXTBUILD)tmp/ +export PYTHON_EXTBUILD_LIB PYTHON_EXTBUILD_TMP + +python-clean := rm -rf $(PYTHON_EXTBUILD) $(OUTPUT)python/perf.so + +PYTHON_EXT_SRCS := $(shell grep -v ^\# util/python-ext-sources) +PYTHON_EXT_DEPS := util/python-ext-sources util/setup.py + +$(OUTPUT)python/perf.so: $(PYTHON_EXT_SRCS) $(PYTHON_EXT_DEPS) + $(QUIET_GEN)CFLAGS='$(BASIC_CFLAGS)' $(PYTHON_WORD) util/setup.py \ + --quiet build_ext; \ + mkdir -p $(OUTPUT)python && \ + cp $(PYTHON_EXTBUILD_LIB)perf.so $(OUTPUT)python/ +# +# No Perl scripts right now: +# + +SCRIPTS = $(patsubst %.sh,%,$(SCRIPT_SH)) + +# +# Single 'perf' binary right now: +# +PROGRAMS += $(OUTPUT)perf + +LANG_BINDINGS = + +# 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 = $(OUTPUT)perf + +# Set paths to tools early so that they can be used for version tests. +ifndef SHELL_PATH + SHELL_PATH = /bin/sh +endif +ifndef PERL_PATH + PERL_PATH = /usr/bin/perl +endif + +export PERL_PATH + +$(OUTPUT)util/parse-events-flex.c: util/parse-events.l $(OUTPUT)util/parse-events-bison.c + $(QUIET_FLEX)$(FLEX) --header-file=$(OUTPUT)util/parse-events-flex.h $(PARSER_DEBUG_FLEX) -t util/parse-events.l > $(OUTPUT)util/parse-events-flex.c + +$(OUTPUT)util/parse-events-bison.c: util/parse-events.y + $(QUIET_BISON)$(BISON) -v util/parse-events.y -d $(PARSER_DEBUG_BISON) -o $(OUTPUT)util/parse-events-bison.c -p parse_events_ + +$(OUTPUT)util/pmu-flex.c: util/pmu.l $(OUTPUT)util/pmu-bison.c + $(QUIET_FLEX)$(FLEX) --header-file=$(OUTPUT)util/pmu-flex.h -t util/pmu.l > $(OUTPUT)util/pmu-flex.c + +$(OUTPUT)util/pmu-bison.c: util/pmu.y + $(QUIET_BISON)$(BISON) -v util/pmu.y -d -o $(OUTPUT)util/pmu-bison.c -p perf_pmu_ + +$(OUTPUT)util/parse-events.o: $(OUTPUT)util/parse-events-flex.c $(OUTPUT)util/parse-events-bison.c +$(OUTPUT)util/pmu.o: $(OUTPUT)util/pmu-flex.c $(OUTPUT)util/pmu-bison.c + +LIB_FILE=$(OUTPUT)libperf.a + +LIB_H += ../../include/uapi/linux/perf_event.h +LIB_H += ../../include/linux/rbtree.h +LIB_H += ../../include/linux/list.h +LIB_H += ../../include/uapi/linux/const.h +LIB_H += ../../include/linux/hash.h +LIB_H += ../../include/linux/stringify.h +LIB_H += util/include/linux/bitmap.h +LIB_H += util/include/linux/bitops.h +LIB_H += util/include/linux/compiler.h +LIB_H += util/include/linux/const.h +LIB_H += util/include/linux/ctype.h +LIB_H += util/include/linux/kernel.h +LIB_H += util/include/linux/list.h +LIB_H += util/include/linux/export.h +LIB_H += util/include/linux/magic.h +LIB_H += util/include/linux/poison.h +LIB_H += util/include/linux/prefetch.h +LIB_H += util/include/linux/rbtree.h +LIB_H += util/include/linux/rbtree_augmented.h +LIB_H += util/include/linux/string.h +LIB_H += util/include/linux/types.h +LIB_H += util/include/linux/linkage.h +LIB_H += util/include/asm/asm-offsets.h +LIB_H += util/include/asm/bug.h +LIB_H += util/include/asm/byteorder.h +LIB_H += util/include/asm/hweight.h +LIB_H += util/include/asm/swab.h +LIB_H += util/include/asm/system.h +LIB_H += util/include/asm/uaccess.h +LIB_H += util/include/dwarf-regs.h +LIB_H += util/include/asm/dwarf2.h +LIB_H += util/include/asm/cpufeature.h +LIB_H += util/include/asm/unistd_32.h +LIB_H += util/include/asm/unistd_64.h +LIB_H += perf.h +LIB_H += util/annotate.h +LIB_H += util/cache.h +LIB_H += util/callchain.h +LIB_H += util/build-id.h +LIB_H += util/debug.h +LIB_H += util/debugfs.h +LIB_H += util/sysfs.h +LIB_H += util/pmu.h +LIB_H += util/event.h +LIB_H += util/evsel.h +LIB_H += util/evlist.h +LIB_H += util/exec_cmd.h +LIB_H += util/types.h +LIB_H += util/levenshtein.h +LIB_H += util/machine.h +LIB_H += util/map.h +LIB_H += util/parse-options.h +LIB_H += util/parse-events.h +LIB_H += util/quote.h +LIB_H += util/util.h +LIB_H += util/xyarray.h +LIB_H += util/header.h +LIB_H += util/help.h +LIB_H += util/session.h +LIB_H += util/strbuf.h +LIB_H += util/strlist.h +LIB_H += util/strfilter.h +LIB_H += util/svghelper.h +LIB_H += util/tool.h +LIB_H += util/run-command.h +LIB_H += util/sigchain.h +LIB_H += util/dso.h +LIB_H += util/symbol.h +LIB_H += util/color.h +LIB_H += util/values.h +LIB_H += util/sort.h +LIB_H += util/hist.h +LIB_H += util/thread.h +LIB_H += util/thread_map.h +LIB_H += util/trace-event.h +LIB_H += util/probe-finder.h +LIB_H += util/dwarf-aux.h +LIB_H += util/probe-event.h +LIB_H += util/pstack.h +LIB_H += util/cpumap.h +LIB_H += util/top.h +LIB_H += $(ARCH_INCLUDE) +LIB_H += util/cgroup.h +LIB_H += $(TRACE_EVENT_DIR)event-parse.h +LIB_H += util/target.h +LIB_H += util/rblist.h +LIB_H += util/intlist.h +LIB_H += util/perf_regs.h +LIB_H += util/unwind.h +LIB_H += util/vdso.h +LIB_H += ui/helpline.h +LIB_H += ui/progress.h +LIB_H += ui/util.h +LIB_H += ui/ui.h + +LIB_OBJS += $(OUTPUT)util/abspath.o +LIB_OBJS += $(OUTPUT)util/alias.o +LIB_OBJS += $(OUTPUT)util/annotate.o +LIB_OBJS += $(OUTPUT)util/build-id.o +LIB_OBJS += $(OUTPUT)util/config.o +LIB_OBJS += $(OUTPUT)util/ctype.o +LIB_OBJS += $(OUTPUT)util/debugfs.o +LIB_OBJS += $(OUTPUT)util/sysfs.o +LIB_OBJS += $(OUTPUT)util/pmu.o +LIB_OBJS += $(OUTPUT)util/environment.o +LIB_OBJS += $(OUTPUT)util/event.o +LIB_OBJS += $(OUTPUT)util/evlist.o +LIB_OBJS += $(OUTPUT)util/evsel.o +LIB_OBJS += $(OUTPUT)util/exec_cmd.o +LIB_OBJS += $(OUTPUT)util/help.o +LIB_OBJS += $(OUTPUT)util/levenshtein.o +LIB_OBJS += $(OUTPUT)util/parse-options.o +LIB_OBJS += $(OUTPUT)util/parse-events.o +LIB_OBJS += $(OUTPUT)util/path.o +LIB_OBJS += $(OUTPUT)util/rbtree.o +LIB_OBJS += $(OUTPUT)util/bitmap.o +LIB_OBJS += $(OUTPUT)util/hweight.o +LIB_OBJS += $(OUTPUT)util/run-command.o +LIB_OBJS += $(OUTPUT)util/quote.o +LIB_OBJS += $(OUTPUT)util/strbuf.o +LIB_OBJS += $(OUTPUT)util/string.o +LIB_OBJS += $(OUTPUT)util/strlist.o +LIB_OBJS += $(OUTPUT)util/strfilter.o +LIB_OBJS += $(OUTPUT)util/top.o +LIB_OBJS += $(OUTPUT)util/usage.o +LIB_OBJS += $(OUTPUT)util/wrapper.o +LIB_OBJS += $(OUTPUT)util/sigchain.o +LIB_OBJS += $(OUTPUT)util/dso.o +LIB_OBJS += $(OUTPUT)util/symbol.o +LIB_OBJS += $(OUTPUT)util/symbol-elf.o +LIB_OBJS += $(OUTPUT)util/color.o +LIB_OBJS += $(OUTPUT)util/pager.o +LIB_OBJS += $(OUTPUT)util/header.o +LIB_OBJS += $(OUTPUT)util/callchain.o +LIB_OBJS += $(OUTPUT)util/values.o +LIB_OBJS += $(OUTPUT)util/debug.o +LIB_OBJS += $(OUTPUT)util/machine.o +LIB_OBJS += $(OUTPUT)util/map.o +LIB_OBJS += $(OUTPUT)util/pstack.o +LIB_OBJS += $(OUTPUT)util/session.o +LIB_OBJS += $(OUTPUT)util/thread.o +LIB_OBJS += $(OUTPUT)util/thread_map.o +LIB_OBJS += $(OUTPUT)util/trace-event-parse.o +LIB_OBJS += $(OUTPUT)util/parse-events-flex.o +LIB_OBJS += $(OUTPUT)util/parse-events-bison.o +LIB_OBJS += $(OUTPUT)util/pmu-flex.o +LIB_OBJS += $(OUTPUT)util/pmu-bison.o +LIB_OBJS += $(OUTPUT)util/trace-event-read.o +LIB_OBJS += $(OUTPUT)util/trace-event-info.o +LIB_OBJS += $(OUTPUT)util/trace-event-scripting.o +LIB_OBJS += $(OUTPUT)util/svghelper.o +LIB_OBJS += $(OUTPUT)util/sort.o +LIB_OBJS += $(OUTPUT)util/hist.o +LIB_OBJS += $(OUTPUT)util/probe-event.o +LIB_OBJS += $(OUTPUT)util/util.o +LIB_OBJS += $(OUTPUT)util/xyarray.o +LIB_OBJS += $(OUTPUT)util/cpumap.o +LIB_OBJS += $(OUTPUT)util/cgroup.o +LIB_OBJS += $(OUTPUT)util/target.o +LIB_OBJS += $(OUTPUT)util/rblist.o +LIB_OBJS += $(OUTPUT)util/intlist.o +LIB_OBJS += $(OUTPUT)util/vdso.o +LIB_OBJS += $(OUTPUT)util/stat.o + +LIB_OBJS += $(OUTPUT)ui/setup.o +LIB_OBJS += $(OUTPUT)ui/helpline.o +LIB_OBJS += $(OUTPUT)ui/progress.o +LIB_OBJS += $(OUTPUT)ui/util.o +LIB_OBJS += $(OUTPUT)ui/hist.o +LIB_OBJS += $(OUTPUT)ui/stdio/hist.o + +LIB_OBJS += $(OUTPUT)arch/common.o + +LIB_OBJS += $(OUTPUT)tests/parse-events.o +LIB_OBJS += $(OUTPUT)tests/dso-data.o +LIB_OBJS += $(OUTPUT)tests/attr.o +LIB_OBJS += $(OUTPUT)tests/vmlinux-kallsyms.o +LIB_OBJS += $(OUTPUT)tests/open-syscall.o +LIB_OBJS += $(OUTPUT)tests/open-syscall-all-cpus.o +LIB_OBJS += $(OUTPUT)tests/open-syscall-tp-fields.o +LIB_OBJS += $(OUTPUT)tests/mmap-basic.o +LIB_OBJS += $(OUTPUT)tests/perf-record.o +LIB_OBJS += $(OUTPUT)tests/rdpmc.o +LIB_OBJS += $(OUTPUT)tests/evsel-roundtrip-name.o +LIB_OBJS += $(OUTPUT)tests/evsel-tp-sched.o +LIB_OBJS += $(OUTPUT)tests/pmu.o +LIB_OBJS += $(OUTPUT)tests/hists_link.o +LIB_OBJS += $(OUTPUT)tests/python-use.o + +BUILTIN_OBJS += $(OUTPUT)builtin-annotate.o +BUILTIN_OBJS += $(OUTPUT)builtin-bench.o +# Benchmark modules +BUILTIN_OBJS += $(OUTPUT)bench/sched-messaging.o +BUILTIN_OBJS += $(OUTPUT)bench/sched-pipe.o +ifeq ($(RAW_ARCH),x86_64) +BUILTIN_OBJS += $(OUTPUT)bench/mem-memcpy-x86-64-asm.o +BUILTIN_OBJS += $(OUTPUT)bench/mem-memset-x86-64-asm.o +endif +BUILTIN_OBJS += $(OUTPUT)bench/mem-memcpy.o +BUILTIN_OBJS += $(OUTPUT)bench/mem-memset.o + +BUILTIN_OBJS += $(OUTPUT)builtin-diff.o +BUILTIN_OBJS += $(OUTPUT)builtin-evlist.o +BUILTIN_OBJS += $(OUTPUT)builtin-help.o +BUILTIN_OBJS += $(OUTPUT)builtin-sched.o +BUILTIN_OBJS += $(OUTPUT)builtin-buildid-list.o +BUILTIN_OBJS += $(OUTPUT)builtin-buildid-cache.o +BUILTIN_OBJS += $(OUTPUT)builtin-list.o +BUILTIN_OBJS += $(OUTPUT)builtin-record.o +BUILTIN_OBJS += $(OUTPUT)builtin-report.o +BUILTIN_OBJS += $(OUTPUT)builtin-stat.o +BUILTIN_OBJS += $(OUTPUT)builtin-timechart.o +BUILTIN_OBJS += $(OUTPUT)builtin-top.o +BUILTIN_OBJS += $(OUTPUT)builtin-script.o +BUILTIN_OBJS += $(OUTPUT)builtin-probe.o +BUILTIN_OBJS += $(OUTPUT)builtin-kmem.o +BUILTIN_OBJS += $(OUTPUT)builtin-lock.o +BUILTIN_OBJS += $(OUTPUT)builtin-kvm.o +BUILTIN_OBJS += $(OUTPUT)builtin-inject.o +BUILTIN_OBJS += $(OUTPUT)tests/builtin-test.o + +PERFLIBS = $(LIB_FILE) $(LIBTRACEEVENT) + +# +# Platform specific tweaks +# +ifneq ($(MAKECMDGOALS),clean) +ifneq ($(MAKECMDGOALS),tags) + +# We choose to avoid "if .. else if .. else .. endif endif" +# because maintaining the nesting to match is a pain. If +# we had "elif" things would have been much nicer... + +ifdef NO_LIBELF + NO_DWARF := 1 + NO_DEMANGLE := 1 + NO_LIBUNWIND := 1 +else +FLAGS_LIBELF=$(ALL_CFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) +ifneq ($(call try-cc,$(SOURCE_LIBELF),$(FLAGS_LIBELF),libelf),y) + FLAGS_GLIBC=$(ALL_CFLAGS) $(ALL_LDFLAGS) + ifeq ($(call try-cc,$(SOURCE_GLIBC),$(FLAGS_GLIBC),glibc),y) + LIBC_SUPPORT := 1 + endif + ifeq ($(BIONIC),1) + LIBC_SUPPORT := 1 + endif + ifeq ($(LIBC_SUPPORT),1) + msg := $(warning No libelf found, disables 'probe' tool, please install elfutils-libelf-devel/libelf-dev); + + NO_LIBELF := 1 + NO_DWARF := 1 + NO_DEMANGLE := 1 + else + msg := $(error No gnu/libc-version.h found, please install glibc-dev[el]/glibc-static); + endif +else + # for linking with debug library, run like: + # make DEBUG=1 LIBDW_DIR=/opt/libdw/ + ifdef LIBDW_DIR + LIBDW_CFLAGS := -I$(LIBDW_DIR)/include + LIBDW_LDFLAGS := -L$(LIBDW_DIR)/lib + endif + + FLAGS_DWARF=$(ALL_CFLAGS) $(LIBDW_CFLAGS) -ldw -lelf $(LIBDW_LDFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) + ifneq ($(call try-cc,$(SOURCE_DWARF),$(FLAGS_DWARF),libdw),y) + msg := $(warning No libdw.h found or old libdw.h found or elfutils is older than 0.138, disables dwarf support. Please install new elfutils-devel/libdw-dev); + NO_DWARF := 1 + endif # Dwarf support +endif # SOURCE_LIBELF +endif # NO_LIBELF + +# There's only x86 (both 32 and 64) support for CFI unwind so far +ifneq ($(ARCH),x86) + NO_LIBUNWIND := 1 +endif + +ifndef NO_LIBUNWIND +# for linking with debug library, run like: +# make DEBUG=1 LIBUNWIND_DIR=/opt/libunwind/ +ifdef LIBUNWIND_DIR + LIBUNWIND_CFLAGS := -I$(LIBUNWIND_DIR)/include + LIBUNWIND_LDFLAGS := -L$(LIBUNWIND_DIR)/lib +endif + +FLAGS_UNWIND=$(LIBUNWIND_CFLAGS) $(ALL_CFLAGS) $(LIBUNWIND_LDFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) $(LIBUNWIND_LIBS) +ifneq ($(call try-cc,$(SOURCE_LIBUNWIND),$(FLAGS_UNWIND),libunwind),y) + msg := $(warning No libunwind found, disabling post unwind support. Please install libunwind-dev[el] >= 0.99); + NO_LIBUNWIND := 1 +endif # Libunwind support +endif # NO_LIBUNWIND + +-include arch/$(ARCH)/Makefile + +ifneq ($(OUTPUT),) + BASIC_CFLAGS += -I$(OUTPUT) +endif + +ifdef NO_LIBELF +EXTLIBS := $(filter-out -lelf,$(EXTLIBS)) + +# Remove ELF/DWARF dependent codes +LIB_OBJS := $(filter-out $(OUTPUT)util/symbol-elf.o,$(LIB_OBJS)) +LIB_OBJS := $(filter-out $(OUTPUT)util/dwarf-aux.o,$(LIB_OBJS)) +LIB_OBJS := $(filter-out $(OUTPUT)util/probe-event.o,$(LIB_OBJS)) +LIB_OBJS := $(filter-out $(OUTPUT)util/probe-finder.o,$(LIB_OBJS)) + +BUILTIN_OBJS := $(filter-out $(OUTPUT)builtin-probe.o,$(BUILTIN_OBJS)) + +# Use minimal symbol handling +LIB_OBJS += $(OUTPUT)util/symbol-minimal.o + +else # NO_LIBELF +BASIC_CFLAGS += -DLIBELF_SUPPORT + +FLAGS_LIBELF=$(ALL_CFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) +ifeq ($(call try-cc,$(SOURCE_ELF_MMAP),$(FLAGS_LIBELF),-DLIBELF_MMAP),y) + BASIC_CFLAGS += -DLIBELF_MMAP +endif + +ifndef NO_DWARF +ifeq ($(origin PERF_HAVE_DWARF_REGS), undefined) + msg := $(warning DWARF register mappings have not been defined for architecture $(ARCH), DWARF support disabled); +else + BASIC_CFLAGS := -DDWARF_SUPPORT $(LIBDW_CFLAGS) $(BASIC_CFLAGS) + BASIC_LDFLAGS := $(LIBDW_LDFLAGS) $(BASIC_LDFLAGS) + EXTLIBS += -lelf -ldw + LIB_OBJS += $(OUTPUT)util/probe-finder.o + LIB_OBJS += $(OUTPUT)util/dwarf-aux.o +endif # PERF_HAVE_DWARF_REGS +endif # NO_DWARF +endif # NO_LIBELF + +ifndef NO_LIBUNWIND + BASIC_CFLAGS += -DLIBUNWIND_SUPPORT + EXTLIBS += $(LIBUNWIND_LIBS) + BASIC_CFLAGS := $(LIBUNWIND_CFLAGS) $(BASIC_CFLAGS) + BASIC_LDFLAGS := $(LIBUNWIND_LDFLAGS) $(BASIC_LDFLAGS) + LIB_OBJS += $(OUTPUT)util/unwind.o +endif + +ifndef NO_LIBAUDIT + FLAGS_LIBAUDIT = $(ALL_CFLAGS) $(ALL_LDFLAGS) -laudit + ifneq ($(call try-cc,$(SOURCE_LIBAUDIT),$(FLAGS_LIBAUDIT),libaudit),y) + msg := $(warning No libaudit.h found, disables 'trace' tool, please install audit-libs-devel or libaudit-dev); + else + BASIC_CFLAGS += -DLIBAUDIT_SUPPORT + BUILTIN_OBJS += $(OUTPUT)builtin-trace.o + EXTLIBS += -laudit + endif +endif + +ifndef NO_NEWT + FLAGS_NEWT=$(ALL_CFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) -lnewt + ifneq ($(call try-cc,$(SOURCE_NEWT),$(FLAGS_NEWT),libnewt),y) + msg := $(warning newt not found, disables TUI support. Please install newt-devel or libnewt-dev); + else + # Fedora has /usr/include/slang/slang.h, but ubuntu /usr/include/slang.h + BASIC_CFLAGS += -I/usr/include/slang + BASIC_CFLAGS += -DNEWT_SUPPORT + EXTLIBS += -lnewt -lslang + LIB_OBJS += $(OUTPUT)ui/browser.o + LIB_OBJS += $(OUTPUT)ui/browsers/annotate.o + LIB_OBJS += $(OUTPUT)ui/browsers/hists.o + LIB_OBJS += $(OUTPUT)ui/browsers/map.o + LIB_OBJS += $(OUTPUT)ui/browsers/scripts.o + LIB_OBJS += $(OUTPUT)ui/tui/setup.o + LIB_OBJS += $(OUTPUT)ui/tui/util.o + LIB_OBJS += $(OUTPUT)ui/tui/helpline.o + LIB_OBJS += $(OUTPUT)ui/tui/progress.o + LIB_H += ui/browser.h + LIB_H += ui/browsers/map.h + LIB_H += ui/keysyms.h + LIB_H += ui/libslang.h + endif +endif + +ifndef NO_GTK2 + FLAGS_GTK2=$(ALL_CFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) $(shell pkg-config --libs --cflags gtk+-2.0 2>/dev/null) + ifneq ($(call try-cc,$(SOURCE_GTK2),$(FLAGS_GTK2),gtk2),y) + msg := $(warning GTK2 not found, disables GTK2 support. Please install gtk2-devel or libgtk2.0-dev); + else + ifeq ($(call try-cc,$(SOURCE_GTK2_INFOBAR),$(FLAGS_GTK2),-DHAVE_GTK_INFO_BAR),y) + BASIC_CFLAGS += -DHAVE_GTK_INFO_BAR + endif + BASIC_CFLAGS += -DGTK2_SUPPORT + BASIC_CFLAGS += $(shell pkg-config --cflags gtk+-2.0 2>/dev/null) + EXTLIBS += $(shell pkg-config --libs gtk+-2.0 2>/dev/null) + LIB_OBJS += $(OUTPUT)ui/gtk/browser.o + LIB_OBJS += $(OUTPUT)ui/gtk/hists.o + LIB_OBJS += $(OUTPUT)ui/gtk/setup.o + LIB_OBJS += $(OUTPUT)ui/gtk/util.o + LIB_OBJS += $(OUTPUT)ui/gtk/helpline.o + LIB_OBJS += $(OUTPUT)ui/gtk/progress.o + LIB_OBJS += $(OUTPUT)ui/gtk/annotate.o + endif +endif + +ifdef NO_LIBPERL + BASIC_CFLAGS += -DNO_LIBPERL +else + PERL_EMBED_LDOPTS = $(shell perl -MExtUtils::Embed -e ldopts 2>/dev/null) + PERL_EMBED_LDFLAGS = $(call strip-libs,$(PERL_EMBED_LDOPTS)) + PERL_EMBED_LIBADD = $(call grep-libs,$(PERL_EMBED_LDOPTS)) + PERL_EMBED_CCOPTS = `perl -MExtUtils::Embed -e ccopts 2>/dev/null` + FLAGS_PERL_EMBED=$(PERL_EMBED_CCOPTS) $(PERL_EMBED_LDOPTS) + + ifneq ($(call try-cc,$(SOURCE_PERL_EMBED),$(FLAGS_PERL_EMBED),perl),y) + BASIC_CFLAGS += -DNO_LIBPERL + else + ALL_LDFLAGS += $(PERL_EMBED_LDFLAGS) + EXTLIBS += $(PERL_EMBED_LIBADD) + LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-perl.o + LIB_OBJS += $(OUTPUT)scripts/perl/Perf-Trace-Util/Context.o + endif +endif + +disable-python = $(eval $(disable-python_code)) +define disable-python_code + BASIC_CFLAGS += -DNO_LIBPYTHON + $(if $(1),$(warning No $(1) was found)) + $(warning Python support will not be built) +endef + +override PYTHON := \ + $(call get-executable-or-default,PYTHON,python) + +ifndef PYTHON + $(call disable-python,python interpreter) +else + + PYTHON_WORD := $(call shell-wordify,$(PYTHON)) + + ifdef NO_LIBPYTHON + $(call disable-python) + else + + override PYTHON_CONFIG := \ + $(call get-executable-or-default,PYTHON_CONFIG,$(PYTHON)-config) + + ifndef PYTHON_CONFIG + $(call disable-python,python-config tool) + else + + PYTHON_CONFIG_SQ := $(call shell-sq,$(PYTHON_CONFIG)) + + PYTHON_EMBED_LDOPTS := $(shell $(PYTHON_CONFIG_SQ) --ldflags 2>/dev/null) + PYTHON_EMBED_LDFLAGS := $(call strip-libs,$(PYTHON_EMBED_LDOPTS)) + PYTHON_EMBED_LIBADD := $(call grep-libs,$(PYTHON_EMBED_LDOPTS)) + PYTHON_EMBED_CCOPTS := $(shell $(PYTHON_CONFIG_SQ) --cflags 2>/dev/null) + FLAGS_PYTHON_EMBED := $(PYTHON_EMBED_CCOPTS) $(PYTHON_EMBED_LDOPTS) + + ifneq ($(call try-cc,$(SOURCE_PYTHON_EMBED),$(FLAGS_PYTHON_EMBED),python),y) + $(call disable-python,Python.h (for Python 2.x)) + else + + ifneq ($(call try-cc,$(SOURCE_PYTHON_VERSION),$(FLAGS_PYTHON_EMBED),python version),y) + $(warning Python 3 is not yet supported; please set) + $(warning PYTHON and/or PYTHON_CONFIG appropriately.) + $(warning If you also have Python 2 installed, then) + $(warning try something like:) + $(warning $(and ,)) + $(warning $(and ,) make PYTHON=python2) + $(warning $(and ,)) + $(warning Otherwise, disable Python support entirely:) + $(warning $(and ,)) + $(warning $(and ,) make NO_LIBPYTHON=1) + $(warning $(and ,)) + $(error $(and ,)) + else + ALL_LDFLAGS += $(PYTHON_EMBED_LDFLAGS) + EXTLIBS += $(PYTHON_EMBED_LIBADD) + LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-python.o + LIB_OBJS += $(OUTPUT)scripts/python/Perf-Trace-Util/Context.o + LANG_BINDINGS += $(OUTPUT)python/perf.so + endif + + endif + endif + endif +endif + +ifdef NO_DEMANGLE + BASIC_CFLAGS += -DNO_DEMANGLE +else + ifdef HAVE_CPLUS_DEMANGLE + EXTLIBS += -liberty + BASIC_CFLAGS += -DHAVE_CPLUS_DEMANGLE + else + FLAGS_BFD=$(ALL_CFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) -DPACKAGE='perf' -lbfd + has_bfd := $(call try-cc,$(SOURCE_BFD),$(FLAGS_BFD),libbfd) + ifeq ($(has_bfd),y) + EXTLIBS += -lbfd + else + FLAGS_BFD_IBERTY=$(FLAGS_BFD) -liberty + has_bfd_iberty := $(call try-cc,$(SOURCE_BFD),$(FLAGS_BFD_IBERTY),liberty) + ifeq ($(has_bfd_iberty),y) + EXTLIBS += -lbfd -liberty + else + FLAGS_BFD_IBERTY_Z=$(FLAGS_BFD_IBERTY) -lz + has_bfd_iberty_z := $(call try-cc,$(SOURCE_BFD),$(FLAGS_BFD_IBERTY_Z),libz) + ifeq ($(has_bfd_iberty_z),y) + EXTLIBS += -lbfd -liberty -lz + else + FLAGS_CPLUS_DEMANGLE=$(ALL_CFLAGS) $(ALL_LDFLAGS) $(EXTLIBS) -liberty + has_cplus_demangle := $(call try-cc,$(SOURCE_CPLUS_DEMANGLE),$(FLAGS_CPLUS_DEMANGLE),demangle) + ifeq ($(has_cplus_demangle),y) + EXTLIBS += -liberty + BASIC_CFLAGS += -DHAVE_CPLUS_DEMANGLE + else + msg := $(warning No bfd.h/libbfd found, install binutils-dev[el]/zlib-static to gain symbol demangling) + BASIC_CFLAGS += -DNO_DEMANGLE + endif + endif + endif + endif + endif +endif + +ifeq ($(NO_PERF_REGS),0) + ifeq ($(ARCH),x86) + LIB_H += arch/x86/include/perf_regs.h + endif + BASIC_CFLAGS += -DHAVE_PERF_REGS +endif + +ifndef NO_STRLCPY + ifeq ($(call try-cc,$(SOURCE_STRLCPY),,-DHAVE_STRLCPY),y) + BASIC_CFLAGS += -DHAVE_STRLCPY + endif +endif + +ifndef NO_ON_EXIT + ifeq ($(call try-cc,$(SOURCE_ON_EXIT),,-DHAVE_ON_EXIT),y) + BASIC_CFLAGS += -DHAVE_ON_EXIT + endif +endif + +ifndef NO_BACKTRACE + ifeq ($(call try-cc,$(SOURCE_BACKTRACE),,-DBACKTRACE_SUPPORT),y) + BASIC_CFLAGS += -DBACKTRACE_SUPPORT + endif +endif + +ifndef NO_LIBNUMA + FLAGS_LIBNUMA = $(ALL_CFLAGS) $(ALL_LDFLAGS) -lnuma + ifneq ($(call try-cc,$(SOURCE_LIBNUMA),$(FLAGS_LIBNUMA),libnuma),y) + msg := $(warning No numa.h found, disables 'perf bench numa mem' benchmark, please install numa-libs-devel or libnuma-dev); + else + BASIC_CFLAGS += -DLIBNUMA_SUPPORT + BUILTIN_OBJS += $(OUTPUT)bench/numa.o + EXTLIBS += -lnuma + endif +endif + +ifdef ASCIIDOC8 + export ASCIIDOC8 +endif + +endif # MAKECMDGOALS != tags +endif # MAKECMDGOALS != clean + +# Shell quote (do not use $(call) to accommodate ancient setups); + +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)) +sysconfdir_SQ = $(subst ','\'',$(sysconfdir)) + +SHELL_PATH_SQ = $(subst ','\'',$(SHELL_PATH)) + +LIBS = -Wl,--whole-archive $(PERFLIBS) -Wl,--no-whole-archive -Wl,--start-group $(EXTLIBS) -Wl,--end-group + +ALL_CFLAGS += $(ANDROID_CFLAGS) +ALL_CFLAGS += $(BASIC_CFLAGS) +ALL_CFLAGS += $(ARCH_CFLAGS) +ALL_LDFLAGS += $(BASIC_LDFLAGS) + +export INSTALL SHELL_PATH + + +### Build rules + +SHELL = $(SHELL_PATH) + +all: shell_compatibility_test $(ALL_PROGRAMS) $(LANG_BINDINGS) $(OTHER_PROGRAMS) + +please_set_SHELL_PATH_to_a_more_modern_shell: + @$$(:) + +shell_compatibility_test: please_set_SHELL_PATH_to_a_more_modern_shell + +strip: $(PROGRAMS) $(OUTPUT)perf + $(STRIP) $(STRIP_OPTS) $(PROGRAMS) $(OUTPUT)perf + +$(OUTPUT)perf.o: perf.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -include $(OUTPUT)PERF-VERSION-FILE \ + '-DPERF_HTML_PATH="$(htmldir_SQ)"' \ + $(ALL_CFLAGS) -c $(filter %.c,$^) -o $@ + +$(OUTPUT)perf: $(OUTPUT)perf.o $(BUILTIN_OBJS) $(PERFLIBS) + $(QUIET_LINK)$(CC) $(ALL_CFLAGS) $(ALL_LDFLAGS) $(OUTPUT)perf.o \ + $(BUILTIN_OBJS) $(LIBS) -o $@ + +$(OUTPUT)builtin-help.o: builtin-help.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \ + '-DPERF_HTML_PATH="$(htmldir_SQ)"' \ + '-DPERF_MAN_PATH="$(mandir_SQ)"' \ + '-DPERF_INFO_PATH="$(infodir_SQ)"' $< + +$(OUTPUT)builtin-timechart.o: builtin-timechart.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \ + '-DPERF_HTML_PATH="$(htmldir_SQ)"' \ + '-DPERF_MAN_PATH="$(mandir_SQ)"' \ + '-DPERF_INFO_PATH="$(infodir_SQ)"' $< + +$(OUTPUT)common-cmds.h: util/generate-cmdlist.sh command-list.txt + +$(OUTPUT)common-cmds.h: $(wildcard Documentation/perf-*.txt) + $(QUIET_GEN). util/generate-cmdlist.sh > $@+ && mv $@+ $@ + +$(SCRIPTS) : % : %.sh + $(QUIET_GEN)$(INSTALL) '$@.sh' '$(OUTPUT)$@' + +# These can record PERF_VERSION +$(OUTPUT)perf.o perf.spec \ + $(SCRIPTS) \ + : $(OUTPUT)PERF-VERSION-FILE + +.SUFFIXES: +.SUFFIXES: .o .c .S .s + +# These two need to be here so that when O= is not used they take precedence +# over the general rule for .o + +$(OUTPUT)util/%-flex.o: $(OUTPUT)util/%-flex.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c -Iutil/ $(ALL_CFLAGS) -w $< + +$(OUTPUT)util/%-bison.o: $(OUTPUT)util/%-bison.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c -Iutil/ $(ALL_CFLAGS) -DYYENABLE_NLS=0 -DYYLTYPE_IS_TRIVIAL=0 -w $< + +$(OUTPUT)%.o: %.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $< +$(OUTPUT)%.i: %.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -E $(ALL_CFLAGS) $< +$(OUTPUT)%.s: %.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -S $(ALL_CFLAGS) $< +$(OUTPUT)%.o: %.S + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $< +$(OUTPUT)%.s: %.S + $(QUIET_CC)$(CC) -o $@ -E $(ALL_CFLAGS) $< + +$(OUTPUT)util/exec_cmd.o: util/exec_cmd.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \ + '-DPERF_EXEC_PATH="$(perfexecdir_SQ)"' \ + '-DPREFIX="$(prefix_SQ)"' \ + $< + +$(OUTPUT)tests/attr.o: tests/attr.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \ + '-DBINDIR="$(bindir_SQ)"' -DPYTHON='"$(PYTHON_WORD)"' \ + $< + +$(OUTPUT)tests/python-use.o: tests/python-use.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \ + -DPYTHONPATH='"$(OUTPUT)python"' \ + -DPYTHON='"$(PYTHON_WORD)"' \ + $< + +$(OUTPUT)util/config.o: util/config.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< + +$(OUTPUT)ui/browser.o: ui/browser.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DENABLE_SLFUTURE_CONST $< + +$(OUTPUT)ui/browsers/annotate.o: ui/browsers/annotate.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DENABLE_SLFUTURE_CONST $< + +$(OUTPUT)ui/browsers/hists.o: ui/browsers/hists.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DENABLE_SLFUTURE_CONST $< + +$(OUTPUT)ui/browsers/map.o: ui/browsers/map.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DENABLE_SLFUTURE_CONST $< + +$(OUTPUT)ui/browsers/scripts.o: ui/browsers/scripts.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DENABLE_SLFUTURE_CONST $< + +$(OUTPUT)util/rbtree.o: ../../lib/rbtree.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -Wno-unused-parameter -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $< + +$(OUTPUT)util/parse-events.o: util/parse-events.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -Wno-redundant-decls $< + +$(OUTPUT)util/scripting-engines/trace-event-perl.o: util/scripting-engines/trace-event-perl.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $< + +$(OUTPUT)scripts/perl/Perf-Trace-Util/Context.o: scripts/perl/Perf-Trace-Util/Context.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $< + +$(OUTPUT)util/scripting-engines/trace-event-python.o: util/scripting-engines/trace-event-python.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $< + +$(OUTPUT)scripts/python/Perf-Trace-Util/Context.o: scripts/python/Perf-Trace-Util/Context.c $(OUTPUT)PERF-CFLAGS + $(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $< + +$(OUTPUT)perf-%: %.o $(PERFLIBS) + $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $(filter %.o,$^) $(LIBS) + +$(LIB_OBJS) $(BUILTIN_OBJS): $(LIB_H) +$(patsubst perf-%,%.o,$(PROGRAMS)): $(LIB_H) $(wildcard */*.h) + +# we compile into subdirectories. if the target directory is not the source directory, they might not exists. So +# we depend the various files onto their directories. +DIRECTORY_DEPS = $(LIB_OBJS) $(BUILTIN_OBJS) $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)common-cmds.h +$(DIRECTORY_DEPS): | $(sort $(dir $(DIRECTORY_DEPS))) +# In the second step, we make a rule to actually create these directories +$(sort $(dir $(DIRECTORY_DEPS))): + $(QUIET_MKDIR)$(MKDIR) -p $@ 2>/dev/null + +$(LIB_FILE): $(LIB_OBJS) + $(QUIET_AR)$(RM) $@ && $(AR) rcs $@ $(LIB_OBJS) + +# libtraceevent.a +$(LIBTRACEEVENT): + $(QUIET_SUBDIR0)$(TRACE_EVENT_DIR) $(QUIET_SUBDIR1) ANDROID_CFLAGS="$(ANDROID_CFLAGS)" O=$(OUTPUT) libtraceevent.a + +$(LIBTRACEEVENT)-clean: + $(QUIET_SUBDIR0)$(TRACE_EVENT_DIR) $(QUIET_SUBDIR1) O=$(OUTPUT) clean + +help: + @echo 'Perf make targets:' + @echo ' doc - make *all* documentation (see below)' + @echo ' man - make manpage documentation (access with man <foo>)' + @echo ' html - make html documentation' + @echo ' info - make GNU info documentation (access with info <foo>)' + @echo ' pdf - make pdf documentation' + @echo ' TAGS - use etags to make tag information for source browsing' + @echo ' tags - use ctags to make tag information for source browsing' + @echo ' cscope - use cscope to make interactive browsing database' + @echo '' + @echo 'Perf install targets:' + @echo ' NOTE: documentation build requires asciidoc, xmlto packages to be installed' + @echo ' HINT: use "make prefix=<path> <install target>" to install to a particular' + @echo ' path like make prefix=/usr/local install install-doc' + @echo ' install - install compiled binaries' + @echo ' install-doc - install *all* documentation' + @echo ' install-man - install manpage documentation' + @echo ' install-html - install html documentation' + @echo ' install-info - install GNU info documentation' + @echo ' install-pdf - install pdf documentation' + @echo '' + @echo ' quick-install-doc - alias for quick-install-man' + @echo ' quick-install-man - install the documentation quickly' + @echo ' quick-install-html - install the html documentation quickly' + @echo '' + @echo 'Perf maintainer targets:' + @echo ' clean - clean all binary objects and build output' + + +DOC_TARGETS := doc man html info pdf + +INSTALL_DOC_TARGETS := $(patsubst %,install-%,$(DOC_TARGETS)) try-install-man +INSTALL_DOC_TARGETS += quick-install-doc quick-install-man quick-install-html + +# 'make doc' should call 'make -C Documentation all' +$(DOC_TARGETS): + $(QUIET_SUBDIR0)Documentation $(QUIET_SUBDIR1) $(@:doc=all) + +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) + +$(OUTPUT)PERF-CFLAGS: .FORCE-PERF-CFLAGS + @FLAGS='$(TRACK_CFLAGS)'; \ + if test x"$$FLAGS" != x"`cat $(OUTPUT)PERF-CFLAGS 2>/dev/null`" ; then \ + echo 1>&2 " * new build flags or prefix"; \ + echo "$$FLAGS" >$(OUTPUT)PERF-CFLAGS; \ + fi + +### Testing rules + +# GNU make supports exporting all variables by "export" without parameters. +# However, the environment gets quite big, and some programs have problems +# with that. + +check: $(OUTPUT)common-cmds.h + if sparse; \ + then \ + for i in *.c */*.c; \ + do \ + sparse $(ALL_CFLAGS) $(SPARSE_FLAGS) $$i || exit; \ + done; \ + else \ + exit 1; \ + fi + +### Installation rules + +ifneq ($(filter /%,$(firstword $(perfexecdir))),) +perfexec_instdir = $(perfexecdir) +else +perfexec_instdir = $(prefix)/$(perfexecdir) +endif +perfexec_instdir_SQ = $(subst ','\'',$(perfexec_instdir)) + +install-bin: all + $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(bindir_SQ)' + $(INSTALL) $(OUTPUT)perf '$(DESTDIR_SQ)$(bindir_SQ)' + $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace' + $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin' + $(INSTALL) $(OUTPUT)perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)' + $(INSTALL) scripts/perl/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace' + $(INSTALL) scripts/perl/*.pl -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl' + $(INSTALL) scripts/perl/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin' + $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/Perf-Trace-Util/lib/Perf/Trace' + $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/bin' + $(INSTALL) scripts/python/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/Perf-Trace-Util/lib/Perf/Trace' + $(INSTALL) scripts/python/*.py -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python' + $(INSTALL) scripts/python/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/bin' + $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(sysconfdir_SQ)/bash_completion.d' + $(INSTALL) bash_completion '$(DESTDIR_SQ)$(sysconfdir_SQ)/bash_completion.d/perf' + $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests' + $(INSTALL) tests/attr.py '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests' + $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests/attr' + $(INSTALL) tests/attr/* '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests/attr' + +install: install-bin try-install-man + +install-python_ext: + $(PYTHON_WORD) util/setup.py --quiet install --root='/$(DESTDIR_SQ)' + +# 'make install-doc' should call 'make -C Documentation install' +$(INSTALL_DOC_TARGETS): + $(QUIET_SUBDIR0)Documentation $(QUIET_SUBDIR1) $(@:-doc=) + +### Cleaning rules + +clean: $(LIBTRACEEVENT)-clean + $(RM) $(LIB_OBJS) $(BUILTIN_OBJS) $(LIB_FILE) $(OUTPUT)perf-archive $(OUTPUT)perf.o $(LANG_BINDINGS) + $(RM) $(ALL_PROGRAMS) perf + $(RM) *.spec *.pyc *.pyo */*.pyc */*.pyo $(OUTPUT)common-cmds.h TAGS tags cscope* + $(QUIET_SUBDIR0)Documentation $(QUIET_SUBDIR1) clean + $(RM) $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)PERF-CFLAGS + $(RM) $(OUTPUT)util/*-bison* + $(RM) $(OUTPUT)util/*-flex* + $(python-clean) + +.PHONY: all install clean strip $(LIBTRACEEVENT) +.PHONY: shell_compatibility_test please_set_SHELL_PATH_to_a_more_modern_shell +.PHONY: .FORCE-PERF-VERSION-FILE TAGS tags cscope .FORCE-PERF-CFLAGS diff --git a/tools/perf/arch/arm/Makefile b/tools/perf/arch/arm/Makefile new file mode 100644 index 00000000..15130b50 --- /dev/null +++ b/tools/perf/arch/arm/Makefile @@ -0,0 +1,4 @@ +ifndef NO_DWARF +PERF_HAVE_DWARF_REGS := 1 +LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o +endif diff --git a/tools/perf/arch/arm/util/dwarf-regs.c b/tools/perf/arch/arm/util/dwarf-regs.c new file mode 100644 index 00000000..e8d5c551 --- /dev/null +++ b/tools/perf/arch/arm/util/dwarf-regs.c @@ -0,0 +1,67 @@ +/* + * Mapping of DWARF debug register numbers into register names. + * + * Copyright (C) 2010 Will Deacon, ARM Ltd. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <stdlib.h> +#ifndef __UCLIBC__ +#include <libio.h> +#endif +#include <dwarf-regs.h> + +struct pt_regs_dwarfnum { + const char *name; + unsigned int dwarfnum; +}; + +#define STR(s) #s +#define REG_DWARFNUM_NAME(r, num) {.name = r, .dwarfnum = num} +#define GPR_DWARFNUM_NAME(num) \ + {.name = STR(%r##num), .dwarfnum = num} +#define REG_DWARFNUM_END {.name = NULL, .dwarfnum = 0} + +/* + * Reference: + * http://infocenter.arm.com/help/topic/com.arm.doc.ihi0040a/IHI0040A_aadwarf.pdf + */ +static const struct pt_regs_dwarfnum regdwarfnum_table[] = { + GPR_DWARFNUM_NAME(0), + GPR_DWARFNUM_NAME(1), + GPR_DWARFNUM_NAME(2), + GPR_DWARFNUM_NAME(3), + GPR_DWARFNUM_NAME(4), + GPR_DWARFNUM_NAME(5), + GPR_DWARFNUM_NAME(6), + GPR_DWARFNUM_NAME(7), + GPR_DWARFNUM_NAME(8), + GPR_DWARFNUM_NAME(9), + GPR_DWARFNUM_NAME(10), + REG_DWARFNUM_NAME("%fp", 11), + REG_DWARFNUM_NAME("%ip", 12), + REG_DWARFNUM_NAME("%sp", 13), + REG_DWARFNUM_NAME("%lr", 14), + REG_DWARFNUM_NAME("%pc", 15), + REG_DWARFNUM_END, +}; + +/** + * get_arch_regstr() - lookup register name from it's DWARF register number + * @n: the DWARF register number + * + * get_arch_regstr() returns the name of the register in struct + * regdwarfnum_table from it's DWARF register number. If the register is not + * found in the table, this returns NULL; + */ +const char *get_arch_regstr(unsigned int n) +{ + const struct pt_regs_dwarfnum *roff; + for (roff = regdwarfnum_table; roff->name != NULL; roff++) + if (roff->dwarfnum == n) + return roff->name; + return NULL; +} diff --git a/tools/perf/arch/common.c b/tools/perf/arch/common.c new file mode 100644 index 00000000..aacef07e --- /dev/null +++ b/tools/perf/arch/common.c @@ -0,0 +1,212 @@ +#include <stdio.h> +#include <sys/utsname.h> +#include "common.h" +#include "../util/debug.h" + +const char *const arm_triplets[] = { + "arm-eabi-", + "arm-linux-androideabi-", + "arm-unknown-linux-", + "arm-unknown-linux-gnu-", + "arm-unknown-linux-gnueabi-", + NULL +}; + +const char *const powerpc_triplets[] = { + "powerpc-unknown-linux-gnu-", + "powerpc64-unknown-linux-gnu-", + NULL +}; + +const char *const s390_triplets[] = { + "s390-ibm-linux-", + NULL +}; + +const char *const sh_triplets[] = { + "sh-unknown-linux-gnu-", + "sh64-unknown-linux-gnu-", + NULL +}; + +const char *const sparc_triplets[] = { + "sparc-unknown-linux-gnu-", + "sparc64-unknown-linux-gnu-", + NULL +}; + +const char *const x86_triplets[] = { + "x86_64-pc-linux-gnu-", + "x86_64-unknown-linux-gnu-", + "i686-pc-linux-gnu-", + "i586-pc-linux-gnu-", + "i486-pc-linux-gnu-", + "i386-pc-linux-gnu-", + "i686-linux-android-", + "i686-android-linux-", + NULL +}; + +const char *const mips_triplets[] = { + "mips-unknown-linux-gnu-", + "mipsel-linux-android-", + NULL +}; + +static bool lookup_path(char *name) +{ + bool found = false; + char *path, *tmp; + char buf[PATH_MAX]; + char *env = getenv("PATH"); + + if (!env) + return false; + + env = strdup(env); + if (!env) + return false; + + path = strtok_r(env, ":", &tmp); + while (path) { + scnprintf(buf, sizeof(buf), "%s/%s", path, name); + if (access(buf, F_OK) == 0) { + found = true; + break; + } + path = strtok_r(NULL, ":", &tmp); + } + free(env); + return found; +} + +static int lookup_triplets(const char *const *triplets, const char *name) +{ + int i; + char buf[PATH_MAX]; + + for (i = 0; triplets[i] != NULL; i++) { + scnprintf(buf, sizeof(buf), "%s%s", triplets[i], name); + if (lookup_path(buf)) + return i; + } + return -1; +} + +/* + * Return architecture name in a normalized form. + * The conversion logic comes from the Makefile. + */ +static const char *normalize_arch(char *arch) +{ + if (!strcmp(arch, "x86_64")) + return "x86"; + if (arch[0] == 'i' && arch[2] == '8' && arch[3] == '6') + return "x86"; + if (!strcmp(arch, "sun4u") || !strncmp(arch, "sparc", 5)) + return "sparc"; + if (!strncmp(arch, "arm", 3) || !strcmp(arch, "sa110")) + return "arm"; + if (!strncmp(arch, "s390", 4)) + return "s390"; + if (!strncmp(arch, "parisc", 6)) + return "parisc"; + if (!strncmp(arch, "powerpc", 7) || !strncmp(arch, "ppc", 3)) + return "powerpc"; + if (!strncmp(arch, "mips", 4)) + return "mips"; + if (!strncmp(arch, "sh", 2) && isdigit(arch[2])) + return "sh"; + + return arch; +} + +static int perf_session_env__lookup_binutils_path(struct perf_session_env *env, + const char *name, + const char **path) +{ + int idx; + const char *arch, *cross_env; + struct utsname uts; + const char *const *path_list; + char *buf = NULL; + + arch = normalize_arch(env->arch); + + if (uname(&uts) < 0) + goto out; + + /* + * We don't need to try to find objdump path for native system. + * Just use default binutils path (e.g.: "objdump"). + */ + if (!strcmp(normalize_arch(uts.machine), arch)) + goto out; + + cross_env = getenv("CROSS_COMPILE"); + if (cross_env) { + if (asprintf(&buf, "%s%s", cross_env, name) < 0) + goto out_error; + if (buf[0] == '/') { + if (access(buf, F_OK) == 0) + goto out; + goto out_error; + } + if (lookup_path(buf)) + goto out; + free(buf); + buf = NULL; + } + + if (!strcmp(arch, "arm")) + path_list = arm_triplets; + else if (!strcmp(arch, "powerpc")) + path_list = powerpc_triplets; + else if (!strcmp(arch, "sh")) + path_list = sh_triplets; + else if (!strcmp(arch, "s390")) + path_list = s390_triplets; + else if (!strcmp(arch, "sparc")) + path_list = sparc_triplets; + else if (!strcmp(arch, "x86")) + path_list = x86_triplets; + else if (!strcmp(arch, "mips")) + path_list = mips_triplets; + else { + ui__error("binutils for %s not supported.\n", arch); + goto out_error; + } + + idx = lookup_triplets(path_list, name); + if (idx < 0) { + ui__error("Please install %s for %s.\n" + "You can add it to PATH, set CROSS_COMPILE or " + "override the default using --%s.\n", + name, arch, name); + goto out_error; + } + + if (asprintf(&buf, "%s%s", path_list[idx], name) < 0) + goto out_error; + +out: + *path = buf; + return 0; +out_error: + free(buf); + *path = NULL; + return -1; +} + +int perf_session_env__lookup_objdump(struct perf_session_env *env) +{ + /* + * For live mode, env->arch will be NULL and we can use + * the native objdump tool. + */ + if (env->arch == NULL) + return 0; + + return perf_session_env__lookup_binutils_path(env, "objdump", + &objdump_path); +} diff --git a/tools/perf/arch/common.h b/tools/perf/arch/common.h new file mode 100644 index 00000000..ede246ed --- /dev/null +++ b/tools/perf/arch/common.h @@ -0,0 +1,10 @@ +#ifndef ARCH_PERF_COMMON_H +#define ARCH_PERF_COMMON_H + +#include "../util/session.h" + +extern const char *objdump_path; + +int perf_session_env__lookup_objdump(struct perf_session_env *env); + +#endif /* ARCH_PERF_COMMON_H */ diff --git a/tools/perf/arch/powerpc/Makefile b/tools/perf/arch/powerpc/Makefile new file mode 100644 index 00000000..744e6297 --- /dev/null +++ b/tools/perf/arch/powerpc/Makefile @@ -0,0 +1,5 @@ +ifndef NO_DWARF +PERF_HAVE_DWARF_REGS := 1 +LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o +endif +LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/header.o diff --git a/tools/perf/arch/powerpc/util/dwarf-regs.c b/tools/perf/arch/powerpc/util/dwarf-regs.c new file mode 100644 index 00000000..7cdd61d0 --- /dev/null +++ b/tools/perf/arch/powerpc/util/dwarf-regs.c @@ -0,0 +1,91 @@ +/* + * Mapping of DWARF debug register numbers into register names. + * + * Copyright (C) 2010 Ian Munsie, IBM Corporation. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include <stdlib.h> +#ifndef __UCLIBC__ +#include <libio.h> +#endif +#include <dwarf-regs.h> + + +struct pt_regs_dwarfnum { + const char *name; + unsigned int dwarfnum; +}; + +#define STR(s) #s +#define REG_DWARFNUM_NAME(r, num) {.name = r, .dwarfnum = num} +#define GPR_DWARFNUM_NAME(num) \ + {.name = STR(%gpr##num), .dwarfnum = num} +#define REG_DWARFNUM_END {.name = NULL, .dwarfnum = 0} + +/* + * Reference: + * http://refspecs.linuxfoundation.org/ELF/ppc64/PPC-elf64abi-1.9.html + */ +static const struct pt_regs_dwarfnum regdwarfnum_table[] = { + GPR_DWARFNUM_NAME(0), + GPR_DWARFNUM_NAME(1), + GPR_DWARFNUM_NAME(2), + GPR_DWARFNUM_NAME(3), + GPR_DWARFNUM_NAME(4), + GPR_DWARFNUM_NAME(5), + GPR_DWARFNUM_NAME(6), + GPR_DWARFNUM_NAME(7), + GPR_DWARFNUM_NAME(8), + GPR_DWARFNUM_NAME(9), + GPR_DWARFNUM_NAME(10), + GPR_DWARFNUM_NAME(11), + GPR_DWARFNUM_NAME(12), + GPR_DWARFNUM_NAME(13), + GPR_DWARFNUM_NAME(14), + GPR_DWARFNUM_NAME(15), + GPR_DWARFNUM_NAME(16), + GPR_DWARFNUM_NAME(17), + GPR_DWARFNUM_NAME(18), + GPR_DWARFNUM_NAME(19), + GPR_DWARFNUM_NAME(20), + GPR_DWARFNUM_NAME(21), + GPR_DWARFNUM_NAME(22), + GPR_DWARFNUM_NAME(23), + GPR_DWARFNUM_NAME(24), + GPR_DWARFNUM_NAME(25), + GPR_DWARFNUM_NAME(26), + GPR_DWARFNUM_NAME(27), + GPR_DWARFNUM_NAME(28), + GPR_DWARFNUM_NAME(29), + GPR_DWARFNUM_NAME(30), + GPR_DWARFNUM_NAME(31), + REG_DWARFNUM_NAME("%msr", 66), + REG_DWARFNUM_NAME("%ctr", 109), + REG_DWARFNUM_NAME("%link", 108), + REG_DWARFNUM_NAME("%xer", 101), + REG_DWARFNUM_NAME("%dar", 119), + REG_DWARFNUM_NAME("%dsisr", 118), + REG_DWARFNUM_END, +}; + +/** + * get_arch_regstr() - lookup register name from it's DWARF register number + * @n: the DWARF register number + * + * get_arch_regstr() returns the name of the register in struct + * regdwarfnum_table from it's DWARF register number. If the register is not + * found in the table, this returns NULL; + */ +const char *get_arch_regstr(unsigned int n) +{ + const struct pt_regs_dwarfnum *roff; + for (roff = regdwarfnum_table; roff->name != NULL; roff++) + if (roff->dwarfnum == n) + return roff->name; + return NULL; +} diff --git a/tools/perf/arch/powerpc/util/header.c b/tools/perf/arch/powerpc/util/header.c new file mode 100644 index 00000000..2f7073d1 --- /dev/null +++ b/tools/perf/arch/powerpc/util/header.c @@ -0,0 +1,36 @@ +#include <sys/types.h> +#include <unistd.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include "../../util/header.h" + +#define __stringify_1(x) #x +#define __stringify(x) __stringify_1(x) + +#define mfspr(rn) ({unsigned long rval; \ + asm volatile("mfspr %0," __stringify(rn) \ + : "=r" (rval)); rval; }) + +#define SPRN_PVR 0x11F /* Processor Version Register */ +#define PVR_VER(pvr) (((pvr) >> 16) & 0xFFFF) /* Version field */ +#define PVR_REV(pvr) (((pvr) >> 0) & 0xFFFF) /* Revison field */ + +int +get_cpuid(char *buffer, size_t sz) +{ + unsigned long pvr; + int nb; + + pvr = mfspr(SPRN_PVR); + + nb = scnprintf(buffer, sz, "%lu,%lu$", PVR_VER(pvr), PVR_REV(pvr)); + + /* look for end marker to ensure the entire data fit */ + if (strchr(buffer, '$')) { + buffer[nb-1] = '\0'; + return 0; + } + return -1; +} diff --git a/tools/perf/arch/s390/Makefile b/tools/perf/arch/s390/Makefile new file mode 100644 index 00000000..15130b50 --- /dev/null +++ b/tools/perf/arch/s390/Makefile @@ -0,0 +1,4 @@ +ifndef NO_DWARF +PERF_HAVE_DWARF_REGS := 1 +LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o +endif diff --git a/tools/perf/arch/s390/util/dwarf-regs.c b/tools/perf/arch/s390/util/dwarf-regs.c new file mode 100644 index 00000000..e19653e0 --- /dev/null +++ b/tools/perf/arch/s390/util/dwarf-regs.c @@ -0,0 +1,22 @@ +/* + * Mapping of DWARF debug register numbers into register names. + * + * Copyright IBM Corp. 2010 + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>, + * + */ + +#include <libio.h> +#include <dwarf-regs.h> + +#define NUM_GPRS 16 + +static const char *gpr_names[NUM_GPRS] = { + "%r0", "%r1", "%r2", "%r3", "%r4", "%r5", "%r6", "%r7", + "%r8", "%r9", "%r10", "%r11", "%r12", "%r13", "%r14", "%r15", +}; + +const char *get_arch_regstr(unsigned int n) +{ + return (n >= NUM_GPRS) ? NULL : gpr_names[n]; +} diff --git a/tools/perf/arch/sh/Makefile b/tools/perf/arch/sh/Makefile new file mode 100644 index 00000000..15130b50 --- /dev/null +++ b/tools/perf/arch/sh/Makefile @@ -0,0 +1,4 @@ +ifndef NO_DWARF +PERF_HAVE_DWARF_REGS := 1 +LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o +endif diff --git a/tools/perf/arch/sh/util/dwarf-regs.c b/tools/perf/arch/sh/util/dwarf-regs.c new file mode 100644 index 00000000..a11edb00 --- /dev/null +++ b/tools/perf/arch/sh/util/dwarf-regs.c @@ -0,0 +1,55 @@ +/* + * Mapping of DWARF debug register numbers into register names. + * + * Copyright (C) 2010 Matt Fleming <matt@console-pimps.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ + +#include <libio.h> +#include <dwarf-regs.h> + +/* + * Generic dwarf analysis helpers + */ + +#define SH_MAX_REGS 18 +const char *sh_regs_table[SH_MAX_REGS] = { + "r0", + "r1", + "r2", + "r3", + "r4", + "r5", + "r6", + "r7", + "r8", + "r9", + "r10", + "r11", + "r12", + "r13", + "r14", + "r15", + "pc", + "pr", +}; + +/* Return architecture dependent register string (for kprobe-tracer) */ +const char *get_arch_regstr(unsigned int n) +{ + return (n <= SH_MAX_REGS) ? sh_regs_table[n] : NULL; +} diff --git a/tools/perf/arch/sparc/Makefile b/tools/perf/arch/sparc/Makefile new file mode 100644 index 00000000..15130b50 --- /dev/null +++ b/tools/perf/arch/sparc/Makefile @@ -0,0 +1,4 @@ +ifndef NO_DWARF +PERF_HAVE_DWARF_REGS := 1 +LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o +endif diff --git a/tools/perf/arch/sparc/util/dwarf-regs.c b/tools/perf/arch/sparc/util/dwarf-regs.c new file mode 100644 index 00000000..0ab88483 --- /dev/null +++ b/tools/perf/arch/sparc/util/dwarf-regs.c @@ -0,0 +1,43 @@ +/* + * Mapping of DWARF debug register numbers into register names. + * + * Copyright (C) 2010 David S. Miller <davem@davemloft.net> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include <libio.h> +#include <dwarf-regs.h> + +#define SPARC_MAX_REGS 96 + +const char *sparc_regs_table[SPARC_MAX_REGS] = { + "%g0", "%g1", "%g2", "%g3", "%g4", "%g5", "%g6", "%g7", + "%o0", "%o1", "%o2", "%o3", "%o4", "%o5", "%sp", "%o7", + "%l0", "%l1", "%l2", "%l3", "%l4", "%l5", "%l6", "%l7", + "%i0", "%i1", "%i2", "%i3", "%i4", "%i5", "%fp", "%i7", + "%f0", "%f1", "%f2", "%f3", "%f4", "%f5", "%f6", "%f7", + "%f8", "%f9", "%f10", "%f11", "%f12", "%f13", "%f14", "%f15", + "%f16", "%f17", "%f18", "%f19", "%f20", "%f21", "%f22", "%f23", + "%f24", "%f25", "%f26", "%f27", "%f28", "%f29", "%f30", "%f31", + "%f32", "%f33", "%f34", "%f35", "%f36", "%f37", "%f38", "%f39", + "%f40", "%f41", "%f42", "%f43", "%f44", "%f45", "%f46", "%f47", + "%f48", "%f49", "%f50", "%f51", "%f52", "%f53", "%f54", "%f55", + "%f56", "%f57", "%f58", "%f59", "%f60", "%f61", "%f62", "%f63", +}; + +/** + * get_arch_regstr() - lookup register name from it's DWARF register number + * @n: the DWARF register number + * + * get_arch_regstr() returns the name of the register in struct + * regdwarfnum_table from it's DWARF register number. If the register is not + * found in the table, this returns NULL; + */ +const char *get_arch_regstr(unsigned int n) +{ + return (n <= SPARC_MAX_REGS) ? sparc_regs_table[n] : NULL; +} diff --git a/tools/perf/arch/x86/Makefile b/tools/perf/arch/x86/Makefile new file mode 100644 index 00000000..815841c0 --- /dev/null +++ b/tools/perf/arch/x86/Makefile @@ -0,0 +1,8 @@ +ifndef NO_DWARF +PERF_HAVE_DWARF_REGS := 1 +LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o +endif +ifndef NO_LIBUNWIND +LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/unwind.o +endif +LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/header.o diff --git a/tools/perf/arch/x86/include/perf_regs.h b/tools/perf/arch/x86/include/perf_regs.h new file mode 100644 index 00000000..7fcdcdbe --- /dev/null +++ b/tools/perf/arch/x86/include/perf_regs.h @@ -0,0 +1,80 @@ +#ifndef ARCH_PERF_REGS_H +#define ARCH_PERF_REGS_H + +#include <stdlib.h> +#include "../../util/types.h" +#include <asm/perf_regs.h> + +#ifndef ARCH_X86_64 +#define PERF_REGS_MASK ((1ULL << PERF_REG_X86_32_MAX) - 1) +#else +#define REG_NOSUPPORT ((1ULL << PERF_REG_X86_DS) | \ + (1ULL << PERF_REG_X86_ES) | \ + (1ULL << PERF_REG_X86_FS) | \ + (1ULL << PERF_REG_X86_GS)) +#define PERF_REGS_MASK (((1ULL << PERF_REG_X86_64_MAX) - 1) & ~REG_NOSUPPORT) +#endif +#define PERF_REG_IP PERF_REG_X86_IP +#define PERF_REG_SP PERF_REG_X86_SP + +static inline const char *perf_reg_name(int id) +{ + switch (id) { + case PERF_REG_X86_AX: + return "AX"; + case PERF_REG_X86_BX: + return "BX"; + case PERF_REG_X86_CX: + return "CX"; + case PERF_REG_X86_DX: + return "DX"; + case PERF_REG_X86_SI: + return "SI"; + case PERF_REG_X86_DI: + return "DI"; + case PERF_REG_X86_BP: + return "BP"; + case PERF_REG_X86_SP: + return "SP"; + case PERF_REG_X86_IP: + return "IP"; + case PERF_REG_X86_FLAGS: + return "FLAGS"; + case PERF_REG_X86_CS: + return "CS"; + case PERF_REG_X86_SS: + return "SS"; + case PERF_REG_X86_DS: + return "DS"; + case PERF_REG_X86_ES: + return "ES"; + case PERF_REG_X86_FS: + return "FS"; + case PERF_REG_X86_GS: + return "GS"; +#ifdef ARCH_X86_64 + case PERF_REG_X86_R8: + return "R8"; + case PERF_REG_X86_R9: + return "R9"; + case PERF_REG_X86_R10: + return "R10"; + case PERF_REG_X86_R11: + return "R11"; + case PERF_REG_X86_R12: + return "R12"; + case PERF_REG_X86_R13: + return "R13"; + case PERF_REG_X86_R14: + return "R14"; + case PERF_REG_X86_R15: + return "R15"; +#endif /* ARCH_X86_64 */ + default: + return NULL; + } + + return NULL; +} + +#endif /* ARCH_PERF_REGS_H */ diff --git a/tools/perf/arch/x86/util/dwarf-regs.c b/tools/perf/arch/x86/util/dwarf-regs.c new file mode 100644 index 00000000..a794d308 --- /dev/null +++ b/tools/perf/arch/x86/util/dwarf-regs.c @@ -0,0 +1,75 @@ +/* + * dwarf-regs.c : Mapping of DWARF debug register numbers into register names. + * Extracted from probe-finder.c + * + * Written by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ + +#include <libio.h> +#include <dwarf-regs.h> + +/* + * Generic dwarf analysis helpers + */ + +#define X86_32_MAX_REGS 8 +const char *x86_32_regs_table[X86_32_MAX_REGS] = { + "%ax", + "%cx", + "%dx", + "%bx", + "$stack", /* Stack address instead of %sp */ + "%bp", + "%si", + "%di", +}; + +#define X86_64_MAX_REGS 16 +const char *x86_64_regs_table[X86_64_MAX_REGS] = { + "%ax", + "%dx", + "%cx", + "%bx", + "%si", + "%di", + "%bp", + "%sp", + "%r8", + "%r9", + "%r10", + "%r11", + "%r12", + "%r13", + "%r14", + "%r15", +}; + +/* TODO: switching by dwarf address size */ +#ifdef __x86_64__ +#define ARCH_MAX_REGS X86_64_MAX_REGS +#define arch_regs_table x86_64_regs_table +#else +#define ARCH_MAX_REGS X86_32_MAX_REGS +#define arch_regs_table x86_32_regs_table +#endif + +/* Return architecture dependent register string (for kprobe-tracer) */ +const char *get_arch_regstr(unsigned int n) +{ + return (n <= ARCH_MAX_REGS) ? arch_regs_table[n] : NULL; +} diff --git a/tools/perf/arch/x86/util/header.c b/tools/perf/arch/x86/util/header.c new file mode 100644 index 00000000..146d12a1 --- /dev/null +++ b/tools/perf/arch/x86/util/header.c @@ -0,0 +1,59 @@ +#include <sys/types.h> +#include <unistd.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include "../../util/header.h" + +static inline void +cpuid(unsigned int op, unsigned int *a, unsigned int *b, unsigned int *c, + unsigned int *d) +{ + __asm__ __volatile__ (".byte 0x53\n\tcpuid\n\t" + "movl %%ebx, %%esi\n\t.byte 0x5b" + : "=a" (*a), + "=S" (*b), + "=c" (*c), + "=d" (*d) + : "a" (op)); +} + +int +get_cpuid(char *buffer, size_t sz) +{ + unsigned int a, b, c, d, lvl; + int family = -1, model = -1, step = -1; + int nb; + char vendor[16]; + + cpuid(0, &lvl, &b, &c, &d); + strncpy(&vendor[0], (char *)(&b), 4); + strncpy(&vendor[4], (char *)(&d), 4); + strncpy(&vendor[8], (char *)(&c), 4); + vendor[12] = '\0'; + + if (lvl >= 1) { + cpuid(1, &a, &b, &c, &d); + + family = (a >> 8) & 0xf; /* bits 11 - 8 */ + model = (a >> 4) & 0xf; /* Bits 7 - 4 */ + step = a & 0xf; + + /* extended family */ + if (family == 0xf) + family += (a >> 20) & 0xff; + + /* extended model */ + if (family >= 0x6) + model += ((a >> 16) & 0xf) << 4; + } + nb = scnprintf(buffer, sz, "%s,%u,%u,%u$", vendor, family, model, step); + + /* look for end marker to ensure the entire data fit */ + if (strchr(buffer, '$')) { + buffer[nb-1] = '\0'; + return 0; + } + return -1; +} diff --git a/tools/perf/arch/x86/util/unwind.c b/tools/perf/arch/x86/util/unwind.c new file mode 100644 index 00000000..78d956ef --- /dev/null +++ b/tools/perf/arch/x86/util/unwind.c @@ -0,0 +1,111 @@ + +#include <errno.h> +#include <libunwind.h> +#include "perf_regs.h" +#include "../../util/unwind.h" + +#ifdef ARCH_X86_64 +int unwind__arch_reg_id(int regnum) +{ + int id; + + switch (regnum) { + case UNW_X86_64_RAX: + id = PERF_REG_X86_AX; + break; + case UNW_X86_64_RDX: + id = PERF_REG_X86_DX; + break; + case UNW_X86_64_RCX: + id = PERF_REG_X86_CX; + break; + case UNW_X86_64_RBX: + id = PERF_REG_X86_BX; + break; + case UNW_X86_64_RSI: + id = PERF_REG_X86_SI; + break; + case UNW_X86_64_RDI: + id = PERF_REG_X86_DI; + break; + case UNW_X86_64_RBP: + id = PERF_REG_X86_BP; + break; + case UNW_X86_64_RSP: + id = PERF_REG_X86_SP; + break; + case UNW_X86_64_R8: + id = PERF_REG_X86_R8; + break; + case UNW_X86_64_R9: + id = PERF_REG_X86_R9; + break; + case UNW_X86_64_R10: + id = PERF_REG_X86_R10; + break; + case UNW_X86_64_R11: + id = PERF_REG_X86_R11; + break; + case UNW_X86_64_R12: + id = PERF_REG_X86_R12; + break; + case UNW_X86_64_R13: + id = PERF_REG_X86_R13; + break; + case UNW_X86_64_R14: + id = PERF_REG_X86_R14; + break; + case UNW_X86_64_R15: + id = PERF_REG_X86_R15; + break; + case UNW_X86_64_RIP: + id = PERF_REG_X86_IP; + break; + default: + pr_err("unwind: invalid reg id %d\n", regnum); + return -EINVAL; + } + + return id; +} +#else +int unwind__arch_reg_id(int regnum) +{ + int id; + + switch (regnum) { + case UNW_X86_EAX: + id = PERF_REG_X86_AX; + break; + case UNW_X86_EDX: + id = PERF_REG_X86_DX; + break; + case UNW_X86_ECX: + id = PERF_REG_X86_CX; + break; + case UNW_X86_EBX: + id = PERF_REG_X86_BX; + break; + case UNW_X86_ESI: + id = PERF_REG_X86_SI; + break; + case UNW_X86_EDI: + id = PERF_REG_X86_DI; + break; + case UNW_X86_EBP: + id = PERF_REG_X86_BP; + break; + case UNW_X86_ESP: + id = PERF_REG_X86_SP; + break; + case UNW_X86_EIP: + id = PERF_REG_X86_IP; + break; + default: + pr_err("unwind: invalid reg id %d\n", regnum); + return -EINVAL; + } + + return id; +} +#endif /* ARCH_X86_64 */ diff --git a/tools/perf/bash_completion b/tools/perf/bash_completion new file mode 100644 index 00000000..56e6a12a --- /dev/null +++ b/tools/perf/bash_completion @@ -0,0 +1,62 @@ +# perf completion + +function_exists() +{ + declare -F $1 > /dev/null + return $? +} + +function_exists __ltrim_colon_completions || +__ltrim_colon_completions() +{ + if [[ "$1" == *:* && "$COMP_WORDBREAKS" == *:* ]]; then + # Remove colon-word prefix from COMPREPLY items + local colon_word=${1%${1##*:}} + local i=${#COMPREPLY[*]} + while [[ $((--i)) -ge 0 ]]; do + COMPREPLY[$i]=${COMPREPLY[$i]#"$colon_word"} + done + fi +} + +have perf && +_perf() +{ + local cur prev cmd + + COMPREPLY=() + if function_exists _get_comp_words_by_ref; then + _get_comp_words_by_ref -n : cur prev + else + cur=$(_get_cword :) + prev=${COMP_WORDS[COMP_CWORD-1]} + fi + + cmd=${COMP_WORDS[0]} + + # List perf subcommands or long options + if [ $COMP_CWORD -eq 1 ]; then + if [[ $cur == --* ]]; then + COMPREPLY=( $( compgen -W '--help --version \ + --exec-path --html-path --paginate --no-pager \ + --perf-dir --work-tree --debugfs-dir' -- "$cur" ) ) + else + cmds=$($cmd --list-cmds) + COMPREPLY=( $( compgen -W '$cmds' -- "$cur" ) ) + fi + # List possible events for -e option + elif [[ $prev == "-e" && "${COMP_WORDS[1]}" == @(record|stat|top) ]]; then + evts=$($cmd list --raw-dump) + COMPREPLY=( $( compgen -W '$evts' -- "$cur" ) ) + __ltrim_colon_completions $cur + # List long option names + elif [[ $cur == --* ]]; then + subcmd=${COMP_WORDS[1]} + opts=$($cmd $subcmd --list-opts) + COMPREPLY=( $( compgen -W '$opts' -- "$cur" ) ) + # Fall down to list regular files + else + _filedir + fi +} && +complete -F _perf perf diff --git a/tools/perf/bench/bench.h b/tools/perf/bench/bench.h new file mode 100644 index 00000000..a5223e6a --- /dev/null +++ b/tools/perf/bench/bench.h @@ -0,0 +1,20 @@ +#ifndef BENCH_H +#define BENCH_H + +extern int bench_numa(int argc, const char **argv, const char *prefix); +extern int bench_sched_messaging(int argc, const char **argv, const char *prefix); +extern int bench_sched_pipe(int argc, const char **argv, const char *prefix); +extern int bench_mem_memcpy(int argc, const char **argv, + const char *prefix __maybe_unused); +extern int bench_mem_memset(int argc, const char **argv, const char *prefix); + +#define BENCH_FORMAT_DEFAULT_STR "default" +#define BENCH_FORMAT_DEFAULT 0 +#define BENCH_FORMAT_SIMPLE_STR "simple" +#define BENCH_FORMAT_SIMPLE 1 + +#define BENCH_FORMAT_UNKNOWN -1 + +extern int bench_format; + +#endif diff --git a/tools/perf/bench/mem-memcpy-arch.h b/tools/perf/bench/mem-memcpy-arch.h new file mode 100644 index 00000000..a72e36cb --- /dev/null +++ b/tools/perf/bench/mem-memcpy-arch.h @@ -0,0 +1,12 @@ + +#ifdef ARCH_X86_64 + +#define MEMCPY_FN(fn, name, desc) \ + extern void *fn(void *, const void *, size_t); + +#include "mem-memcpy-x86-64-asm-def.h" + +#undef MEMCPY_FN + +#endif + diff --git a/tools/perf/bench/mem-memcpy-x86-64-asm-def.h b/tools/perf/bench/mem-memcpy-x86-64-asm-def.h new file mode 100644 index 00000000..d66ab799 --- /dev/null +++ b/tools/perf/bench/mem-memcpy-x86-64-asm-def.h @@ -0,0 +1,12 @@ + +MEMCPY_FN(__memcpy, + "x86-64-unrolled", + "unrolled memcpy() in arch/x86/lib/memcpy_64.S") + +MEMCPY_FN(memcpy_c, + "x86-64-movsq", + "movsq-based memcpy() in arch/x86/lib/memcpy_64.S") + +MEMCPY_FN(memcpy_c_e, + "x86-64-movsb", + "movsb-based memcpy() in arch/x86/lib/memcpy_64.S") diff --git a/tools/perf/bench/mem-memcpy-x86-64-asm.S b/tools/perf/bench/mem-memcpy-x86-64-asm.S new file mode 100644 index 00000000..fcd9cf00 --- /dev/null +++ b/tools/perf/bench/mem-memcpy-x86-64-asm.S @@ -0,0 +1,12 @@ +#define memcpy MEMCPY /* don't hide glibc's memcpy() */ +#define altinstr_replacement text +#define globl p2align 4; .globl +#define Lmemcpy_c globl memcpy_c; memcpy_c +#define Lmemcpy_c_e globl memcpy_c_e; memcpy_c_e +#include "../../../arch/x86/lib/memcpy_64.S" +/* + * We need to provide note.GNU-stack section, saying that we want + * NOT executable stack. Otherwise the final linking will assume that + * the ELF stack should not be restricted at all and set it RWX. + */ +.section .note.GNU-stack,"",@progbits diff --git a/tools/perf/bench/mem-memcpy.c b/tools/perf/bench/mem-memcpy.c new file mode 100644 index 00000000..93c83e3c --- /dev/null +++ b/tools/perf/bench/mem-memcpy.c @@ -0,0 +1,303 @@ +/* + * mem-memcpy.c + * + * memcpy: Simple memory copy in various ways + * + * Written by Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp> + */ + +#include "../perf.h" +#include "../util/util.h" +#include "../util/parse-options.h" +#include "../util/header.h" +#include "bench.h" +#include "mem-memcpy-arch.h" + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <sys/time.h> +#include <errno.h> + +#define K 1024 + +static const char *length_str = "1MB"; +static const char *routine = "default"; +static int iterations = 1; +static bool use_cycle; +static int cycle_fd; +static bool only_prefault; +static bool no_prefault; + +static const struct option options[] = { + OPT_STRING('l', "length", &length_str, "1MB", + "Specify length of memory to copy. " + "Available units: B, KB, MB, GB and TB (upper and lower)"), + OPT_STRING('r', "routine", &routine, "default", + "Specify routine to copy"), + OPT_INTEGER('i', "iterations", &iterations, + "repeat memcpy() invocation this number of times"), + OPT_BOOLEAN('c', "cycle", &use_cycle, + "Use cycles event instead of gettimeofday() for measuring"), + OPT_BOOLEAN('o', "only-prefault", &only_prefault, + "Show only the result with page faults before memcpy()"), + OPT_BOOLEAN('n', "no-prefault", &no_prefault, + "Show only the result without page faults before memcpy()"), + OPT_END() +}; + +typedef void *(*memcpy_t)(void *, const void *, size_t); + +struct routine { + const char *name; + const char *desc; + memcpy_t fn; +}; + +struct routine routines[] = { + { "default", + "Default memcpy() provided by glibc", + memcpy }, +#ifdef ARCH_X86_64 + +#define MEMCPY_FN(fn, name, desc) { name, desc, fn }, +#include "mem-memcpy-x86-64-asm-def.h" +#undef MEMCPY_FN + +#endif + + { NULL, + NULL, + NULL } +}; + +static const char * const bench_mem_memcpy_usage[] = { + "perf bench mem memcpy <options>", + NULL +}; + +static struct perf_event_attr cycle_attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES +}; + +static void init_cycle(void) +{ + cycle_fd = sys_perf_event_open(&cycle_attr, getpid(), -1, -1, 0); + + if (cycle_fd < 0 && errno == ENOSYS) + die("No CONFIG_PERF_EVENTS=y kernel support configured?\n"); + else + BUG_ON(cycle_fd < 0); +} + +static u64 get_cycle(void) +{ + int ret; + u64 clk; + + ret = read(cycle_fd, &clk, sizeof(u64)); + BUG_ON(ret != sizeof(u64)); + + return clk; +} + +static double timeval2double(struct timeval *ts) +{ + return (double)ts->tv_sec + + (double)ts->tv_usec / (double)1000000; +} + +static void alloc_mem(void **dst, void **src, size_t length) +{ + *dst = zalloc(length); + if (!dst) + die("memory allocation failed - maybe length is too large?\n"); + + *src = zalloc(length); + if (!src) + die("memory allocation failed - maybe length is too large?\n"); +} + +static u64 do_memcpy_cycle(memcpy_t fn, size_t len, bool prefault) +{ + u64 cycle_start = 0ULL, cycle_end = 0ULL; + void *src = NULL, *dst = NULL; + int i; + + alloc_mem(&src, &dst, len); + + if (prefault) + fn(dst, src, len); + + cycle_start = get_cycle(); + for (i = 0; i < iterations; ++i) + fn(dst, src, len); + cycle_end = get_cycle(); + + free(src); + free(dst); + return cycle_end - cycle_start; +} + +static double do_memcpy_gettimeofday(memcpy_t fn, size_t len, bool prefault) +{ + struct timeval tv_start, tv_end, tv_diff; + void *src = NULL, *dst = NULL; + int i; + + alloc_mem(&src, &dst, len); + + if (prefault) + fn(dst, src, len); + + BUG_ON(gettimeofday(&tv_start, NULL)); + for (i = 0; i < iterations; ++i) + fn(dst, src, len); + BUG_ON(gettimeofday(&tv_end, NULL)); + + timersub(&tv_end, &tv_start, &tv_diff); + + free(src); + free(dst); + return (double)((double)len / timeval2double(&tv_diff)); +} + +#define pf (no_prefault ? 0 : 1) + +#define print_bps(x) do { \ + if (x < K) \ + printf(" %14lf B/Sec", x); \ + else if (x < K * K) \ + printf(" %14lfd KB/Sec", x / K); \ + else if (x < K * K * K) \ + printf(" %14lf MB/Sec", x / K / K); \ + else \ + printf(" %14lf GB/Sec", x / K / K / K); \ + } while (0) + +int bench_mem_memcpy(int argc, const char **argv, + const char *prefix __maybe_unused) +{ + int i; + size_t len; + double result_bps[2]; + u64 result_cycle[2]; + + argc = parse_options(argc, argv, options, + bench_mem_memcpy_usage, 0); + + if (use_cycle) + init_cycle(); + + len = (size_t)perf_atoll((char *)length_str); + + result_cycle[0] = result_cycle[1] = 0ULL; + result_bps[0] = result_bps[1] = 0.0; + + if ((s64)len <= 0) { + fprintf(stderr, "Invalid length:%s\n", length_str); + return 1; + } + + /* same to without specifying either of prefault and no-prefault */ + if (only_prefault && no_prefault) + only_prefault = no_prefault = false; + + for (i = 0; routines[i].name; i++) { + if (!strcmp(routines[i].name, routine)) + break; + } + if (!routines[i].name) { + printf("Unknown routine:%s\n", routine); + printf("Available routines...\n"); + for (i = 0; routines[i].name; i++) { + printf("\t%s ... %s\n", + routines[i].name, routines[i].desc); + } + return 1; + } + + if (bench_format == BENCH_FORMAT_DEFAULT) + printf("# Copying %s Bytes ...\n\n", length_str); + + if (!only_prefault && !no_prefault) { + /* show both of results */ + if (use_cycle) { + result_cycle[0] = + do_memcpy_cycle(routines[i].fn, len, false); + result_cycle[1] = + do_memcpy_cycle(routines[i].fn, len, true); + } else { + result_bps[0] = + do_memcpy_gettimeofday(routines[i].fn, + len, false); + result_bps[1] = + do_memcpy_gettimeofday(routines[i].fn, + len, true); + } + } else { + if (use_cycle) { + result_cycle[pf] = + do_memcpy_cycle(routines[i].fn, + len, only_prefault); + } else { + result_bps[pf] = + do_memcpy_gettimeofday(routines[i].fn, + len, only_prefault); + } + } + + switch (bench_format) { + case BENCH_FORMAT_DEFAULT: + if (!only_prefault && !no_prefault) { + if (use_cycle) { + printf(" %14lf Cycle/Byte\n", + (double)result_cycle[0] + / (double)len); + printf(" %14lf Cycle/Byte (with prefault)\n", + (double)result_cycle[1] + / (double)len); + } else { + print_bps(result_bps[0]); + printf("\n"); + print_bps(result_bps[1]); + printf(" (with prefault)\n"); + } + } else { + if (use_cycle) { + printf(" %14lf Cycle/Byte", + (double)result_cycle[pf] + / (double)len); + } else + print_bps(result_bps[pf]); + + printf("%s\n", only_prefault ? " (with prefault)" : ""); + } + break; + case BENCH_FORMAT_SIMPLE: + if (!only_prefault && !no_prefault) { + if (use_cycle) { + printf("%lf %lf\n", + (double)result_cycle[0] / (double)len, + (double)result_cycle[1] / (double)len); + } else { + printf("%lf %lf\n", + result_bps[0], result_bps[1]); + } + } else { + if (use_cycle) { + printf("%lf\n", (double)result_cycle[pf] + / (double)len); + } else + printf("%lf\n", result_bps[pf]); + } + break; + default: + /* reaching this means there's some disaster: */ + die("unknown format: %d\n", bench_format); + break; + } + + return 0; +} diff --git a/tools/perf/bench/mem-memset-arch.h b/tools/perf/bench/mem-memset-arch.h new file mode 100644 index 00000000..a040fa77 --- /dev/null +++ b/tools/perf/bench/mem-memset-arch.h @@ -0,0 +1,12 @@ + +#ifdef ARCH_X86_64 + +#define MEMSET_FN(fn, name, desc) \ + extern void *fn(void *, int, size_t); + +#include "mem-memset-x86-64-asm-def.h" + +#undef MEMSET_FN + +#endif + diff --git a/tools/perf/bench/mem-memset-x86-64-asm-def.h b/tools/perf/bench/mem-memset-x86-64-asm-def.h new file mode 100644 index 00000000..a71dff97 --- /dev/null +++ b/tools/perf/bench/mem-memset-x86-64-asm-def.h @@ -0,0 +1,12 @@ + +MEMSET_FN(__memset, + "x86-64-unrolled", + "unrolled memset() in arch/x86/lib/memset_64.S") + +MEMSET_FN(memset_c, + "x86-64-stosq", + "movsq-based memset() in arch/x86/lib/memset_64.S") + +MEMSET_FN(memset_c_e, + "x86-64-stosb", + "movsb-based memset() in arch/x86/lib/memset_64.S") diff --git a/tools/perf/bench/mem-memset-x86-64-asm.S b/tools/perf/bench/mem-memset-x86-64-asm.S new file mode 100644 index 00000000..9e5af89e --- /dev/null +++ b/tools/perf/bench/mem-memset-x86-64-asm.S @@ -0,0 +1,13 @@ +#define memset MEMSET /* don't hide glibc's memset() */ +#define altinstr_replacement text +#define globl p2align 4; .globl +#define Lmemset_c globl memset_c; memset_c +#define Lmemset_c_e globl memset_c_e; memset_c_e +#include "../../../arch/x86/lib/memset_64.S" + +/* + * We need to provide note.GNU-stack section, saying that we want + * NOT executable stack. Otherwise the final linking will assume that + * the ELF stack should not be restricted at all and set it RWX. + */ +.section .note.GNU-stack,"",@progbits diff --git a/tools/perf/bench/mem-memset.c b/tools/perf/bench/mem-memset.c new file mode 100644 index 00000000..c6e4bc52 --- /dev/null +++ b/tools/perf/bench/mem-memset.c @@ -0,0 +1,297 @@ +/* + * mem-memset.c + * + * memset: Simple memory set in various ways + * + * Trivial clone of mem-memcpy.c. + */ + +#include "../perf.h" +#include "../util/util.h" +#include "../util/parse-options.h" +#include "../util/header.h" +#include "bench.h" +#include "mem-memset-arch.h" + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <sys/time.h> +#include <errno.h> + +#define K 1024 + +static const char *length_str = "1MB"; +static const char *routine = "default"; +static int iterations = 1; +static bool use_cycle; +static int cycle_fd; +static bool only_prefault; +static bool no_prefault; + +static const struct option options[] = { + OPT_STRING('l', "length", &length_str, "1MB", + "Specify length of memory to set. " + "Available units: B, KB, MB, GB and TB (upper and lower)"), + OPT_STRING('r', "routine", &routine, "default", + "Specify routine to set"), + OPT_INTEGER('i', "iterations", &iterations, + "repeat memset() invocation this number of times"), + OPT_BOOLEAN('c', "cycle", &use_cycle, + "Use cycles event instead of gettimeofday() for measuring"), + OPT_BOOLEAN('o', "only-prefault", &only_prefault, + "Show only the result with page faults before memset()"), + OPT_BOOLEAN('n', "no-prefault", &no_prefault, + "Show only the result without page faults before memset()"), + OPT_END() +}; + +typedef void *(*memset_t)(void *, int, size_t); + +struct routine { + const char *name; + const char *desc; + memset_t fn; +}; + +static const struct routine routines[] = { + { "default", + "Default memset() provided by glibc", + memset }, +#ifdef ARCH_X86_64 + +#define MEMSET_FN(fn, name, desc) { name, desc, fn }, +#include "mem-memset-x86-64-asm-def.h" +#undef MEMSET_FN + +#endif + + { NULL, + NULL, + NULL } +}; + +static const char * const bench_mem_memset_usage[] = { + "perf bench mem memset <options>", + NULL +}; + +static struct perf_event_attr cycle_attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES +}; + +static void init_cycle(void) +{ + cycle_fd = sys_perf_event_open(&cycle_attr, getpid(), -1, -1, 0); + + if (cycle_fd < 0 && errno == ENOSYS) + die("No CONFIG_PERF_EVENTS=y kernel support configured?\n"); + else + BUG_ON(cycle_fd < 0); +} + +static u64 get_cycle(void) +{ + int ret; + u64 clk; + + ret = read(cycle_fd, &clk, sizeof(u64)); + BUG_ON(ret != sizeof(u64)); + + return clk; +} + +static double timeval2double(struct timeval *ts) +{ + return (double)ts->tv_sec + + (double)ts->tv_usec / (double)1000000; +} + +static void alloc_mem(void **dst, size_t length) +{ + *dst = zalloc(length); + if (!dst) + die("memory allocation failed - maybe length is too large?\n"); +} + +static u64 do_memset_cycle(memset_t fn, size_t len, bool prefault) +{ + u64 cycle_start = 0ULL, cycle_end = 0ULL; + void *dst = NULL; + int i; + + alloc_mem(&dst, len); + + if (prefault) + fn(dst, -1, len); + + cycle_start = get_cycle(); + for (i = 0; i < iterations; ++i) + fn(dst, i, len); + cycle_end = get_cycle(); + + free(dst); + return cycle_end - cycle_start; +} + +static double do_memset_gettimeofday(memset_t fn, size_t len, bool prefault) +{ + struct timeval tv_start, tv_end, tv_diff; + void *dst = NULL; + int i; + + alloc_mem(&dst, len); + + if (prefault) + fn(dst, -1, len); + + BUG_ON(gettimeofday(&tv_start, NULL)); + for (i = 0; i < iterations; ++i) + fn(dst, i, len); + BUG_ON(gettimeofday(&tv_end, NULL)); + + timersub(&tv_end, &tv_start, &tv_diff); + + free(dst); + return (double)((double)len / timeval2double(&tv_diff)); +} + +#define pf (no_prefault ? 0 : 1) + +#define print_bps(x) do { \ + if (x < K) \ + printf(" %14lf B/Sec", x); \ + else if (x < K * K) \ + printf(" %14lfd KB/Sec", x / K); \ + else if (x < K * K * K) \ + printf(" %14lf MB/Sec", x / K / K); \ + else \ + printf(" %14lf GB/Sec", x / K / K / K); \ + } while (0) + +int bench_mem_memset(int argc, const char **argv, + const char *prefix __maybe_unused) +{ + int i; + size_t len; + double result_bps[2]; + u64 result_cycle[2]; + + argc = parse_options(argc, argv, options, + bench_mem_memset_usage, 0); + + if (use_cycle) + init_cycle(); + + len = (size_t)perf_atoll((char *)length_str); + + result_cycle[0] = result_cycle[1] = 0ULL; + result_bps[0] = result_bps[1] = 0.0; + + if ((s64)len <= 0) { + fprintf(stderr, "Invalid length:%s\n", length_str); + return 1; + } + + /* same to without specifying either of prefault and no-prefault */ + if (only_prefault && no_prefault) + only_prefault = no_prefault = false; + + for (i = 0; routines[i].name; i++) { + if (!strcmp(routines[i].name, routine)) + break; + } + if (!routines[i].name) { + printf("Unknown routine:%s\n", routine); + printf("Available routines...\n"); + for (i = 0; routines[i].name; i++) { + printf("\t%s ... %s\n", + routines[i].name, routines[i].desc); + } + return 1; + } + + if (bench_format == BENCH_FORMAT_DEFAULT) + printf("# Copying %s Bytes ...\n\n", length_str); + + if (!only_prefault && !no_prefault) { + /* show both of results */ + if (use_cycle) { + result_cycle[0] = + do_memset_cycle(routines[i].fn, len, false); + result_cycle[1] = + do_memset_cycle(routines[i].fn, len, true); + } else { + result_bps[0] = + do_memset_gettimeofday(routines[i].fn, + len, false); + result_bps[1] = + do_memset_gettimeofday(routines[i].fn, + len, true); + } + } else { + if (use_cycle) { + result_cycle[pf] = + do_memset_cycle(routines[i].fn, + len, only_prefault); + } else { + result_bps[pf] = + do_memset_gettimeofday(routines[i].fn, + len, only_prefault); + } + } + + switch (bench_format) { + case BENCH_FORMAT_DEFAULT: + if (!only_prefault && !no_prefault) { + if (use_cycle) { + printf(" %14lf Cycle/Byte\n", + (double)result_cycle[0] + / (double)len); + printf(" %14lf Cycle/Byte (with prefault)\n ", + (double)result_cycle[1] + / (double)len); + } else { + print_bps(result_bps[0]); + printf("\n"); + print_bps(result_bps[1]); + printf(" (with prefault)\n"); + } + } else { + if (use_cycle) { + printf(" %14lf Cycle/Byte", + (double)result_cycle[pf] + / (double)len); + } else + print_bps(result_bps[pf]); + + printf("%s\n", only_prefault ? " (with prefault)" : ""); + } + break; + case BENCH_FORMAT_SIMPLE: + if (!only_prefault && !no_prefault) { + if (use_cycle) { + printf("%lf %lf\n", + (double)result_cycle[0] / (double)len, + (double)result_cycle[1] / (double)len); + } else { + printf("%lf %lf\n", + result_bps[0], result_bps[1]); + } + } else { + if (use_cycle) { + printf("%lf\n", (double)result_cycle[pf] + / (double)len); + } else + printf("%lf\n", result_bps[pf]); + } + break; + default: + /* reaching this means there's some disaster: */ + die("unknown format: %d\n", bench_format); + break; + } + + return 0; +} diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c new file mode 100644 index 00000000..30d1c322 --- /dev/null +++ b/tools/perf/bench/numa.c @@ -0,0 +1,1731 @@ +/* + * numa.c + * + * numa: Simulate NUMA-sensitive workload and measure their NUMA performance + */ + +#include "../perf.h" +#include "../builtin.h" +#include "../util/util.h" +#include "../util/parse-options.h" + +#include "bench.h" + +#include <errno.h> +#include <sched.h> +#include <stdio.h> +#include <assert.h> +#include <malloc.h> +#include <signal.h> +#include <stdlib.h> +#include <string.h> +#include <unistd.h> +#include <pthread.h> +#include <sys/mman.h> +#include <sys/time.h> +#include <sys/wait.h> +#include <sys/prctl.h> +#include <sys/types.h> + +#include <numa.h> +#include <numaif.h> + +/* + * Regular printout to the terminal, supressed if -q is specified: + */ +#define tprintf(x...) do { if (g && g->p.show_details >= 0) printf(x); } while (0) + +/* + * Debug printf: + */ +#define dprintf(x...) do { if (g && g->p.show_details >= 1) printf(x); } while (0) + +struct thread_data { + int curr_cpu; + cpu_set_t bind_cpumask; + int bind_node; + u8 *process_data; + int process_nr; + int thread_nr; + int task_nr; + unsigned int loops_done; + u64 val; + u64 runtime_ns; + pthread_mutex_t *process_lock; +}; + +/* Parameters set by options: */ + +struct params { + /* Startup synchronization: */ + bool serialize_startup; + + /* Task hierarchy: */ + int nr_proc; + int nr_threads; + + /* Working set sizes: */ + const char *mb_global_str; + const char *mb_proc_str; + const char *mb_proc_locked_str; + const char *mb_thread_str; + + double mb_global; + double mb_proc; + double mb_proc_locked; + double mb_thread; + + /* Access patterns to the working set: */ + bool data_reads; + bool data_writes; + bool data_backwards; + bool data_zero_memset; + bool data_rand_walk; + u32 nr_loops; + u32 nr_secs; + u32 sleep_usecs; + + /* Working set initialization: */ + bool init_zero; + bool init_random; + bool init_cpu0; + + /* Misc options: */ + int show_details; + int run_all; + int thp; + + long bytes_global; + long bytes_process; + long bytes_process_locked; + long bytes_thread; + + int nr_tasks; + bool show_quiet; + + bool show_convergence; + bool measure_convergence; + + int perturb_secs; + int nr_cpus; + int nr_nodes; + + /* Affinity options -C and -N: */ + char *cpu_list_str; + char *node_list_str; +}; + + +/* Global, read-writable area, accessible to all processes and threads: */ + +struct global_info { + u8 *data; + + pthread_mutex_t startup_mutex; + int nr_tasks_started; + + pthread_mutex_t startup_done_mutex; + + pthread_mutex_t start_work_mutex; + int nr_tasks_working; + + pthread_mutex_t stop_work_mutex; + u64 bytes_done; + + struct thread_data *threads; + + /* Convergence latency measurement: */ + bool all_converged; + bool stop_work; + + int print_once; + + struct params p; +}; + +static struct global_info *g = NULL; + +static int parse_cpus_opt(const struct option *opt, const char *arg, int unset); +static int parse_nodes_opt(const struct option *opt, const char *arg, int unset); + +struct params p0; + +static const struct option options[] = { + OPT_INTEGER('p', "nr_proc" , &p0.nr_proc, "number of processes"), + OPT_INTEGER('t', "nr_threads" , &p0.nr_threads, "number of threads per process"), + + OPT_STRING('G', "mb_global" , &p0.mb_global_str, "MB", "global memory (MBs)"), + OPT_STRING('P', "mb_proc" , &p0.mb_proc_str, "MB", "process memory (MBs)"), + OPT_STRING('L', "mb_proc_locked", &p0.mb_proc_locked_str,"MB", "process serialized/locked memory access (MBs), <= process_memory"), + OPT_STRING('T', "mb_thread" , &p0.mb_thread_str, "MB", "thread memory (MBs)"), + + OPT_UINTEGER('l', "nr_loops" , &p0.nr_loops, "max number of loops to run"), + OPT_UINTEGER('s', "nr_secs" , &p0.nr_secs, "max number of seconds to run"), + OPT_UINTEGER('u', "usleep" , &p0.sleep_usecs, "usecs to sleep per loop iteration"), + + OPT_BOOLEAN('R', "data_reads" , &p0.data_reads, "access the data via writes (can be mixed with -W)"), + OPT_BOOLEAN('W', "data_writes" , &p0.data_writes, "access the data via writes (can be mixed with -R)"), + OPT_BOOLEAN('B', "data_backwards", &p0.data_backwards, "access the data backwards as well"), + OPT_BOOLEAN('Z', "data_zero_memset", &p0.data_zero_memset,"access the data via glibc bzero only"), + OPT_BOOLEAN('r', "data_rand_walk", &p0.data_rand_walk, "access the data with random (32bit LFSR) walk"), + + + OPT_BOOLEAN('z', "init_zero" , &p0.init_zero, "bzero the initial allocations"), + OPT_BOOLEAN('I', "init_random" , &p0.init_random, "randomize the contents of the initial allocations"), + OPT_BOOLEAN('0', "init_cpu0" , &p0.init_cpu0, "do the initial allocations on CPU#0"), + OPT_INTEGER('x', "perturb_secs", &p0.perturb_secs, "perturb thread 0/0 every X secs, to test convergence stability"), + + OPT_INCR ('d', "show_details" , &p0.show_details, "Show details"), + OPT_INCR ('a', "all" , &p0.run_all, "Run all tests in the suite"), + OPT_INTEGER('H', "thp" , &p0.thp, "MADV_NOHUGEPAGE < 0 < MADV_HUGEPAGE"), + OPT_BOOLEAN('c', "show_convergence", &p0.show_convergence, "show convergence details"), + OPT_BOOLEAN('m', "measure_convergence", &p0.measure_convergence, "measure convergence latency"), + OPT_BOOLEAN('q', "quiet" , &p0.show_quiet, "bzero the initial allocations"), + OPT_BOOLEAN('S', "serialize-startup", &p0.serialize_startup,"serialize thread startup"), + + /* Special option string parsing callbacks: */ + OPT_CALLBACK('C', "cpus", NULL, "cpu[,cpu2,...cpuN]", + "bind the first N tasks to these specific cpus (the rest is unbound)", + parse_cpus_opt), + OPT_CALLBACK('M', "memnodes", NULL, "node[,node2,...nodeN]", + "bind the first N tasks to these specific memory nodes (the rest is unbound)", + parse_nodes_opt), + OPT_END() +}; + +static const char * const bench_numa_usage[] = { + "perf bench numa <options>", + NULL +}; + +static const char * const numa_usage[] = { + "perf bench numa mem [<options>]", + NULL +}; + +static cpu_set_t bind_to_cpu(int target_cpu) +{ + cpu_set_t orig_mask, mask; + int ret; + + ret = sched_getaffinity(0, sizeof(orig_mask), &orig_mask); + BUG_ON(ret); + + CPU_ZERO(&mask); + + if (target_cpu == -1) { + int cpu; + + for (cpu = 0; cpu < g->p.nr_cpus; cpu++) + CPU_SET(cpu, &mask); + } else { + BUG_ON(target_cpu < 0 || target_cpu >= g->p.nr_cpus); + CPU_SET(target_cpu, &mask); + } + + ret = sched_setaffinity(0, sizeof(mask), &mask); + BUG_ON(ret); + + return orig_mask; +} + +static cpu_set_t bind_to_node(int target_node) +{ + int cpus_per_node = g->p.nr_cpus/g->p.nr_nodes; + cpu_set_t orig_mask, mask; + int cpu; + int ret; + + BUG_ON(cpus_per_node*g->p.nr_nodes != g->p.nr_cpus); + BUG_ON(!cpus_per_node); + + ret = sched_getaffinity(0, sizeof(orig_mask), &orig_mask); + BUG_ON(ret); + + CPU_ZERO(&mask); + + if (target_node == -1) { + for (cpu = 0; cpu < g->p.nr_cpus; cpu++) + CPU_SET(cpu, &mask); + } else { + int cpu_start = (target_node + 0) * cpus_per_node; + int cpu_stop = (target_node + 1) * cpus_per_node; + + BUG_ON(cpu_stop > g->p.nr_cpus); + + for (cpu = cpu_start; cpu < cpu_stop; cpu++) + CPU_SET(cpu, &mask); + } + + ret = sched_setaffinity(0, sizeof(mask), &mask); + BUG_ON(ret); + + return orig_mask; +} + +static void bind_to_cpumask(cpu_set_t mask) +{ + int ret; + + ret = sched_setaffinity(0, sizeof(mask), &mask); + BUG_ON(ret); +} + +static void mempol_restore(void) +{ + int ret; + + ret = set_mempolicy(MPOL_DEFAULT, NULL, g->p.nr_nodes-1); + + BUG_ON(ret); +} + +static void bind_to_memnode(int node) +{ + unsigned long nodemask; + int ret; + + if (node == -1) + return; + + BUG_ON(g->p.nr_nodes > (int)sizeof(nodemask)); + nodemask = 1L << node; + + ret = set_mempolicy(MPOL_BIND, &nodemask, sizeof(nodemask)*8); + dprintf("binding to node %d, mask: %016lx => %d\n", node, nodemask, ret); + + BUG_ON(ret); +} + +#define HPSIZE (2*1024*1024) + +#define set_taskname(fmt...) \ +do { \ + char name[20]; \ + \ + snprintf(name, 20, fmt); \ + prctl(PR_SET_NAME, name); \ +} while (0) + +static u8 *alloc_data(ssize_t bytes0, int map_flags, + int init_zero, int init_cpu0, int thp, int init_random) +{ + cpu_set_t orig_mask; + ssize_t bytes; + u8 *buf; + int ret; + + if (!bytes0) + return NULL; + + /* Allocate and initialize all memory on CPU#0: */ + if (init_cpu0) { + orig_mask = bind_to_node(0); + bind_to_memnode(0); + } + + bytes = bytes0 + HPSIZE; + + buf = (void *)mmap(0, bytes, PROT_READ|PROT_WRITE, MAP_ANON|map_flags, -1, 0); + BUG_ON(buf == (void *)-1); + + if (map_flags == MAP_PRIVATE) { + if (thp > 0) { + ret = madvise(buf, bytes, MADV_HUGEPAGE); + if (ret && !g->print_once) { + g->print_once = 1; + printf("WARNING: Could not enable THP - do: 'echo madvise > /sys/kernel/mm/transparent_hugepage/enabled'\n"); + } + } + if (thp < 0) { + ret = madvise(buf, bytes, MADV_NOHUGEPAGE); + if (ret && !g->print_once) { + g->print_once = 1; + printf("WARNING: Could not disable THP: run a CONFIG_TRANSPARENT_HUGEPAGE kernel?\n"); + } + } + } + + if (init_zero) { + bzero(buf, bytes); + } else { + /* Initialize random contents, different in each word: */ + if (init_random) { + u64 *wbuf = (void *)buf; + long off = rand(); + long i; + + for (i = 0; i < bytes/8; i++) + wbuf[i] = i + off; + } + } + + /* Align to 2MB boundary: */ + buf = (void *)(((unsigned long)buf + HPSIZE-1) & ~(HPSIZE-1)); + + /* Restore affinity: */ + if (init_cpu0) { + bind_to_cpumask(orig_mask); + mempol_restore(); + } + + return buf; +} + +static void free_data(void *data, ssize_t bytes) +{ + int ret; + + if (!data) + return; + + ret = munmap(data, bytes); + BUG_ON(ret); +} + +/* + * Create a shared memory buffer that can be shared between processes, zeroed: + */ +static void * zalloc_shared_data(ssize_t bytes) +{ + return alloc_data(bytes, MAP_SHARED, 1, g->p.init_cpu0, g->p.thp, g->p.init_random); +} + +/* + * Create a shared memory buffer that can be shared between processes: + */ +static void * setup_shared_data(ssize_t bytes) +{ + return alloc_data(bytes, MAP_SHARED, 0, g->p.init_cpu0, g->p.thp, g->p.init_random); +} + +/* + * Allocate process-local memory - this will either be shared between + * threads of this process, or only be accessed by this thread: + */ +static void * setup_private_data(ssize_t bytes) +{ + return alloc_data(bytes, MAP_PRIVATE, 0, g->p.init_cpu0, g->p.thp, g->p.init_random); +} + +/* + * Return a process-shared (global) mutex: + */ +static void init_global_mutex(pthread_mutex_t *mutex) +{ + pthread_mutexattr_t attr; + + pthread_mutexattr_init(&attr); + pthread_mutexattr_setpshared(&attr, PTHREAD_PROCESS_SHARED); + pthread_mutex_init(mutex, &attr); +} + +static int parse_cpu_list(const char *arg) +{ + p0.cpu_list_str = strdup(arg); + + dprintf("got CPU list: {%s}\n", p0.cpu_list_str); + + return 0; +} + +static void parse_setup_cpu_list(void) +{ + struct thread_data *td; + char *str0, *str; + int t; + + if (!g->p.cpu_list_str) + return; + + dprintf("g->p.nr_tasks: %d\n", g->p.nr_tasks); + + str0 = str = strdup(g->p.cpu_list_str); + t = 0; + + BUG_ON(!str); + + tprintf("# binding tasks to CPUs:\n"); + tprintf("# "); + + while (true) { + int bind_cpu, bind_cpu_0, bind_cpu_1; + char *tok, *tok_end, *tok_step, *tok_len, *tok_mul; + int bind_len; + int step; + int mul; + + tok = strsep(&str, ","); + if (!tok) + break; + + tok_end = strstr(tok, "-"); + + dprintf("\ntoken: {%s}, end: {%s}\n", tok, tok_end); + if (!tok_end) { + /* Single CPU specified: */ + bind_cpu_0 = bind_cpu_1 = atol(tok); + } else { + /* CPU range specified (for example: "5-11"): */ + bind_cpu_0 = atol(tok); + bind_cpu_1 = atol(tok_end + 1); + } + + step = 1; + tok_step = strstr(tok, "#"); + if (tok_step) { + step = atol(tok_step + 1); + BUG_ON(step <= 0 || step >= g->p.nr_cpus); + } + + /* + * Mask length. + * Eg: "--cpus 8_4-16#4" means: '--cpus 8_4,12_4,16_4', + * where the _4 means the next 4 CPUs are allowed. + */ + bind_len = 1; + tok_len = strstr(tok, "_"); + if (tok_len) { + bind_len = atol(tok_len + 1); + BUG_ON(bind_len <= 0 || bind_len > g->p.nr_cpus); + } + + /* Multiplicator shortcut, "0x8" is a shortcut for: "0,0,0,0,0,0,0,0" */ + mul = 1; + tok_mul = strstr(tok, "x"); + if (tok_mul) { + mul = atol(tok_mul + 1); + BUG_ON(mul <= 0); + } + + dprintf("CPUs: %d_%d-%d#%dx%d\n", bind_cpu_0, bind_len, bind_cpu_1, step, mul); + + BUG_ON(bind_cpu_0 < 0 || bind_cpu_0 >= g->p.nr_cpus); + BUG_ON(bind_cpu_1 < 0 || bind_cpu_1 >= g->p.nr_cpus); + BUG_ON(bind_cpu_0 > bind_cpu_1); + + for (bind_cpu = bind_cpu_0; bind_cpu <= bind_cpu_1; bind_cpu += step) { + int i; + + for (i = 0; i < mul; i++) { + int cpu; + + if (t >= g->p.nr_tasks) { + printf("\n# NOTE: ignoring bind CPUs starting at CPU#%d\n #", bind_cpu); + goto out; + } + td = g->threads + t; + + if (t) + tprintf(","); + if (bind_len > 1) { + tprintf("%2d/%d", bind_cpu, bind_len); + } else { + tprintf("%2d", bind_cpu); + } + + CPU_ZERO(&td->bind_cpumask); + for (cpu = bind_cpu; cpu < bind_cpu+bind_len; cpu++) { + BUG_ON(cpu < 0 || cpu >= g->p.nr_cpus); + CPU_SET(cpu, &td->bind_cpumask); + } + t++; + } + } + } +out: + + tprintf("\n"); + + if (t < g->p.nr_tasks) + printf("# NOTE: %d tasks bound, %d tasks unbound\n", t, g->p.nr_tasks - t); + + free(str0); +} + +static int parse_cpus_opt(const struct option *opt __maybe_unused, + const char *arg, int unset __maybe_unused) +{ + if (!arg) + return -1; + + return parse_cpu_list(arg); +} + +static int parse_node_list(const char *arg) +{ + p0.node_list_str = strdup(arg); + + dprintf("got NODE list: {%s}\n", p0.node_list_str); + + return 0; +} + +static void parse_setup_node_list(void) +{ + struct thread_data *td; + char *str0, *str; + int t; + + if (!g->p.node_list_str) + return; + + dprintf("g->p.nr_tasks: %d\n", g->p.nr_tasks); + + str0 = str = strdup(g->p.node_list_str); + t = 0; + + BUG_ON(!str); + + tprintf("# binding tasks to NODEs:\n"); + tprintf("# "); + + while (true) { + int bind_node, bind_node_0, bind_node_1; + char *tok, *tok_end, *tok_step, *tok_mul; + int step; + int mul; + + tok = strsep(&str, ","); + if (!tok) + break; + + tok_end = strstr(tok, "-"); + + dprintf("\ntoken: {%s}, end: {%s}\n", tok, tok_end); + if (!tok_end) { + /* Single NODE specified: */ + bind_node_0 = bind_node_1 = atol(tok); + } else { + /* NODE range specified (for example: "5-11"): */ + bind_node_0 = atol(tok); + bind_node_1 = atol(tok_end + 1); + } + + step = 1; + tok_step = strstr(tok, "#"); + if (tok_step) { + step = atol(tok_step + 1); + BUG_ON(step <= 0 || step >= g->p.nr_nodes); + } + + /* Multiplicator shortcut, "0x8" is a shortcut for: "0,0,0,0,0,0,0,0" */ + mul = 1; + tok_mul = strstr(tok, "x"); + if (tok_mul) { + mul = atol(tok_mul + 1); + BUG_ON(mul <= 0); + } + + dprintf("NODEs: %d-%d #%d\n", bind_node_0, bind_node_1, step); + + BUG_ON(bind_node_0 < 0 || bind_node_0 >= g->p.nr_nodes); + BUG_ON(bind_node_1 < 0 || bind_node_1 >= g->p.nr_nodes); + BUG_ON(bind_node_0 > bind_node_1); + + for (bind_node = bind_node_0; bind_node <= bind_node_1; bind_node += step) { + int i; + + for (i = 0; i < mul; i++) { + if (t >= g->p.nr_tasks) { + printf("\n# NOTE: ignoring bind NODEs starting at NODE#%d\n", bind_node); + goto out; + } + td = g->threads + t; + + if (!t) + tprintf(" %2d", bind_node); + else + tprintf(",%2d", bind_node); + + td->bind_node = bind_node; + t++; + } + } + } +out: + + tprintf("\n"); + + if (t < g->p.nr_tasks) + printf("# NOTE: %d tasks mem-bound, %d tasks unbound\n", t, g->p.nr_tasks - t); + + free(str0); +} + +static int parse_nodes_opt(const struct option *opt __maybe_unused, + const char *arg, int unset __maybe_unused) +{ + if (!arg) + return -1; + + return parse_node_list(arg); + + return 0; +} + +#define BIT(x) (1ul << x) + +static inline uint32_t lfsr_32(uint32_t lfsr) +{ + const uint32_t taps = BIT(1) | BIT(5) | BIT(6) | BIT(31); + return (lfsr>>1) ^ ((0x0u - (lfsr & 0x1u)) & taps); +} + +/* + * Make sure there's real data dependency to RAM (when read + * accesses are enabled), so the compiler, the CPU and the + * kernel (KSM, zero page, etc.) cannot optimize away RAM + * accesses: + */ +static inline u64 access_data(u64 *data __attribute__((unused)), u64 val) +{ + if (g->p.data_reads) + val += *data; + if (g->p.data_writes) + *data = val + 1; + return val; +} + +/* + * The worker process does two types of work, a forwards going + * loop and a backwards going loop. + * + * We do this so that on multiprocessor systems we do not create + * a 'train' of processing, with highly synchronized processes, + * skewing the whole benchmark. + */ +static u64 do_work(u8 *__data, long bytes, int nr, int nr_max, int loop, u64 val) +{ + long words = bytes/sizeof(u64); + u64 *data = (void *)__data; + long chunk_0, chunk_1; + u64 *d0, *d, *d1; + long off; + long i; + + BUG_ON(!data && words); + BUG_ON(data && !words); + + if (!data) + return val; + + /* Very simple memset() work variant: */ + if (g->p.data_zero_memset && !g->p.data_rand_walk) { + bzero(data, bytes); + return val; + } + + /* Spread out by PID/TID nr and by loop nr: */ + chunk_0 = words/nr_max; + chunk_1 = words/g->p.nr_loops; + off = nr*chunk_0 + loop*chunk_1; + + while (off >= words) + off -= words; + + if (g->p.data_rand_walk) { + u32 lfsr = nr + loop + val; + int j; + + for (i = 0; i < words/1024; i++) { + long start, end; + + lfsr = lfsr_32(lfsr); + + start = lfsr % words; + end = min(start + 1024, words-1); + + if (g->p.data_zero_memset) { + bzero(data + start, (end-start) * sizeof(u64)); + } else { + for (j = start; j < end; j++) + val = access_data(data + j, val); + } + } + } else if (!g->p.data_backwards || (nr + loop) & 1) { + + d0 = data + off; + d = data + off + 1; + d1 = data + words; + + /* Process data forwards: */ + for (;;) { + if (unlikely(d >= d1)) + d = data; + if (unlikely(d == d0)) + break; + + val = access_data(d, val); + + d++; + } + } else { + /* Process data backwards: */ + + d0 = data + off; + d = data + off - 1; + d1 = data + words; + + /* Process data forwards: */ + for (;;) { + if (unlikely(d < data)) + d = data + words-1; + if (unlikely(d == d0)) + break; + + val = access_data(d, val); + + d--; + } + } + + return val; +} + +static void update_curr_cpu(int task_nr, unsigned long bytes_worked) +{ + unsigned int cpu; + + cpu = sched_getcpu(); + + g->threads[task_nr].curr_cpu = cpu; + prctl(0, bytes_worked); +} + +#define MAX_NR_NODES 64 + +/* + * Count the number of nodes a process's threads + * are spread out on. + * + * A count of 1 means that the process is compressed + * to a single node. A count of g->p.nr_nodes means it's + * spread out on the whole system. + */ +static int count_process_nodes(int process_nr) +{ + char node_present[MAX_NR_NODES] = { 0, }; + int nodes; + int n, t; + + for (t = 0; t < g->p.nr_threads; t++) { + struct thread_data *td; + int task_nr; + int node; + + task_nr = process_nr*g->p.nr_threads + t; + td = g->threads + task_nr; + + node = numa_node_of_cpu(td->curr_cpu); + node_present[node] = 1; + } + + nodes = 0; + + for (n = 0; n < MAX_NR_NODES; n++) + nodes += node_present[n]; + + return nodes; +} + +/* + * Count the number of distinct process-threads a node contains. + * + * A count of 1 means that the node contains only a single + * process. If all nodes on the system contain at most one + * process then we are well-converged. + */ +static int count_node_processes(int node) +{ + int processes = 0; + int t, p; + + for (p = 0; p < g->p.nr_proc; p++) { + for (t = 0; t < g->p.nr_threads; t++) { + struct thread_data *td; + int task_nr; + int n; + + task_nr = p*g->p.nr_threads + t; + td = g->threads + task_nr; + + n = numa_node_of_cpu(td->curr_cpu); + if (n == node) { + processes++; + break; + } + } + } + + return processes; +} + +static void calc_convergence_compression(int *strong) +{ + unsigned int nodes_min, nodes_max; + int p; + + nodes_min = -1; + nodes_max = 0; + + for (p = 0; p < g->p.nr_proc; p++) { + unsigned int nodes = count_process_nodes(p); + + nodes_min = min(nodes, nodes_min); + nodes_max = max(nodes, nodes_max); + } + + /* Strong convergence: all threads compress on a single node: */ + if (nodes_min == 1 && nodes_max == 1) { + *strong = 1; + } else { + *strong = 0; + tprintf(" {%d-%d}", nodes_min, nodes_max); + } +} + +static void calc_convergence(double runtime_ns_max, double *convergence) +{ + unsigned int loops_done_min, loops_done_max; + int process_groups; + int nodes[MAX_NR_NODES]; + int distance; + int nr_min; + int nr_max; + int strong; + int sum; + int nr; + int node; + int cpu; + int t; + + if (!g->p.show_convergence && !g->p.measure_convergence) + return; + + for (node = 0; node < g->p.nr_nodes; node++) + nodes[node] = 0; + + loops_done_min = -1; + loops_done_max = 0; + + for (t = 0; t < g->p.nr_tasks; t++) { + struct thread_data *td = g->threads + t; + unsigned int loops_done; + + cpu = td->curr_cpu; + + /* Not all threads have written it yet: */ + if (cpu < 0) + continue; + + node = numa_node_of_cpu(cpu); + + nodes[node]++; + + loops_done = td->loops_done; + loops_done_min = min(loops_done, loops_done_min); + loops_done_max = max(loops_done, loops_done_max); + } + + nr_max = 0; + nr_min = g->p.nr_tasks; + sum = 0; + + for (node = 0; node < g->p.nr_nodes; node++) { + nr = nodes[node]; + nr_min = min(nr, nr_min); + nr_max = max(nr, nr_max); + sum += nr; + } + BUG_ON(nr_min > nr_max); + + BUG_ON(sum > g->p.nr_tasks); + + if (0 && (sum < g->p.nr_tasks)) + return; + + /* + * Count the number of distinct process groups present + * on nodes - when we are converged this will decrease + * to g->p.nr_proc: + */ + process_groups = 0; + + for (node = 0; node < g->p.nr_nodes; node++) { + int processes = count_node_processes(node); + + nr = nodes[node]; + tprintf(" %2d/%-2d", nr, processes); + + process_groups += processes; + } + + distance = nr_max - nr_min; + + tprintf(" [%2d/%-2d]", distance, process_groups); + + tprintf(" l:%3d-%-3d (%3d)", + loops_done_min, loops_done_max, loops_done_max-loops_done_min); + + if (loops_done_min && loops_done_max) { + double skew = 1.0 - (double)loops_done_min/loops_done_max; + + tprintf(" [%4.1f%%]", skew * 100.0); + } + + calc_convergence_compression(&strong); + + if (strong && process_groups == g->p.nr_proc) { + if (!*convergence) { + *convergence = runtime_ns_max; + tprintf(" (%6.1fs converged)\n", *convergence/1e9); + if (g->p.measure_convergence) { + g->all_converged = true; + g->stop_work = true; + } + } + } else { + if (*convergence) { + tprintf(" (%6.1fs de-converged)", runtime_ns_max/1e9); + *convergence = 0; + } + tprintf("\n"); + } +} + +static void show_summary(double runtime_ns_max, int l, double *convergence) +{ + tprintf("\r # %5.1f%% [%.1f mins]", + (double)(l+1)/g->p.nr_loops*100.0, runtime_ns_max/1e9 / 60.0); + + calc_convergence(runtime_ns_max, convergence); + + if (g->p.show_details >= 0) + fflush(stdout); +} + +static void *worker_thread(void *__tdata) +{ + struct thread_data *td = __tdata; + struct timeval start0, start, stop, diff; + int process_nr = td->process_nr; + int thread_nr = td->thread_nr; + unsigned long last_perturbance; + int task_nr = td->task_nr; + int details = g->p.show_details; + int first_task, last_task; + double convergence = 0; + u64 val = td->val; + double runtime_ns_max; + u8 *global_data; + u8 *process_data; + u8 *thread_data; + u64 bytes_done; + long work_done; + u32 l; + + bind_to_cpumask(td->bind_cpumask); + bind_to_memnode(td->bind_node); + + set_taskname("thread %d/%d", process_nr, thread_nr); + + global_data = g->data; + process_data = td->process_data; + thread_data = setup_private_data(g->p.bytes_thread); + + bytes_done = 0; + + last_task = 0; + if (process_nr == g->p.nr_proc-1 && thread_nr == g->p.nr_threads-1) + last_task = 1; + + first_task = 0; + if (process_nr == 0 && thread_nr == 0) + first_task = 1; + + if (details >= 2) { + printf("# thread %2d / %2d global mem: %p, process mem: %p, thread mem: %p\n", + process_nr, thread_nr, global_data, process_data, thread_data); + } + + if (g->p.serialize_startup) { + pthread_mutex_lock(&g->startup_mutex); + g->nr_tasks_started++; + pthread_mutex_unlock(&g->startup_mutex); + + /* Here we will wait for the main process to start us all at once: */ + pthread_mutex_lock(&g->start_work_mutex); + g->nr_tasks_working++; + + /* Last one wake the main process: */ + if (g->nr_tasks_working == g->p.nr_tasks) + pthread_mutex_unlock(&g->startup_done_mutex); + + pthread_mutex_unlock(&g->start_work_mutex); + } + + gettimeofday(&start0, NULL); + + start = stop = start0; + last_perturbance = start.tv_sec; + + for (l = 0; l < g->p.nr_loops; l++) { + start = stop; + + if (g->stop_work) + break; + + val += do_work(global_data, g->p.bytes_global, process_nr, g->p.nr_proc, l, val); + val += do_work(process_data, g->p.bytes_process, thread_nr, g->p.nr_threads, l, val); + val += do_work(thread_data, g->p.bytes_thread, 0, 1, l, val); + + if (g->p.sleep_usecs) { + pthread_mutex_lock(td->process_lock); + usleep(g->p.sleep_usecs); + pthread_mutex_unlock(td->process_lock); + } + /* + * Amount of work to be done under a process-global lock: + */ + if (g->p.bytes_process_locked) { + pthread_mutex_lock(td->process_lock); + val += do_work(process_data, g->p.bytes_process_locked, thread_nr, g->p.nr_threads, l, val); + pthread_mutex_unlock(td->process_lock); + } + + work_done = g->p.bytes_global + g->p.bytes_process + + g->p.bytes_process_locked + g->p.bytes_thread; + + update_curr_cpu(task_nr, work_done); + bytes_done += work_done; + + if (details < 0 && !g->p.perturb_secs && !g->p.measure_convergence && !g->p.nr_secs) + continue; + + td->loops_done = l; + + gettimeofday(&stop, NULL); + + /* Check whether our max runtime timed out: */ + if (g->p.nr_secs) { + timersub(&stop, &start0, &diff); + if (diff.tv_sec >= g->p.nr_secs) { + g->stop_work = true; + break; + } + } + + /* Update the summary at most once per second: */ + if (start.tv_sec == stop.tv_sec) + continue; + + /* + * Perturb the first task's equilibrium every g->p.perturb_secs seconds, + * by migrating to CPU#0: + */ + if (first_task && g->p.perturb_secs && (int)(stop.tv_sec - last_perturbance) >= g->p.perturb_secs) { + cpu_set_t orig_mask; + int target_cpu; + int this_cpu; + + last_perturbance = stop.tv_sec; + + /* + * Depending on where we are running, move into + * the other half of the system, to create some + * real disturbance: + */ + this_cpu = g->threads[task_nr].curr_cpu; + if (this_cpu < g->p.nr_cpus/2) + target_cpu = g->p.nr_cpus-1; + else + target_cpu = 0; + + orig_mask = bind_to_cpu(target_cpu); + + /* Here we are running on the target CPU already */ + if (details >= 1) + printf(" (injecting perturbalance, moved to CPU#%d)\n", target_cpu); + + bind_to_cpumask(orig_mask); + } + + if (details >= 3) { + timersub(&stop, &start, &diff); + runtime_ns_max = diff.tv_sec * 1000000000; + runtime_ns_max += diff.tv_usec * 1000; + + if (details >= 0) { + printf(" #%2d / %2d: %14.2lf nsecs/op [val: %016lx]\n", + process_nr, thread_nr, runtime_ns_max / bytes_done, val); + } + fflush(stdout); + } + if (!last_task) + continue; + + timersub(&stop, &start0, &diff); + runtime_ns_max = diff.tv_sec * 1000000000ULL; + runtime_ns_max += diff.tv_usec * 1000ULL; + + show_summary(runtime_ns_max, l, &convergence); + } + + gettimeofday(&stop, NULL); + timersub(&stop, &start0, &diff); + td->runtime_ns = diff.tv_sec * 1000000000ULL; + td->runtime_ns += diff.tv_usec * 1000ULL; + + free_data(thread_data, g->p.bytes_thread); + + pthread_mutex_lock(&g->stop_work_mutex); + g->bytes_done += bytes_done; + pthread_mutex_unlock(&g->stop_work_mutex); + + return NULL; +} + +/* + * A worker process starts a couple of threads: + */ +static void worker_process(int process_nr) +{ + pthread_mutex_t process_lock; + struct thread_data *td; + pthread_t *pthreads; + u8 *process_data; + int task_nr; + int ret; + int t; + + pthread_mutex_init(&process_lock, NULL); + set_taskname("process %d", process_nr); + + /* + * Pick up the memory policy and the CPU binding of our first thread, + * so that we initialize memory accordingly: + */ + task_nr = process_nr*g->p.nr_threads; + td = g->threads + task_nr; + + bind_to_memnode(td->bind_node); + bind_to_cpumask(td->bind_cpumask); + + pthreads = zalloc(g->p.nr_threads * sizeof(pthread_t)); + process_data = setup_private_data(g->p.bytes_process); + + if (g->p.show_details >= 3) { + printf(" # process %2d global mem: %p, process mem: %p\n", + process_nr, g->data, process_data); + } + + for (t = 0; t < g->p.nr_threads; t++) { + task_nr = process_nr*g->p.nr_threads + t; + td = g->threads + task_nr; + + td->process_data = process_data; + td->process_nr = process_nr; + td->thread_nr = t; + td->task_nr = task_nr; + td->val = rand(); + td->curr_cpu = -1; + td->process_lock = &process_lock; + + ret = pthread_create(pthreads + t, NULL, worker_thread, td); + BUG_ON(ret); + } + + for (t = 0; t < g->p.nr_threads; t++) { + ret = pthread_join(pthreads[t], NULL); + BUG_ON(ret); + } + + free_data(process_data, g->p.bytes_process); + free(pthreads); +} + +static void print_summary(void) +{ + if (g->p.show_details < 0) + return; + + printf("\n ###\n"); + printf(" # %d %s will execute (on %d nodes, %d CPUs):\n", + g->p.nr_tasks, g->p.nr_tasks == 1 ? "task" : "tasks", g->p.nr_nodes, g->p.nr_cpus); + printf(" # %5dx %5ldMB global shared mem operations\n", + g->p.nr_loops, g->p.bytes_global/1024/1024); + printf(" # %5dx %5ldMB process shared mem operations\n", + g->p.nr_loops, g->p.bytes_process/1024/1024); + printf(" # %5dx %5ldMB thread local mem operations\n", + g->p.nr_loops, g->p.bytes_thread/1024/1024); + + printf(" ###\n"); + + printf("\n ###\n"); fflush(stdout); +} + +static void init_thread_data(void) +{ + ssize_t size = sizeof(*g->threads)*g->p.nr_tasks; + int t; + + g->threads = zalloc_shared_data(size); + + for (t = 0; t < g->p.nr_tasks; t++) { + struct thread_data *td = g->threads + t; + int cpu; + + /* Allow all nodes by default: */ + td->bind_node = -1; + + /* Allow all CPUs by default: */ + CPU_ZERO(&td->bind_cpumask); + for (cpu = 0; cpu < g->p.nr_cpus; cpu++) + CPU_SET(cpu, &td->bind_cpumask); + } +} + +static void deinit_thread_data(void) +{ + ssize_t size = sizeof(*g->threads)*g->p.nr_tasks; + + free_data(g->threads, size); +} + +static int init(void) +{ + g = (void *)alloc_data(sizeof(*g), MAP_SHARED, 1, 0, 0 /* THP */, 0); + + /* Copy over options: */ + g->p = p0; + + g->p.nr_cpus = numa_num_configured_cpus(); + + g->p.nr_nodes = numa_max_node() + 1; + + /* char array in count_process_nodes(): */ + BUG_ON(g->p.nr_nodes > MAX_NR_NODES || g->p.nr_nodes < 0); + + if (g->p.show_quiet && !g->p.show_details) + g->p.show_details = -1; + + /* Some memory should be specified: */ + if (!g->p.mb_global_str && !g->p.mb_proc_str && !g->p.mb_thread_str) + return -1; + + if (g->p.mb_global_str) { + g->p.mb_global = atof(g->p.mb_global_str); + BUG_ON(g->p.mb_global < 0); + } + + if (g->p.mb_proc_str) { + g->p.mb_proc = atof(g->p.mb_proc_str); + BUG_ON(g->p.mb_proc < 0); + } + + if (g->p.mb_proc_locked_str) { + g->p.mb_proc_locked = atof(g->p.mb_proc_locked_str); + BUG_ON(g->p.mb_proc_locked < 0); + BUG_ON(g->p.mb_proc_locked > g->p.mb_proc); + } + + if (g->p.mb_thread_str) { + g->p.mb_thread = atof(g->p.mb_thread_str); + BUG_ON(g->p.mb_thread < 0); + } + + BUG_ON(g->p.nr_threads <= 0); + BUG_ON(g->p.nr_proc <= 0); + + g->p.nr_tasks = g->p.nr_proc*g->p.nr_threads; + + g->p.bytes_global = g->p.mb_global *1024L*1024L; + g->p.bytes_process = g->p.mb_proc *1024L*1024L; + g->p.bytes_process_locked = g->p.mb_proc_locked *1024L*1024L; + g->p.bytes_thread = g->p.mb_thread *1024L*1024L; + + g->data = setup_shared_data(g->p.bytes_global); + + /* Startup serialization: */ + init_global_mutex(&g->start_work_mutex); + init_global_mutex(&g->startup_mutex); + init_global_mutex(&g->startup_done_mutex); + init_global_mutex(&g->stop_work_mutex); + + init_thread_data(); + + tprintf("#\n"); + parse_setup_cpu_list(); + parse_setup_node_list(); + tprintf("#\n"); + + print_summary(); + + return 0; +} + +static void deinit(void) +{ + free_data(g->data, g->p.bytes_global); + g->data = NULL; + + deinit_thread_data(); + + free_data(g, sizeof(*g)); + g = NULL; +} + +/* + * Print a short or long result, depending on the verbosity setting: + */ +static void print_res(const char *name, double val, + const char *txt_unit, const char *txt_short, const char *txt_long) +{ + if (!name) + name = "main,"; + + if (g->p.show_quiet) + printf(" %-30s %15.3f, %-15s %s\n", name, val, txt_unit, txt_short); + else + printf(" %14.3f %s\n", val, txt_long); +} + +static int __bench_numa(const char *name) +{ + struct timeval start, stop, diff; + u64 runtime_ns_min, runtime_ns_sum; + pid_t *pids, pid, wpid; + double delta_runtime; + double runtime_avg; + double runtime_sec_max; + double runtime_sec_min; + int wait_stat; + double bytes; + int i, t; + + if (init()) + return -1; + + pids = zalloc(g->p.nr_proc * sizeof(*pids)); + pid = -1; + + /* All threads try to acquire it, this way we can wait for them to start up: */ + pthread_mutex_lock(&g->start_work_mutex); + + if (g->p.serialize_startup) { + tprintf(" #\n"); + tprintf(" # Startup synchronization: ..."); fflush(stdout); + } + + gettimeofday(&start, NULL); + + for (i = 0; i < g->p.nr_proc; i++) { + pid = fork(); + dprintf(" # process %2d: PID %d\n", i, pid); + + BUG_ON(pid < 0); + if (!pid) { + /* Child process: */ + worker_process(i); + + exit(0); + } + pids[i] = pid; + + } + /* Wait for all the threads to start up: */ + while (g->nr_tasks_started != g->p.nr_tasks) + usleep(1000); + + BUG_ON(g->nr_tasks_started != g->p.nr_tasks); + + if (g->p.serialize_startup) { + double startup_sec; + + pthread_mutex_lock(&g->startup_done_mutex); + + /* This will start all threads: */ + pthread_mutex_unlock(&g->start_work_mutex); + + /* This mutex is locked - the last started thread will wake us: */ + pthread_mutex_lock(&g->startup_done_mutex); + + gettimeofday(&stop, NULL); + + timersub(&stop, &start, &diff); + + startup_sec = diff.tv_sec * 1000000000.0; + startup_sec += diff.tv_usec * 1000.0; + startup_sec /= 1e9; + + tprintf(" threads initialized in %.6f seconds.\n", startup_sec); + tprintf(" #\n"); + + start = stop; + pthread_mutex_unlock(&g->startup_done_mutex); + } else { + gettimeofday(&start, NULL); + } + + /* Parent process: */ + + + for (i = 0; i < g->p.nr_proc; i++) { + wpid = waitpid(pids[i], &wait_stat, 0); + BUG_ON(wpid < 0); + BUG_ON(!WIFEXITED(wait_stat)); + + } + + runtime_ns_sum = 0; + runtime_ns_min = -1LL; + + for (t = 0; t < g->p.nr_tasks; t++) { + u64 thread_runtime_ns = g->threads[t].runtime_ns; + + runtime_ns_sum += thread_runtime_ns; + runtime_ns_min = min(thread_runtime_ns, runtime_ns_min); + } + + gettimeofday(&stop, NULL); + timersub(&stop, &start, &diff); + + BUG_ON(bench_format != BENCH_FORMAT_DEFAULT); + + tprintf("\n ###\n"); + tprintf("\n"); + + runtime_sec_max = diff.tv_sec * 1000000000.0; + runtime_sec_max += diff.tv_usec * 1000.0; + runtime_sec_max /= 1e9; + + runtime_sec_min = runtime_ns_min/1e9; + + bytes = g->bytes_done; + runtime_avg = (double)runtime_ns_sum / g->p.nr_tasks / 1e9; + + if (g->p.measure_convergence) { + print_res(name, runtime_sec_max, + "secs,", "NUMA-convergence-latency", "secs latency to NUMA-converge"); + } + + print_res(name, runtime_sec_max, + "secs,", "runtime-max/thread", "secs slowest (max) thread-runtime"); + + print_res(name, runtime_sec_min, + "secs,", "runtime-min/thread", "secs fastest (min) thread-runtime"); + + print_res(name, runtime_avg, + "secs,", "runtime-avg/thread", "secs average thread-runtime"); + + delta_runtime = (runtime_sec_max - runtime_sec_min)/2.0; + print_res(name, delta_runtime / runtime_sec_max * 100.0, + "%,", "spread-runtime/thread", "% difference between max/avg runtime"); + + print_res(name, bytes / g->p.nr_tasks / 1e9, + "GB,", "data/thread", "GB data processed, per thread"); + + print_res(name, bytes / 1e9, + "GB,", "data-total", "GB data processed, total"); + + print_res(name, runtime_sec_max * 1e9 / (bytes / g->p.nr_tasks), + "nsecs,", "runtime/byte/thread","nsecs/byte/thread runtime"); + + print_res(name, bytes / g->p.nr_tasks / 1e9 / runtime_sec_max, + "GB/sec,", "thread-speed", "GB/sec/thread speed"); + + print_res(name, bytes / runtime_sec_max / 1e9, + "GB/sec,", "total-speed", "GB/sec total speed"); + + free(pids); + + deinit(); + + return 0; +} + +#define MAX_ARGS 50 + +static int command_size(const char **argv) +{ + int size = 0; + + while (*argv) { + size++; + argv++; + } + + BUG_ON(size >= MAX_ARGS); + + return size; +} + +static void init_params(struct params *p, const char *name, int argc, const char **argv) +{ + int i; + + printf("\n # Running %s \"perf bench numa", name); + + for (i = 0; i < argc; i++) + printf(" %s", argv[i]); + + printf("\"\n"); + + memset(p, 0, sizeof(*p)); + + /* Initialize nonzero defaults: */ + + p->serialize_startup = 1; + p->data_reads = true; + p->data_writes = true; + p->data_backwards = true; + p->data_rand_walk = true; + p->nr_loops = -1; + p->init_random = true; +} + +static int run_bench_numa(const char *name, const char **argv) +{ + int argc = command_size(argv); + + init_params(&p0, name, argc, argv); + argc = parse_options(argc, argv, options, bench_numa_usage, 0); + if (argc) + goto err; + + if (__bench_numa(name)) + goto err; + + return 0; + +err: + usage_with_options(numa_usage, options); + return -1; +} + +#define OPT_BW_RAM "-s", "20", "-zZq", "--thp", " 1", "--no-data_rand_walk" +#define OPT_BW_RAM_NOTHP OPT_BW_RAM, "--thp", "-1" + +#define OPT_CONV "-s", "100", "-zZ0qcm", "--thp", " 1" +#define OPT_CONV_NOTHP OPT_CONV, "--thp", "-1" + +#define OPT_BW "-s", "20", "-zZ0q", "--thp", " 1" +#define OPT_BW_NOTHP OPT_BW, "--thp", "-1" + +/* + * The built-in test-suite executed by "perf bench numa -a". + * + * (A minimum of 4 nodes and 16 GB of RAM is recommended.) + */ +static const char *tests[][MAX_ARGS] = { + /* Basic single-stream NUMA bandwidth measurements: */ + { "RAM-bw-local,", "mem", "-p", "1", "-t", "1", "-P", "1024", + "-C" , "0", "-M", "0", OPT_BW_RAM }, + { "RAM-bw-local-NOTHP,", + "mem", "-p", "1", "-t", "1", "-P", "1024", + "-C" , "0", "-M", "0", OPT_BW_RAM_NOTHP }, + { "RAM-bw-remote,", "mem", "-p", "1", "-t", "1", "-P", "1024", + "-C" , "0", "-M", "1", OPT_BW_RAM }, + + /* 2-stream NUMA bandwidth measurements: */ + { "RAM-bw-local-2x,", "mem", "-p", "2", "-t", "1", "-P", "1024", + "-C", "0,2", "-M", "0x2", OPT_BW_RAM }, + { "RAM-bw-remote-2x,", "mem", "-p", "2", "-t", "1", "-P", "1024", + "-C", "0,2", "-M", "1x2", OPT_BW_RAM }, + + /* Cross-stream NUMA bandwidth measurement: */ + { "RAM-bw-cross,", "mem", "-p", "2", "-t", "1", "-P", "1024", + "-C", "0,8", "-M", "1,0", OPT_BW_RAM }, + + /* Convergence latency measurements: */ + { " 1x3-convergence,", "mem", "-p", "1", "-t", "3", "-P", "512", OPT_CONV }, + { " 1x4-convergence,", "mem", "-p", "1", "-t", "4", "-P", "512", OPT_CONV }, + { " 1x6-convergence,", "mem", "-p", "1", "-t", "6", "-P", "1020", OPT_CONV }, + { " 2x3-convergence,", "mem", "-p", "3", "-t", "3", "-P", "1020", OPT_CONV }, + { " 3x3-convergence,", "mem", "-p", "3", "-t", "3", "-P", "1020", OPT_CONV }, + { " 4x4-convergence,", "mem", "-p", "4", "-t", "4", "-P", "512", OPT_CONV }, + { " 4x4-convergence-NOTHP,", + "mem", "-p", "4", "-t", "4", "-P", "512", OPT_CONV_NOTHP }, + { " 4x6-convergence,", "mem", "-p", "4", "-t", "6", "-P", "1020", OPT_CONV }, + { " 4x8-convergence,", "mem", "-p", "4", "-t", "8", "-P", "512", OPT_CONV }, + { " 8x4-convergence,", "mem", "-p", "8", "-t", "4", "-P", "512", OPT_CONV }, + { " 8x4-convergence-NOTHP,", + "mem", "-p", "8", "-t", "4", "-P", "512", OPT_CONV_NOTHP }, + { " 3x1-convergence,", "mem", "-p", "3", "-t", "1", "-P", "512", OPT_CONV }, + { " 4x1-convergence,", "mem", "-p", "4", "-t", "1", "-P", "512", OPT_CONV }, + { " 8x1-convergence,", "mem", "-p", "8", "-t", "1", "-P", "512", OPT_CONV }, + { "16x1-convergence,", "mem", "-p", "16", "-t", "1", "-P", "256", OPT_CONV }, + { "32x1-convergence,", "mem", "-p", "32", "-t", "1", "-P", "128", OPT_CONV }, + + /* Various NUMA process/thread layout bandwidth measurements: */ + { " 2x1-bw-process,", "mem", "-p", "2", "-t", "1", "-P", "1024", OPT_BW }, + { " 3x1-bw-process,", "mem", "-p", "3", "-t", "1", "-P", "1024", OPT_BW }, + { " 4x1-bw-process,", "mem", "-p", "4", "-t", "1", "-P", "1024", OPT_BW }, + { " 8x1-bw-process,", "mem", "-p", "8", "-t", "1", "-P", " 512", OPT_BW }, + { " 8x1-bw-process-NOTHP,", + "mem", "-p", "8", "-t", "1", "-P", " 512", OPT_BW_NOTHP }, + { "16x1-bw-process,", "mem", "-p", "16", "-t", "1", "-P", "256", OPT_BW }, + + { " 4x1-bw-thread,", "mem", "-p", "1", "-t", "4", "-T", "256", OPT_BW }, + { " 8x1-bw-thread,", "mem", "-p", "1", "-t", "8", "-T", "256", OPT_BW }, + { "16x1-bw-thread,", "mem", "-p", "1", "-t", "16", "-T", "128", OPT_BW }, + { "32x1-bw-thread,", "mem", "-p", "1", "-t", "32", "-T", "64", OPT_BW }, + + { " 2x3-bw-thread,", "mem", "-p", "2", "-t", "3", "-P", "512", OPT_BW }, + { " 4x4-bw-thread,", "mem", "-p", "4", "-t", "4", "-P", "512", OPT_BW }, + { " 4x6-bw-thread,", "mem", "-p", "4", "-t", "6", "-P", "512", OPT_BW }, + { " 4x8-bw-thread,", "mem", "-p", "4", "-t", "8", "-P", "512", OPT_BW }, + { " 4x8-bw-thread-NOTHP,", + "mem", "-p", "4", "-t", "8", "-P", "512", OPT_BW_NOTHP }, + { " 3x3-bw-thread,", "mem", "-p", "3", "-t", "3", "-P", "512", OPT_BW }, + { " 5x5-bw-thread,", "mem", "-p", "5", "-t", "5", "-P", "512", OPT_BW }, + + { "2x16-bw-thread,", "mem", "-p", "2", "-t", "16", "-P", "512", OPT_BW }, + { "1x32-bw-thread,", "mem", "-p", "1", "-t", "32", "-P", "2048", OPT_BW }, + + { "numa02-bw,", "mem", "-p", "1", "-t", "32", "-T", "32", OPT_BW }, + { "numa02-bw-NOTHP,", "mem", "-p", "1", "-t", "32", "-T", "32", OPT_BW_NOTHP }, + { "numa01-bw-thread,", "mem", "-p", "2", "-t", "16", "-T", "192", OPT_BW }, + { "numa01-bw-thread-NOTHP,", + "mem", "-p", "2", "-t", "16", "-T", "192", OPT_BW_NOTHP }, +}; + +static int bench_all(void) +{ + int nr = ARRAY_SIZE(tests); + int ret; + int i; + + ret = system("echo ' #'; echo ' # Running test on: '$(uname -a); echo ' #'"); + BUG_ON(ret < 0); + + for (i = 0; i < nr; i++) { + if (run_bench_numa(tests[i][0], tests[i] + 1)) + return -1; + } + + printf("\n"); + + return 0; +} + +int bench_numa(int argc, const char **argv, const char *prefix __maybe_unused) +{ + init_params(&p0, "main,", argc, argv); + argc = parse_options(argc, argv, options, bench_numa_usage, 0); + if (argc) + goto err; + + if (p0.run_all) + return bench_all(); + + if (__bench_numa(NULL)) + goto err; + + return 0; + +err: + usage_with_options(numa_usage, options); + return -1; +} diff --git a/tools/perf/bench/sched-messaging.c b/tools/perf/bench/sched-messaging.c new file mode 100644 index 00000000..cc1190a0 --- /dev/null +++ b/tools/perf/bench/sched-messaging.c @@ -0,0 +1,336 @@ +/* + * + * sched-messaging.c + * + * messaging: Benchmark for scheduler and IPC mechanisms + * + * Based on hackbench by Rusty Russell <rusty@rustcorp.com.au> + * Ported to perf by Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp> + * + */ + +#include "../perf.h" +#include "../util/util.h" +#include "../util/parse-options.h" +#include "../builtin.h" +#include "bench.h" + +/* Test groups of 20 processes spraying to 20 receivers */ +#include <pthread.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <errno.h> +#include <unistd.h> +#include <sys/types.h> +#include <sys/socket.h> +#include <sys/wait.h> +#include <sys/time.h> +#include <sys/poll.h> +#include <limits.h> + +#define DATASIZE 100 + +static bool use_pipes = false; +static unsigned int loops = 100; +static bool thread_mode = false; +static unsigned int num_groups = 10; + +struct sender_context { + unsigned int num_fds; + int ready_out; + int wakefd; + int out_fds[0]; +}; + +struct receiver_context { + unsigned int num_packets; + int in_fds[2]; + int ready_out; + int wakefd; +}; + +static void barf(const char *msg) +{ + fprintf(stderr, "%s (error: %s)\n", msg, strerror(errno)); + exit(1); +} + +static void fdpair(int fds[2]) +{ + if (use_pipes) { + if (pipe(fds) == 0) + return; + } else { + if (socketpair(AF_UNIX, SOCK_STREAM, 0, fds) == 0) + return; + } + + barf(use_pipes ? "pipe()" : "socketpair()"); +} + +/* Block until we're ready to go */ +static void ready(int ready_out, int wakefd) +{ + char dummy; + struct pollfd pollfd = { .fd = wakefd, .events = POLLIN }; + + /* Tell them we're ready. */ + if (write(ready_out, &dummy, 1) != 1) + barf("CLIENT: ready write"); + + /* Wait for "GO" signal */ + if (poll(&pollfd, 1, -1) != 1) + barf("poll"); +} + +/* Sender sprays loops messages down each file descriptor */ +static void *sender(struct sender_context *ctx) +{ + char data[DATASIZE]; + unsigned int i, j; + + ready(ctx->ready_out, ctx->wakefd); + + /* Now pump to every receiver. */ + for (i = 0; i < loops; i++) { + for (j = 0; j < ctx->num_fds; j++) { + int ret, done = 0; + +again: + ret = write(ctx->out_fds[j], data + done, + sizeof(data)-done); + if (ret < 0) + barf("SENDER: write"); + done += ret; + if (done < DATASIZE) + goto again; + } + } + + return NULL; +} + + +/* One receiver per fd */ +static void *receiver(struct receiver_context* ctx) +{ + unsigned int i; + + if (!thread_mode) + close(ctx->in_fds[1]); + + /* Wait for start... */ + ready(ctx->ready_out, ctx->wakefd); + + /* Receive them all */ + for (i = 0; i < ctx->num_packets; i++) { + char data[DATASIZE]; + int ret, done = 0; + +again: + ret = read(ctx->in_fds[0], data + done, DATASIZE - done); + if (ret < 0) + barf("SERVER: read"); + done += ret; + if (done < DATASIZE) + goto again; + } + + return NULL; +} + +static pthread_t create_worker(void *ctx, void *(*func)(void *)) +{ + pthread_attr_t attr; + pthread_t childid; + int err; + + if (!thread_mode) { + /* process mode */ + /* Fork the receiver. */ + switch (fork()) { + case -1: + barf("fork()"); + break; + case 0: + (*func) (ctx); + exit(0); + break; + default: + break; + } + + return (pthread_t)0; + } + + if (pthread_attr_init(&attr) != 0) + barf("pthread_attr_init:"); + +#ifndef __ia64__ + if (pthread_attr_setstacksize(&attr, PTHREAD_STACK_MIN) != 0) + barf("pthread_attr_setstacksize"); +#endif + + err = pthread_create(&childid, &attr, func, ctx); + if (err != 0) { + fprintf(stderr, "pthread_create failed: %s (%d)\n", + strerror(err), err); + exit(-1); + } + return childid; +} + +static void reap_worker(pthread_t id) +{ + int proc_status; + void *thread_status; + + if (!thread_mode) { + /* process mode */ + wait(&proc_status); + if (!WIFEXITED(proc_status)) + exit(1); + } else { + pthread_join(id, &thread_status); + } +} + +/* One group of senders and receivers */ +static unsigned int group(pthread_t *pth, + unsigned int num_fds, + int ready_out, + int wakefd) +{ + unsigned int i; + struct sender_context *snd_ctx = malloc(sizeof(struct sender_context) + + num_fds * sizeof(int)); + + if (!snd_ctx) + barf("malloc()"); + + for (i = 0; i < num_fds; i++) { + int fds[2]; + struct receiver_context *ctx = malloc(sizeof(*ctx)); + + if (!ctx) + barf("malloc()"); + + + /* Create the pipe between client and server */ + fdpair(fds); + + ctx->num_packets = num_fds * loops; + ctx->in_fds[0] = fds[0]; + ctx->in_fds[1] = fds[1]; + ctx->ready_out = ready_out; + ctx->wakefd = wakefd; + + pth[i] = create_worker(ctx, (void *)receiver); + + snd_ctx->out_fds[i] = fds[1]; + if (!thread_mode) + close(fds[0]); + } + + /* Now we have all the fds, fork the senders */ + for (i = 0; i < num_fds; i++) { + snd_ctx->ready_out = ready_out; + snd_ctx->wakefd = wakefd; + snd_ctx->num_fds = num_fds; + + pth[num_fds+i] = create_worker(snd_ctx, (void *)sender); + } + + /* Close the fds we have left */ + if (!thread_mode) + for (i = 0; i < num_fds; i++) + close(snd_ctx->out_fds[i]); + + /* Return number of children to reap */ + return num_fds * 2; +} + +static const struct option options[] = { + OPT_BOOLEAN('p', "pipe", &use_pipes, + "Use pipe() instead of socketpair()"), + OPT_BOOLEAN('t', "thread", &thread_mode, + "Be multi thread instead of multi process"), + OPT_UINTEGER('g', "group", &num_groups, "Specify number of groups"), + OPT_UINTEGER('l', "loop", &loops, "Specify number of loops"), + OPT_END() +}; + +static const char * const bench_sched_message_usage[] = { + "perf bench sched messaging <options>", + NULL +}; + +int bench_sched_messaging(int argc, const char **argv, + const char *prefix __maybe_unused) +{ + unsigned int i, total_children; + struct timeval start, stop, diff; + unsigned int num_fds = 20; + int readyfds[2], wakefds[2]; + char dummy; + pthread_t *pth_tab; + + argc = parse_options(argc, argv, options, + bench_sched_message_usage, 0); + + pth_tab = malloc(num_fds * 2 * num_groups * sizeof(pthread_t)); + if (!pth_tab) + barf("main:malloc()"); + + fdpair(readyfds); + fdpair(wakefds); + + total_children = 0; + for (i = 0; i < num_groups; i++) + total_children += group(pth_tab+total_children, num_fds, + readyfds[1], wakefds[0]); + + /* Wait for everyone to be ready */ + for (i = 0; i < total_children; i++) + if (read(readyfds[0], &dummy, 1) != 1) + barf("Reading for readyfds"); + + gettimeofday(&start, NULL); + + /* Kick them off */ + if (write(wakefds[1], &dummy, 1) != 1) + barf("Writing to start them"); + + /* Reap them all */ + for (i = 0; i < total_children; i++) + reap_worker(pth_tab[i]); + + gettimeofday(&stop, NULL); + + timersub(&stop, &start, &diff); + + switch (bench_format) { + case BENCH_FORMAT_DEFAULT: + printf("# %d sender and receiver %s per group\n", + num_fds, thread_mode ? "threads" : "processes"); + printf("# %d groups == %d %s run\n\n", + num_groups, num_groups * 2 * num_fds, + thread_mode ? "threads" : "processes"); + printf(" %14s: %lu.%03lu [sec]\n", "Total time", + diff.tv_sec, + (unsigned long) (diff.tv_usec/1000)); + break; + case BENCH_FORMAT_SIMPLE: + printf("%lu.%03lu\n", diff.tv_sec, + (unsigned long) (diff.tv_usec/1000)); + break; + default: + /* reaching here is something disaster */ + fprintf(stderr, "Unknown format:%d\n", bench_format); + exit(1); + break; + } + + return 0; +} diff --git a/tools/perf/bench/sched-pipe.c b/tools/perf/bench/sched-pipe.c new file mode 100644 index 00000000..69cfba8d --- /dev/null +++ b/tools/perf/bench/sched-pipe.c @@ -0,0 +1,127 @@ +/* + * + * sched-pipe.c + * + * pipe: Benchmark for pipe() + * + * Based on pipe-test-1m.c by Ingo Molnar <mingo@redhat.com> + * http://people.redhat.com/mingo/cfs-scheduler/tools/pipe-test-1m.c + * Ported to perf by Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp> + * + */ + +#include "../perf.h" +#include "../util/util.h" +#include "../util/parse-options.h" +#include "../builtin.h" +#include "bench.h" + +#include <unistd.h> +#include <stdio.h> +#include <stdlib.h> +#include <signal.h> +#include <sys/wait.h> +#include <linux/unistd.h> +#include <string.h> +#include <errno.h> +#include <assert.h> +#include <sys/time.h> +#include <sys/types.h> + +#define LOOPS_DEFAULT 1000000 +static int loops = LOOPS_DEFAULT; + +static const struct option options[] = { + OPT_INTEGER('l', "loop", &loops, + "Specify number of loops"), + OPT_END() +}; + +static const char * const bench_sched_pipe_usage[] = { + "perf bench sched pipe <options>", + NULL +}; + +int bench_sched_pipe(int argc, const char **argv, + const char *prefix __maybe_unused) +{ + int pipe_1[2], pipe_2[2]; + int m = 0, i; + struct timeval start, stop, diff; + unsigned long long result_usec = 0; + + /* + * why does "ret" exist? + * discarding returned value of read(), write() + * causes error in building environment for perf + */ + int __maybe_unused ret, wait_stat; + pid_t pid, retpid __maybe_unused; + + argc = parse_options(argc, argv, options, + bench_sched_pipe_usage, 0); + + BUG_ON(pipe(pipe_1)); + BUG_ON(pipe(pipe_2)); + + pid = fork(); + assert(pid >= 0); + + gettimeofday(&start, NULL); + + if (!pid) { + for (i = 0; i < loops; i++) { + ret = read(pipe_1[0], &m, sizeof(int)); + ret = write(pipe_2[1], &m, sizeof(int)); + } + } else { + for (i = 0; i < loops; i++) { + ret = write(pipe_1[1], &m, sizeof(int)); + ret = read(pipe_2[0], &m, sizeof(int)); + } + } + + gettimeofday(&stop, NULL); + timersub(&stop, &start, &diff); + + if (pid) { + retpid = waitpid(pid, &wait_stat, 0); + assert((retpid == pid) && WIFEXITED(wait_stat)); + } else { + exit(0); + } + + switch (bench_format) { + case BENCH_FORMAT_DEFAULT: + printf("# Executed %d pipe operations between two tasks\n\n", + loops); + + result_usec = diff.tv_sec * 1000000; + result_usec += diff.tv_usec; + + printf(" %14s: %lu.%03lu [sec]\n\n", "Total time", + diff.tv_sec, + (unsigned long) (diff.tv_usec/1000)); + + printf(" %14lf usecs/op\n", + (double)result_usec / (double)loops); + printf(" %14d ops/sec\n", + (int)((double)loops / + ((double)result_usec / (double)1000000))); + break; + + case BENCH_FORMAT_SIMPLE: + printf("%lu.%03lu\n", + diff.tv_sec, + (unsigned long) (diff.tv_usec / 1000)); + break; + + default: + /* reaching here is something disaster */ + fprintf(stderr, "Unknown format:%d\n", bench_format); + exit(1); + break; + } + + return 0; +} diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c new file mode 100644 index 00000000..2e6961ea --- /dev/null +++ b/tools/perf/builtin-annotate.c @@ -0,0 +1,350 @@ +/* + * 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 <linux/list.h> +#include "util/cache.h" +#include <linux/rbtree.h> +#include "util/symbol.h" + +#include "perf.h" +#include "util/debug.h" + +#include "util/evlist.h" +#include "util/evsel.h" +#include "util/annotate.h" +#include "util/event.h" +#include "util/parse-options.h" +#include "util/parse-events.h" +#include "util/thread.h" +#include "util/sort.h" +#include "util/hist.h" +#include "util/session.h" +#include "util/tool.h" +#include "arch/common.h" + +#include <linux/bitmap.h> + +struct perf_annotate { + struct perf_tool tool; + bool force, use_tui, use_stdio, use_gtk; + bool full_paths; + bool print_line; + bool skip_missing; + const char *sym_hist_filter; + const char *cpu_list; + DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS); +}; + +static int perf_evsel__add_sample(struct perf_evsel *evsel, + struct perf_sample *sample, + struct addr_location *al, + struct perf_annotate *ann) +{ + struct hist_entry *he; + int ret; + + if (ann->sym_hist_filter != NULL && + (al->sym == NULL || + strcmp(ann->sym_hist_filter, al->sym->name) != 0)) { + /* We're only interested in a symbol named sym_hist_filter */ + if (al->sym != NULL) { + rb_erase(&al->sym->rb_node, + &al->map->dso->symbols[al->map->type]); + symbol__delete(al->sym); + } + return 0; + } + + he = __hists__add_entry(&evsel->hists, al, NULL, 1); + if (he == NULL) + return -ENOMEM; + + ret = 0; + if (he->ms.sym != NULL) { + struct annotation *notes = symbol__annotation(he->ms.sym); + if (notes->src == NULL && symbol__alloc_hist(he->ms.sym) < 0) + return -ENOMEM; + + ret = hist_entry__inc_addr_samples(he, evsel->idx, al->addr); + } + + evsel->hists.stats.total_period += sample->period; + hists__inc_nr_events(&evsel->hists, PERF_RECORD_SAMPLE); + return ret; +} + +static int process_sample_event(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct perf_annotate *ann = container_of(tool, struct perf_annotate, tool); + struct addr_location al; + + if (perf_event__preprocess_sample(event, machine, &al, sample, + symbol__annotate_init) < 0) { + pr_warning("problem processing %d event, skipping it.\n", + event->header.type); + return -1; + } + + if (ann->cpu_list && !test_bit(sample->cpu, ann->cpu_bitmap)) + return 0; + + if (!al.filtered && perf_evsel__add_sample(evsel, sample, &al, ann)) { + pr_warning("problem incrementing symbol count, " + "skipping event\n"); + return -1; + } + + return 0; +} + +static int hist_entry__tty_annotate(struct hist_entry *he, int evidx, + struct perf_annotate *ann) +{ + return symbol__tty_annotate(he->ms.sym, he->ms.map, evidx, + ann->print_line, ann->full_paths, 0, 0); +} + +static void hists__find_annotations(struct hists *self, int evidx, + struct perf_annotate *ann) +{ + struct rb_node *nd = rb_first(&self->entries), *next; + int key = K_RIGHT; + + while (nd) { + struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node); + struct annotation *notes; + + if (he->ms.sym == NULL || he->ms.map->dso->annotate_warned) + goto find_next; + + notes = symbol__annotation(he->ms.sym); + if (notes->src == NULL) { +find_next: + if (key == K_LEFT) + nd = rb_prev(nd); + else + nd = rb_next(nd); + continue; + } + + if (use_browser == 2) { + int ret; + + ret = hist_entry__gtk_annotate(he, evidx, NULL); + if (!ret || !ann->skip_missing) + return; + + /* skip missing symbols */ + nd = rb_next(nd); + } else if (use_browser == 1) { + key = hist_entry__tui_annotate(he, evidx, NULL); + switch (key) { + case -1: + if (!ann->skip_missing) + return; + /* fall through */ + case K_RIGHT: + next = rb_next(nd); + break; + case K_LEFT: + next = rb_prev(nd); + break; + default: + return; + } + + if (next != NULL) + nd = next; + } else { + hist_entry__tty_annotate(he, evidx, ann); + nd = rb_next(nd); + /* + * Since we have a hist_entry per IP for the same + * symbol, free he->ms.sym->src to signal we already + * processed this symbol. + */ + free(notes->src); + notes->src = NULL; + } + } +} + +static int __cmd_annotate(struct perf_annotate *ann) +{ + int ret; + struct perf_session *session; + struct perf_evsel *pos; + u64 total_nr_samples; + + session = perf_session__new(input_name, O_RDONLY, + ann->force, false, &ann->tool); + if (session == NULL) + return -ENOMEM; + + if (ann->cpu_list) { + ret = perf_session__cpu_bitmap(session, ann->cpu_list, + ann->cpu_bitmap); + if (ret) + goto out_delete; + } + + if (!objdump_path) { + ret = perf_session_env__lookup_objdump(&session->header.env); + if (ret) + goto out_delete; + } + + ret = perf_session__process_events(session, &ann->tool); + if (ret) + goto out_delete; + + if (dump_trace) { + perf_session__fprintf_nr_events(session, stdout); + goto out_delete; + } + + if (verbose > 3) + perf_session__fprintf(session, stdout); + + if (verbose > 2) + perf_session__fprintf_dsos(session, stdout); + + total_nr_samples = 0; + list_for_each_entry(pos, &session->evlist->entries, node) { + struct hists *hists = &pos->hists; + u32 nr_samples = hists->stats.nr_events[PERF_RECORD_SAMPLE]; + + if (nr_samples > 0) { + total_nr_samples += nr_samples; + hists__collapse_resort(hists); + hists__output_resort(hists); + hists__find_annotations(hists, pos->idx, ann); + } + } + + if (total_nr_samples == 0) { + ui__error("The %s file has no samples!\n", session->filename); + goto out_delete; + } + + if (use_browser == 2) + perf_gtk__show_annotations(); + +out_delete: + /* + * Speed up the exit process, for large files this can + * take quite a while. + * + * XXX Enable this when using valgrind or if we ever + * librarize this command. + * + * Also experiment with obstacks to see how much speed + * up we'll get here. + * + * perf_session__delete(session); + */ + return ret; +} + +static const char * const annotate_usage[] = { + "perf annotate [<options>]", + NULL +}; + +int cmd_annotate(int argc, const char **argv, const char *prefix __maybe_unused) +{ + struct perf_annotate annotate = { + .tool = { + .sample = process_sample_event, + .mmap = perf_event__process_mmap, + .comm = perf_event__process_comm, + .exit = perf_event__process_exit, + .fork = perf_event__process_fork, + .ordered_samples = true, + .ordering_requires_timestamps = true, + }, + }; + const struct option options[] = { + OPT_STRING('i', "input", &input_name, "file", + "input file name"), + OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]", + "only consider symbols in these dsos"), + OPT_STRING('s', "symbol", &annotate.sym_hist_filter, "symbol", + "symbol to annotate"), + OPT_BOOLEAN('f', "force", &annotate.force, "don't complain, do it"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show symbol address, etc)"), + OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, + "dump raw trace in ASCII"), + OPT_BOOLEAN(0, "gtk", &annotate.use_gtk, "Use the GTK interface"), + OPT_BOOLEAN(0, "tui", &annotate.use_tui, "Use the TUI interface"), + OPT_BOOLEAN(0, "stdio", &annotate.use_stdio, "Use the stdio interface"), + OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name, + "file", "vmlinux pathname"), + OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules, + "load module symbols - WARNING: use only with -k and LIVE kernel"), + OPT_BOOLEAN('l', "print-line", &annotate.print_line, + "print matching source lines (may be slow)"), + OPT_BOOLEAN('P', "full-paths", &annotate.full_paths, + "Don't shorten the displayed pathnames"), + OPT_BOOLEAN(0, "skip-missing", &annotate.skip_missing, + "Skip symbols that cannot be annotated"), + OPT_STRING('C', "cpu", &annotate.cpu_list, "cpu", "list of cpus to profile"), + OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory", + "Look for files with symbols relative to this directory"), + OPT_BOOLEAN(0, "source", &symbol_conf.annotate_src, + "Interleave source code with assembly code (default)"), + OPT_BOOLEAN(0, "asm-raw", &symbol_conf.annotate_asm_raw, + "Display raw encoding of assembly instructions (default)"), + OPT_STRING('M', "disassembler-style", &disassembler_style, "disassembler style", + "Specify disassembler style (e.g. -M intel for intel syntax)"), + OPT_STRING(0, "objdump", &objdump_path, "path", + "objdump binary to use for disassembly and annotations"), + OPT_END() + }; + + argc = parse_options(argc, argv, options, annotate_usage, 0); + + if (annotate.use_stdio) + use_browser = 0; + else if (annotate.use_tui) + use_browser = 1; + else if (annotate.use_gtk) + use_browser = 2; + + setup_browser(true); + + symbol_conf.priv_size = sizeof(struct annotation); + symbol_conf.try_vmlinux_path = true; + + if (symbol__init() < 0) + return -1; + + if (setup_sorting() < 0) + usage_with_options(annotate_usage, options); + + 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); + + annotate.sym_hist_filter = argv[0]; + } + + return __cmd_annotate(&annotate); +} diff --git a/tools/perf/builtin-bench.c b/tools/perf/builtin-bench.c new file mode 100644 index 00000000..77298bf8 --- /dev/null +++ b/tools/perf/builtin-bench.c @@ -0,0 +1,267 @@ +/* + * + * builtin-bench.c + * + * General benchmarking subsystem provided by perf + * + * Copyright (C) 2009, Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp> + * + */ + +/* + * + * Available subsystem list: + * sched ... scheduler and IPC mechanism + * mem ... memory access performance + * + */ + +#include "perf.h" +#include "util/util.h" +#include "util/parse-options.h" +#include "builtin.h" +#include "bench/bench.h" + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +struct bench_suite { + const char *name; + const char *summary; + int (*fn)(int, const char **, const char *); +}; + \ +/* sentinel: easy for help */ +#define suite_all { "all", "Test all benchmark suites", NULL } + +#ifdef LIBNUMA_SUPPORT +static struct bench_suite numa_suites[] = { + { "mem", + "Benchmark for NUMA workloads", + bench_numa }, + suite_all, + { NULL, + NULL, + NULL } +}; +#endif + +static struct bench_suite sched_suites[] = { + { "messaging", + "Benchmark for scheduler and IPC mechanisms", + bench_sched_messaging }, + { "pipe", + "Flood of communication over pipe() between two processes", + bench_sched_pipe }, + suite_all, + { NULL, + NULL, + NULL } +}; + +static struct bench_suite mem_suites[] = { + { "memcpy", + "Simple memory copy in various ways", + bench_mem_memcpy }, + { "memset", + "Simple memory set in various ways", + bench_mem_memset }, + suite_all, + { NULL, + NULL, + NULL } +}; + +struct bench_subsys { + const char *name; + const char *summary; + struct bench_suite *suites; +}; + +static struct bench_subsys subsystems[] = { +#ifdef LIBNUMA_SUPPORT + { "numa", + "NUMA scheduling and MM behavior", + numa_suites }, +#endif + { "sched", + "scheduler and IPC mechanism", + sched_suites }, + { "mem", + "memory access performance", + mem_suites }, + { "all", /* sentinel: easy for help */ + "all benchmark subsystem", + NULL }, + { NULL, + NULL, + NULL } +}; + +static void dump_suites(int subsys_index) +{ + int i; + + printf("# List of available suites for %s...\n\n", + subsystems[subsys_index].name); + + for (i = 0; subsystems[subsys_index].suites[i].name; i++) + printf("%14s: %s\n", + subsystems[subsys_index].suites[i].name, + subsystems[subsys_index].suites[i].summary); + + printf("\n"); + return; +} + +static const char *bench_format_str; +int bench_format = BENCH_FORMAT_DEFAULT; + +static const struct option bench_options[] = { + OPT_STRING('f', "format", &bench_format_str, "default", + "Specify format style"), + OPT_END() +}; + +static const char * const bench_usage[] = { + "perf bench [<common options>] <subsystem> <suite> [<options>]", + NULL +}; + +static void print_usage(void) +{ + int i; + + printf("Usage: \n"); + for (i = 0; bench_usage[i]; i++) + printf("\t%s\n", bench_usage[i]); + printf("\n"); + + printf("# List of available subsystems...\n\n"); + + for (i = 0; subsystems[i].name; i++) + printf("%14s: %s\n", + subsystems[i].name, subsystems[i].summary); + printf("\n"); +} + +static int bench_str2int(const char *str) +{ + if (!str) + return BENCH_FORMAT_DEFAULT; + + if (!strcmp(str, BENCH_FORMAT_DEFAULT_STR)) + return BENCH_FORMAT_DEFAULT; + else if (!strcmp(str, BENCH_FORMAT_SIMPLE_STR)) + return BENCH_FORMAT_SIMPLE; + + return BENCH_FORMAT_UNKNOWN; +} + +static void all_suite(struct bench_subsys *subsys) /* FROM HERE */ +{ + int i; + const char *argv[2]; + struct bench_suite *suites = subsys->suites; + + argv[1] = NULL; + /* + * TODO: + * preparing preset parameters for + * embedded, ordinary PC, HPC, etc... + * will be helpful + */ + for (i = 0; suites[i].fn; i++) { + printf("# Running %s/%s benchmark...\n", + subsys->name, + suites[i].name); + fflush(stdout); + + argv[1] = suites[i].name; + suites[i].fn(1, argv, NULL); + printf("\n"); + } +} + +static void all_subsystem(void) +{ + int i; + for (i = 0; subsystems[i].suites; i++) + all_suite(&subsystems[i]); +} + +int cmd_bench(int argc, const char **argv, const char *prefix __maybe_unused) +{ + int i, j, status = 0; + + if (argc < 2) { + /* No subsystem specified. */ + print_usage(); + goto end; + } + + argc = parse_options(argc, argv, bench_options, bench_usage, + PARSE_OPT_STOP_AT_NON_OPTION); + + bench_format = bench_str2int(bench_format_str); + if (bench_format == BENCH_FORMAT_UNKNOWN) { + printf("Unknown format descriptor:%s\n", bench_format_str); + goto end; + } + + if (argc < 1) { + print_usage(); + goto end; + } + + if (!strcmp(argv[0], "all")) { + all_subsystem(); + goto end; + } + + for (i = 0; subsystems[i].name; i++) { + if (strcmp(subsystems[i].name, argv[0])) + continue; + + if (argc < 2) { + /* No suite specified. */ + dump_suites(i); + goto end; + } + + if (!strcmp(argv[1], "all")) { + all_suite(&subsystems[i]); + goto end; + } + + for (j = 0; subsystems[i].suites[j].name; j++) { + if (strcmp(subsystems[i].suites[j].name, argv[1])) + continue; + + if (bench_format == BENCH_FORMAT_DEFAULT) + printf("# Running %s/%s benchmark...\n", + subsystems[i].name, + subsystems[i].suites[j].name); + fflush(stdout); + status = subsystems[i].suites[j].fn(argc - 1, + argv + 1, prefix); + goto end; + } + + if (!strcmp(argv[1], "-h") || !strcmp(argv[1], "--help")) { + dump_suites(i); + goto end; + } + + printf("Unknown suite:%s for %s\n", argv[1], argv[0]); + status = 1; + goto end; + } + + printf("Unknown subsystem:%s\n", argv[0]); + status = 1; + +end: + return status; +} diff --git a/tools/perf/builtin-buildid-cache.c b/tools/perf/builtin-buildid-cache.c new file mode 100644 index 00000000..c96c8fa3 --- /dev/null +++ b/tools/perf/builtin-buildid-cache.c @@ -0,0 +1,221 @@ +/* + * builtin-buildid-cache.c + * + * Builtin buildid-cache command: Manages build-id cache + * + * Copyright (C) 2010, Red Hat Inc. + * Copyright (C) 2010, Arnaldo Carvalho de Melo <acme@redhat.com> + */ +#include "builtin.h" +#include "perf.h" +#include "util/cache.h" +#include "util/debug.h" +#include "util/header.h" +#include "util/parse-options.h" +#include "util/strlist.h" +#include "util/build-id.h" +#include "util/session.h" +#include "util/symbol.h" + +static int build_id_cache__add_file(const char *filename, const char *debugdir) +{ + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + u8 build_id[BUILD_ID_SIZE]; + int err; + + if (filename__read_build_id(filename, &build_id, sizeof(build_id)) < 0) { + pr_debug("Couldn't read a build-id in %s\n", filename); + return -1; + } + + build_id__sprintf(build_id, sizeof(build_id), sbuild_id); + err = build_id_cache__add_s(sbuild_id, debugdir, filename, + false, false); + if (verbose) + pr_info("Adding %s %s: %s\n", sbuild_id, filename, + err ? "FAIL" : "Ok"); + return err; +} + +static int build_id_cache__remove_file(const char *filename, + const char *debugdir) +{ + u8 build_id[BUILD_ID_SIZE]; + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + int err; + + if (filename__read_build_id(filename, &build_id, sizeof(build_id)) < 0) { + pr_debug("Couldn't read a build-id in %s\n", filename); + return -1; + } + + build_id__sprintf(build_id, sizeof(build_id), sbuild_id); + err = build_id_cache__remove_s(sbuild_id, debugdir); + if (verbose) + pr_info("Removing %s %s: %s\n", sbuild_id, filename, + err ? "FAIL" : "Ok"); + + return err; +} + +static bool dso__missing_buildid_cache(struct dso *dso, int parm __maybe_unused) +{ + char filename[PATH_MAX]; + u8 build_id[BUILD_ID_SIZE]; + + if (dso__build_id_filename(dso, filename, sizeof(filename)) && + filename__read_build_id(filename, build_id, + sizeof(build_id)) != sizeof(build_id)) { + if (errno == ENOENT) + return false; + + pr_warning("Problems with %s file, consider removing it from the cache\n", + filename); + } else if (memcmp(dso->build_id, build_id, sizeof(dso->build_id))) { + pr_warning("Problems with %s file, consider removing it from the cache\n", + filename); + } + + return true; +} + +static int build_id_cache__fprintf_missing(const char *filename, bool force, FILE *fp) +{ + struct perf_session *session = perf_session__new(filename, O_RDONLY, + force, false, NULL); + if (session == NULL) + return -1; + + perf_session__fprintf_dsos_buildid(session, fp, dso__missing_buildid_cache, 0); + perf_session__delete(session); + + return 0; +} + +static int build_id_cache__update_file(const char *filename, + const char *debugdir) +{ + u8 build_id[BUILD_ID_SIZE]; + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + int err; + + if (filename__read_build_id(filename, &build_id, sizeof(build_id)) < 0) { + pr_debug("Couldn't read a build-id in %s\n", filename); + return -1; + } + + build_id__sprintf(build_id, sizeof(build_id), sbuild_id); + err = build_id_cache__remove_s(sbuild_id, debugdir); + if (!err) { + err = build_id_cache__add_s(sbuild_id, debugdir, filename, + false, false); + } + if (verbose) + pr_info("Updating %s %s: %s\n", sbuild_id, filename, + err ? "FAIL" : "Ok"); + + return err; +} + +int cmd_buildid_cache(int argc, const char **argv, + const char *prefix __maybe_unused) +{ + struct strlist *list; + struct str_node *pos; + int ret = 0; + bool force = false; + char debugdir[PATH_MAX]; + char const *add_name_list_str = NULL, + *remove_name_list_str = NULL, + *missing_filename = NULL, + *update_name_list_str = NULL; + + const struct option buildid_cache_options[] = { + OPT_STRING('a', "add", &add_name_list_str, + "file list", "file(s) to add"), + OPT_STRING('r', "remove", &remove_name_list_str, "file list", + "file(s) to remove"), + OPT_STRING('M', "missing", &missing_filename, "file", + "to find missing build ids in the cache"), + OPT_BOOLEAN('f', "force", &force, "don't complain, do it"), + OPT_STRING('u', "update", &update_name_list_str, "file list", + "file(s) to update"), + OPT_INCR('v', "verbose", &verbose, "be more verbose"), + OPT_END() + }; + const char * const buildid_cache_usage[] = { + "perf buildid-cache [<options>]", + NULL + }; + + argc = parse_options(argc, argv, buildid_cache_options, + buildid_cache_usage, 0); + + if (symbol__init() < 0) + return -1; + + setup_pager(); + + snprintf(debugdir, sizeof(debugdir), "%s", buildid_dir); + + if (add_name_list_str) { + list = strlist__new(true, add_name_list_str); + if (list) { + strlist__for_each(pos, list) + if (build_id_cache__add_file(pos->s, debugdir)) { + if (errno == EEXIST) { + pr_debug("%s already in the cache\n", + pos->s); + continue; + } + pr_warning("Couldn't add %s: %s\n", + pos->s, strerror(errno)); + } + + strlist__delete(list); + } + } + + if (remove_name_list_str) { + list = strlist__new(true, remove_name_list_str); + if (list) { + strlist__for_each(pos, list) + if (build_id_cache__remove_file(pos->s, debugdir)) { + if (errno == ENOENT) { + pr_debug("%s wasn't in the cache\n", + pos->s); + continue; + } + pr_warning("Couldn't remove %s: %s\n", + pos->s, strerror(errno)); + } + + strlist__delete(list); + } + } + + if (missing_filename) + ret = build_id_cache__fprintf_missing(missing_filename, force, stdout); + + if (update_name_list_str) { + list = strlist__new(true, update_name_list_str); + if (list) { + strlist__for_each(pos, list) + if (build_id_cache__update_file(pos->s, debugdir)) { + if (errno == ENOENT) { + pr_debug("%s wasn't in the cache\n", + pos->s); + continue; + } + pr_warning("Couldn't update %s: %s\n", + pos->s, strerror(errno)); + } + + strlist__delete(list); + } + } + + return ret; +} diff --git a/tools/perf/builtin-buildid-list.c b/tools/perf/builtin-buildid-list.c new file mode 100644 index 00000000..e74366a1 --- /dev/null +++ b/tools/perf/builtin-buildid-list.c @@ -0,0 +1,106 @@ +/* + * builtin-buildid-list.c + * + * Builtin buildid-list command: list buildids in perf.data, in the running + * kernel and in ELF files. + * + * Copyright (C) 2009, Red Hat Inc. + * Copyright (C) 2009, Arnaldo Carvalho de Melo <acme@redhat.com> + */ +#include "builtin.h" +#include "perf.h" +#include "util/build-id.h" +#include "util/cache.h" +#include "util/debug.h" +#include "util/parse-options.h" +#include "util/session.h" +#include "util/symbol.h" + +static int sysfs__fprintf_build_id(FILE *fp) +{ + u8 kallsyms_build_id[BUILD_ID_SIZE]; + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + if (sysfs__read_build_id("/sys/kernel/notes", kallsyms_build_id, + sizeof(kallsyms_build_id)) != 0) + return -1; + + build_id__sprintf(kallsyms_build_id, sizeof(kallsyms_build_id), + sbuild_id); + fprintf(fp, "%s\n", sbuild_id); + return 0; +} + +static int filename__fprintf_build_id(const char *name, FILE *fp) +{ + u8 build_id[BUILD_ID_SIZE]; + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + if (filename__read_build_id(name, build_id, + sizeof(build_id)) != sizeof(build_id)) + return 0; + + build_id__sprintf(build_id, sizeof(build_id), sbuild_id); + return fprintf(fp, "%s\n", sbuild_id); +} + +static bool dso__skip_buildid(struct dso *dso, int with_hits) +{ + return with_hits && !dso->hit; +} + +static int perf_session__list_build_ids(bool force, bool with_hits) +{ + struct perf_session *session; + + symbol__elf_init(); + /* + * See if this is an ELF file first: + */ + if (filename__fprintf_build_id(input_name, stdout)) + goto out; + + session = perf_session__new(input_name, O_RDONLY, force, false, + &build_id__mark_dso_hit_ops); + if (session == NULL) + return -1; + /* + * in pipe-mode, the only way to get the buildids is to parse + * the record stream. Buildids are stored as RECORD_HEADER_BUILD_ID + */ + if (with_hits || session->fd_pipe) + perf_session__process_events(session, &build_id__mark_dso_hit_ops); + + perf_session__fprintf_dsos_buildid(session, stdout, dso__skip_buildid, with_hits); + perf_session__delete(session); +out: + return 0; +} + +int cmd_buildid_list(int argc, const char **argv, + const char *prefix __maybe_unused) +{ + bool show_kernel = false; + bool with_hits = false; + bool force = false; + const struct option options[] = { + OPT_BOOLEAN('H', "with-hits", &with_hits, "Show only DSOs with hits"), + OPT_STRING('i', "input", &input_name, "file", "input file name"), + OPT_BOOLEAN('f', "force", &force, "don't complain, do it"), + OPT_BOOLEAN('k', "kernel", &show_kernel, "Show current kernel build id"), + OPT_INCR('v', "verbose", &verbose, "be more verbose"), + OPT_END() + }; + const char * const buildid_list_usage[] = { + "perf buildid-list [<options>]", + NULL + }; + + argc = parse_options(argc, argv, options, buildid_list_usage, 0); + setup_pager(); + + if (show_kernel) + return sysfs__fprintf_build_id(stdout); + + return perf_session__list_build_ids(force, with_hits); +} diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c new file mode 100644 index 00000000..d207a97a --- /dev/null +++ b/tools/perf/builtin-diff.c @@ -0,0 +1,618 @@ +/* + * builtin-diff.c + * + * Builtin diff command: Analyze two perf.data input files, look up and read + * DSOs and symbol information, sort them and produce a diff. + */ +#include "builtin.h" + +#include "util/debug.h" +#include "util/event.h" +#include "util/hist.h" +#include "util/evsel.h" +#include "util/evlist.h" +#include "util/session.h" +#include "util/tool.h" +#include "util/sort.h" +#include "util/symbol.h" +#include "util/util.h" + +#include <stdlib.h> + +static char const *input_old = "perf.data.old", + *input_new = "perf.data"; +static char diff__default_sort_order[] = "dso,symbol"; +static bool force; +static bool show_period; +static bool show_formula; +static bool show_baseline_only; +static bool sort_compute; + +static s64 compute_wdiff_w1; +static s64 compute_wdiff_w2; + +enum { + COMPUTE_DELTA, + COMPUTE_RATIO, + COMPUTE_WEIGHTED_DIFF, + COMPUTE_MAX, +}; + +const char *compute_names[COMPUTE_MAX] = { + [COMPUTE_DELTA] = "delta", + [COMPUTE_RATIO] = "ratio", + [COMPUTE_WEIGHTED_DIFF] = "wdiff", +}; + +static int compute; + +static int setup_compute_opt_wdiff(char *opt) +{ + char *w1_str = opt; + char *w2_str; + + int ret = -EINVAL; + + if (!opt) + goto out; + + w2_str = strchr(opt, ','); + if (!w2_str) + goto out; + + *w2_str++ = 0x0; + if (!*w2_str) + goto out; + + compute_wdiff_w1 = strtol(w1_str, NULL, 10); + compute_wdiff_w2 = strtol(w2_str, NULL, 10); + + if (!compute_wdiff_w1 || !compute_wdiff_w2) + goto out; + + pr_debug("compute wdiff w1(%" PRId64 ") w2(%" PRId64 ")\n", + compute_wdiff_w1, compute_wdiff_w2); + + ret = 0; + + out: + if (ret) + pr_err("Failed: wrong weight data, use 'wdiff:w1,w2'\n"); + + return ret; +} + +static int setup_compute_opt(char *opt) +{ + if (compute == COMPUTE_WEIGHTED_DIFF) + return setup_compute_opt_wdiff(opt); + + if (opt) { + pr_err("Failed: extra option specified '%s'", opt); + return -EINVAL; + } + + return 0; +} + +static int setup_compute(const struct option *opt, const char *str, + int unset __maybe_unused) +{ + int *cp = (int *) opt->value; + char *cstr = (char *) str; + char buf[50]; + unsigned i; + char *option; + + if (!str) { + *cp = COMPUTE_DELTA; + return 0; + } + + if (*str == '+') { + sort_compute = true; + cstr = (char *) ++str; + if (!*str) + return 0; + } + + option = strchr(str, ':'); + if (option) { + unsigned len = option++ - str; + + /* + * The str data are not writeable, so we need + * to use another buffer. + */ + + /* No option value is longer. */ + if (len >= sizeof(buf)) + return -EINVAL; + + strncpy(buf, str, len); + buf[len] = 0x0; + cstr = buf; + } + + for (i = 0; i < COMPUTE_MAX; i++) + if (!strcmp(cstr, compute_names[i])) { + *cp = i; + return setup_compute_opt(option); + } + + pr_err("Failed: '%s' is not computation method " + "(use 'delta','ratio' or 'wdiff')\n", str); + return -EINVAL; +} + +double perf_diff__period_percent(struct hist_entry *he, u64 period) +{ + u64 total = he->hists->stats.total_period; + return (period * 100.0) / total; +} + +double perf_diff__compute_delta(struct hist_entry *he, struct hist_entry *pair) +{ + double new_percent = perf_diff__period_percent(he, he->stat.period); + double old_percent = perf_diff__period_percent(pair, pair->stat.period); + + he->diff.period_ratio_delta = new_percent - old_percent; + he->diff.computed = true; + return he->diff.period_ratio_delta; +} + +double perf_diff__compute_ratio(struct hist_entry *he, struct hist_entry *pair) +{ + double new_period = he->stat.period; + double old_period = pair->stat.period; + + he->diff.computed = true; + he->diff.period_ratio = new_period / old_period; + return he->diff.period_ratio; +} + +s64 perf_diff__compute_wdiff(struct hist_entry *he, struct hist_entry *pair) +{ + u64 new_period = he->stat.period; + u64 old_period = pair->stat.period; + + he->diff.computed = true; + he->diff.wdiff = new_period * compute_wdiff_w2 - + old_period * compute_wdiff_w1; + + return he->diff.wdiff; +} + +static int formula_delta(struct hist_entry *he, struct hist_entry *pair, + char *buf, size_t size) +{ + return scnprintf(buf, size, + "(%" PRIu64 " * 100 / %" PRIu64 ") - " + "(%" PRIu64 " * 100 / %" PRIu64 ")", + he->stat.period, he->hists->stats.total_period, + pair->stat.period, pair->hists->stats.total_period); +} + +static int formula_ratio(struct hist_entry *he, struct hist_entry *pair, + char *buf, size_t size) +{ + double new_period = he->stat.period; + double old_period = pair->stat.period; + + return scnprintf(buf, size, "%.0F / %.0F", new_period, old_period); +} + +static int formula_wdiff(struct hist_entry *he, struct hist_entry *pair, + char *buf, size_t size) +{ + u64 new_period = he->stat.period; + u64 old_period = pair->stat.period; + + return scnprintf(buf, size, + "(%" PRIu64 " * " "%" PRId64 ") - (%" PRIu64 " * " "%" PRId64 ")", + new_period, compute_wdiff_w2, old_period, compute_wdiff_w1); +} + +int perf_diff__formula(struct hist_entry *he, struct hist_entry *pair, + char *buf, size_t size) +{ + switch (compute) { + case COMPUTE_DELTA: + return formula_delta(he, pair, buf, size); + case COMPUTE_RATIO: + return formula_ratio(he, pair, buf, size); + case COMPUTE_WEIGHTED_DIFF: + return formula_wdiff(he, pair, buf, size); + default: + BUG_ON(1); + } + + return -1; +} + +static int hists__add_entry(struct hists *self, + struct addr_location *al, u64 period) +{ + if (__hists__add_entry(self, al, NULL, period) != NULL) + return 0; + return -ENOMEM; +} + +static int diff__process_sample_event(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct addr_location al; + + if (perf_event__preprocess_sample(event, machine, &al, sample, NULL) < 0) { + pr_warning("problem processing %d event, skipping it.\n", + event->header.type); + return -1; + } + + if (al.filtered) + return 0; + + if (hists__add_entry(&evsel->hists, &al, sample->period)) { + pr_warning("problem incrementing symbol period, skipping event\n"); + return -1; + } + + evsel->hists.stats.total_period += sample->period; + return 0; +} + +static struct perf_tool tool = { + .sample = diff__process_sample_event, + .mmap = perf_event__process_mmap, + .comm = perf_event__process_comm, + .exit = perf_event__process_exit, + .fork = perf_event__process_fork, + .lost = perf_event__process_lost, + .ordered_samples = true, + .ordering_requires_timestamps = true, +}; + +static struct perf_evsel *evsel_match(struct perf_evsel *evsel, + struct perf_evlist *evlist) +{ + struct perf_evsel *e; + + list_for_each_entry(e, &evlist->entries, node) + if (perf_evsel__match2(evsel, e)) + return e; + + return NULL; +} + +static void perf_evlist__collapse_resort(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel; + + list_for_each_entry(evsel, &evlist->entries, node) { + struct hists *hists = &evsel->hists; + + hists__collapse_resort(hists); + } +} + +static void hists__baseline_only(struct hists *hists) +{ + struct rb_root *root; + struct rb_node *next; + + if (sort__need_collapse) + root = &hists->entries_collapsed; + else + root = hists->entries_in; + + next = rb_first(root); + while (next != NULL) { + struct hist_entry *he = rb_entry(next, struct hist_entry, rb_node_in); + + next = rb_next(&he->rb_node_in); + if (!hist_entry__next_pair(he)) { + rb_erase(&he->rb_node_in, root); + hist_entry__free(he); + } + } +} + +static void hists__precompute(struct hists *hists) +{ + struct rb_node *next = rb_first(&hists->entries); + + while (next != NULL) { + struct hist_entry *he = rb_entry(next, struct hist_entry, rb_node); + struct hist_entry *pair = hist_entry__next_pair(he); + + next = rb_next(&he->rb_node); + if (!pair) + continue; + + switch (compute) { + case COMPUTE_DELTA: + perf_diff__compute_delta(he, pair); + break; + case COMPUTE_RATIO: + perf_diff__compute_ratio(he, pair); + break; + case COMPUTE_WEIGHTED_DIFF: + perf_diff__compute_wdiff(he, pair); + break; + default: + BUG_ON(1); + } + } +} + +static int64_t cmp_doubles(double l, double r) +{ + if (l > r) + return -1; + else if (l < r) + return 1; + else + return 0; +} + +static int64_t +hist_entry__cmp_compute(struct hist_entry *left, struct hist_entry *right, + int c) +{ + switch (c) { + case COMPUTE_DELTA: + { + double l = left->diff.period_ratio_delta; + double r = right->diff.period_ratio_delta; + + return cmp_doubles(l, r); + } + case COMPUTE_RATIO: + { + double l = left->diff.period_ratio; + double r = right->diff.period_ratio; + + return cmp_doubles(l, r); + } + case COMPUTE_WEIGHTED_DIFF: + { + s64 l = left->diff.wdiff; + s64 r = right->diff.wdiff; + + return r - l; + } + default: + BUG_ON(1); + } + + return 0; +} + +static void insert_hist_entry_by_compute(struct rb_root *root, + struct hist_entry *he, + int c) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent = NULL; + struct hist_entry *iter; + + while (*p != NULL) { + parent = *p; + iter = rb_entry(parent, struct hist_entry, rb_node); + if (hist_entry__cmp_compute(he, iter, c) < 0) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + rb_link_node(&he->rb_node, parent, p); + rb_insert_color(&he->rb_node, root); +} + +static void hists__compute_resort(struct hists *hists) +{ + struct rb_root *root; + struct rb_node *next; + + if (sort__need_collapse) + root = &hists->entries_collapsed; + else + root = hists->entries_in; + + hists->entries = RB_ROOT; + next = rb_first(root); + + hists->nr_entries = 0; + hists->stats.total_period = 0; + hists__reset_col_len(hists); + + while (next != NULL) { + struct hist_entry *he; + + he = rb_entry(next, struct hist_entry, rb_node_in); + next = rb_next(&he->rb_node_in); + + insert_hist_entry_by_compute(&hists->entries, he, compute); + hists__inc_nr_entries(hists, he); + } +} + +static void hists__process(struct hists *old, struct hists *new) +{ + hists__match(new, old); + + if (show_baseline_only) + hists__baseline_only(new); + else + hists__link(new, old); + + if (sort_compute) { + hists__precompute(new); + hists__compute_resort(new); + } else { + hists__output_resort(new); + } + + hists__fprintf(new, true, 0, 0, stdout); +} + +static int __cmd_diff(void) +{ + int ret, i; +#define older (session[0]) +#define newer (session[1]) + struct perf_session *session[2]; + struct perf_evlist *evlist_new, *evlist_old; + struct perf_evsel *evsel; + bool first = true; + + older = perf_session__new(input_old, O_RDONLY, force, false, + &tool); + newer = perf_session__new(input_new, O_RDONLY, force, false, + &tool); + if (session[0] == NULL || session[1] == NULL) + return -ENOMEM; + + for (i = 0; i < 2; ++i) { + ret = perf_session__process_events(session[i], &tool); + if (ret) + goto out_delete; + } + + evlist_old = older->evlist; + evlist_new = newer->evlist; + + perf_evlist__collapse_resort(evlist_old); + perf_evlist__collapse_resort(evlist_new); + + list_for_each_entry(evsel, &evlist_new->entries, node) { + struct perf_evsel *evsel_old; + + evsel_old = evsel_match(evsel, evlist_old); + if (!evsel_old) + continue; + + fprintf(stdout, "%s# Event '%s'\n#\n", first ? "" : "\n", + perf_evsel__name(evsel)); + + first = false; + + hists__process(&evsel_old->hists, &evsel->hists); + } + +out_delete: + for (i = 0; i < 2; ++i) + perf_session__delete(session[i]); + return ret; +#undef older +#undef newer +} + +static const char * const diff_usage[] = { + "perf diff [<options>] [old_file] [new_file]", + NULL, +}; + +static const struct option options[] = { + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show symbol address, etc)"), + OPT_BOOLEAN('b', "baseline-only", &show_baseline_only, + "Show only items with match in baseline"), + OPT_CALLBACK('c', "compute", &compute, + "delta,ratio,wdiff:w1,w2 (default delta)", + "Entries differential computation selection", + setup_compute), + OPT_BOOLEAN('p', "period", &show_period, + "Show period values."), + OPT_BOOLEAN('F', "formula", &show_formula, + "Show formula."), + OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, + "dump raw trace in ASCII"), + OPT_BOOLEAN('f', "force", &force, "don't complain, do it"), + OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules, + "load module symbols - WARNING: use only with -k and LIVE kernel"), + OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]", + "only consider symbols in these dsos"), + OPT_STRING('C', "comms", &symbol_conf.comm_list_str, "comm[,comm...]", + "only consider symbols in these comms"), + OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]", + "only consider these symbols"), + OPT_STRING('s', "sort", &sort_order, "key[,key2...]", + "sort by key(s): pid, comm, dso, symbol, parent"), + OPT_STRING('t', "field-separator", &symbol_conf.field_sep, "separator", + "separator for columns, no spaces will be added between " + "columns '.' is reserved."), + OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory", + "Look for files with symbols relative to this directory"), + OPT_END() +}; + +static void ui_init(void) +{ + /* + * Display baseline/delta/ratio + * formula/periods columns. + */ + perf_hpp__column_enable(PERF_HPP__BASELINE); + + switch (compute) { + case COMPUTE_DELTA: + perf_hpp__column_enable(PERF_HPP__DELTA); + break; + case COMPUTE_RATIO: + perf_hpp__column_enable(PERF_HPP__RATIO); + break; + case COMPUTE_WEIGHTED_DIFF: + perf_hpp__column_enable(PERF_HPP__WEIGHTED_DIFF); + break; + default: + BUG_ON(1); + }; + + if (show_formula) + perf_hpp__column_enable(PERF_HPP__FORMULA); + + if (show_period) { + perf_hpp__column_enable(PERF_HPP__PERIOD); + perf_hpp__column_enable(PERF_HPP__PERIOD_BASELINE); + } +} + +int cmd_diff(int argc, const char **argv, const char *prefix __maybe_unused) +{ + sort_order = diff__default_sort_order; + argc = parse_options(argc, argv, options, diff_usage, 0); + if (argc) { + if (argc > 2) + usage_with_options(diff_usage, options); + if (argc == 2) { + input_old = argv[0]; + input_new = argv[1]; + } else + input_new = argv[0]; + } else if (symbol_conf.default_guest_vmlinux_name || + symbol_conf.default_guest_kallsyms) { + input_old = "perf.data.host"; + input_new = "perf.data.guest"; + } + + symbol_conf.exclude_other = false; + if (symbol__init() < 0) + return -1; + + ui_init(); + + if (setup_sorting() < 0) + usage_with_options(diff_usage, options); + + setup_pager(); + + sort_entry__setup_elide(&sort_dso, symbol_conf.dso_list, "dso", NULL); + sort_entry__setup_elide(&sort_comm, symbol_conf.comm_list, "comm", NULL); + sort_entry__setup_elide(&sort_sym, symbol_conf.sym_list, "symbol", NULL); + + return __cmd_diff(); +} diff --git a/tools/perf/builtin-evlist.c b/tools/perf/builtin-evlist.c new file mode 100644 index 00000000..05bd9dfe --- /dev/null +++ b/tools/perf/builtin-evlist.c @@ -0,0 +1,61 @@ +/* + * Builtin evlist command: Show the list of event selectors present + * in a perf.data file. + */ +#include "builtin.h" + +#include "util/util.h" + +#include <linux/list.h> + +#include "perf.h" +#include "util/evlist.h" +#include "util/evsel.h" +#include "util/parse-events.h" +#include "util/parse-options.h" +#include "util/session.h" + +static int __cmd_evlist(const char *file_name, struct perf_attr_details *details) +{ + struct perf_session *session; + struct perf_evsel *pos; + + session = perf_session__new(file_name, O_RDONLY, 0, false, NULL); + if (session == NULL) + return -ENOMEM; + + list_for_each_entry(pos, &session->evlist->entries, node) + perf_evsel__fprintf(pos, details, stdout); + + perf_session__delete(session); + return 0; +} + +int cmd_evlist(int argc, const char **argv, const char *prefix __maybe_unused) +{ + struct perf_attr_details details = { .verbose = false, }; + const struct option options[] = { + OPT_STRING('i', "input", &input_name, "file", "Input file name"), + OPT_BOOLEAN('F', "freq", &details.freq, "Show the sample frequency"), + OPT_BOOLEAN('v', "verbose", &details.verbose, + "Show all event attr details"), + OPT_BOOLEAN('g', "group", &details.event_group, + "Show event group information"), + OPT_END() + }; + const char * const evlist_usage[] = { + "perf evlist [<options>]", + NULL + }; + + argc = parse_options(argc, argv, options, evlist_usage, 0); + if (argc) + usage_with_options(evlist_usage, options); + + if (details.event_group && (details.verbose || details.freq)) { + pr_err("--group option is not compatible with other options\n"); + usage_with_options(evlist_usage, options); + } + + return __cmd_evlist(input_name, &details); +} diff --git a/tools/perf/builtin-help.c b/tools/perf/builtin-help.c new file mode 100644 index 00000000..178b88ae --- /dev/null +++ b/tools/perf/builtin-help.c @@ -0,0 +1,479 @@ +/* + * builtin-help.c + * + * Builtin help command + */ +#include "perf.h" +#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_NONE, + HELP_FORMAT_MAN, + HELP_FORMAT_INFO, + HELP_FORMAT_WEB, +}; + +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; + + pr_err("unrecognized help format '%s'", format); + return HELP_FORMAT_NONE; +} + +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 = zalloc(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 = zalloc(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) +{ + enum help_format *help_formatp = cb; + + if (!strcmp(var, "help.format")) { + if (!value) + return config_error_nonbool(var); + *help_formatp = parse_help_format(value); + if (*help_formatp == HELP_FORMAT_NONE) + return -1; + 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) +{ + unsigned 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 ", longest, 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 + 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 int 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); + + pr_err("no man viewer handled the request"); + return -1; +} + +static int 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); + return -1; +} + +static int 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)) { + pr_err("'%s': not a documentation directory.", html_path); + return -1; + } + + strbuf_init(page_path, 0); + strbuf_addf(page_path, "%s/%s.html", html_path, page); + + return 0; +} + +/* + * 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 int 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 */ + + if (get_html_page_path(&page_path, page) != 0) + return -1; + + open_html(page_path.buf); + + return 0; +} + +int cmd_help(int argc, const char **argv, const char *prefix __maybe_unused) +{ + bool show_all = false; + enum help_format help_format = HELP_FORMAT_MAN; + struct option builtin_help_options[] = { + OPT_BOOLEAN('a', "all", &show_all, "print all available commands"), + OPT_SET_UINT('m', "man", &help_format, "show man page", HELP_FORMAT_MAN), + OPT_SET_UINT('w', "web", &help_format, "show manual in web browser", + HELP_FORMAT_WEB), + OPT_SET_UINT('i', "info", &help_format, "show info page", + HELP_FORMAT_INFO), + OPT_END(), + }; + const char * const builtin_help_usage[] = { + "perf help [--all] [--man|--web|--info] [command]", + NULL + }; + const char *alias; + int rc = 0; + + load_command_list("perf-", &main_cmds, &other_cmds); + + perf_config(perf_help_config, &help_format); + + 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: + rc = show_man_page(argv[0]); + break; + case HELP_FORMAT_INFO: + rc = show_info_page(argv[0]); + break; + case HELP_FORMAT_WEB: + rc = show_html_page(argv[0]); + break; + case HELP_FORMAT_NONE: + /* fall-through */ + default: + rc = -1; + break; + } + + return rc; +} diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c new file mode 100644 index 00000000..84ad6abe --- /dev/null +++ b/tools/perf/builtin-inject.c @@ -0,0 +1,461 @@ +/* + * builtin-inject.c + * + * Builtin inject command: Examine the live mode (stdin) event stream + * and repipe it to stdout while optionally injecting additional + * events into it. + */ +#include "builtin.h" + +#include "perf.h" +#include "util/color.h" +#include "util/evlist.h" +#include "util/evsel.h" +#include "util/session.h" +#include "util/tool.h" +#include "util/debug.h" +#include "util/build-id.h" + +#include "util/parse-options.h" + +#include <linux/list.h> + +struct perf_inject { + struct perf_tool tool; + bool build_ids; + bool sched_stat; + const char *input_name; + int pipe_output, + output; + u64 bytes_written; + struct list_head samples; +}; + +struct event_entry { + struct list_head node; + u32 tid; + union perf_event event[0]; +}; + +static int perf_event__repipe_synth(struct perf_tool *tool, + union perf_event *event, + struct machine *machine __maybe_unused) +{ + struct perf_inject *inject = container_of(tool, struct perf_inject, tool); + uint32_t size; + void *buf = event; + + size = event->header.size; + + while (size) { + int ret = write(inject->output, buf, size); + if (ret < 0) + return -errno; + + size -= ret; + buf += ret; + inject->bytes_written += ret; + } + + return 0; +} + +static int perf_event__repipe_op2_synth(struct perf_tool *tool, + union perf_event *event, + struct perf_session *session + __maybe_unused) +{ + return perf_event__repipe_synth(tool, event, NULL); +} + +static int perf_event__repipe_event_type_synth(struct perf_tool *tool, + union perf_event *event) +{ + return perf_event__repipe_synth(tool, event, NULL); +} + +static int perf_event__repipe_tracing_data_synth(union perf_event *event, + struct perf_session *session + __maybe_unused) +{ + return perf_event__repipe_synth(NULL, event, NULL); +} + +static int perf_event__repipe_attr(union perf_event *event, + struct perf_evlist **pevlist __maybe_unused) +{ + int ret; + ret = perf_event__process_attr(event, pevlist); + if (ret) + return ret; + + return perf_event__repipe_synth(NULL, event, NULL); +} + +static int perf_event__repipe(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine) +{ + return perf_event__repipe_synth(tool, event, machine); +} + +typedef int (*inject_handler)(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine); + +static int perf_event__repipe_sample(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + if (evsel->handler.func) { + inject_handler f = evsel->handler.func; + return f(tool, event, sample, evsel, machine); + } + + build_id__mark_dso_hit(tool, event, sample, evsel, machine); + + return perf_event__repipe_synth(tool, event, machine); +} + +static int perf_event__repipe_mmap(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct machine *machine) +{ + int err; + + err = perf_event__process_mmap(tool, event, sample, machine); + perf_event__repipe(tool, event, sample, machine); + + return err; +} + +static int perf_event__repipe_fork(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct machine *machine) +{ + int err; + + err = perf_event__process_fork(tool, event, sample, machine); + perf_event__repipe(tool, event, sample, machine); + + return err; +} + +static int perf_event__repipe_tracing_data(union perf_event *event, + struct perf_session *session) +{ + int err; + + perf_event__repipe_synth(NULL, event, NULL); + err = perf_event__process_tracing_data(event, session); + + return err; +} + +static int dso__read_build_id(struct dso *self) +{ + if (self->has_build_id) + return 0; + + if (filename__read_build_id(self->long_name, self->build_id, + sizeof(self->build_id)) > 0) { + self->has_build_id = true; + return 0; + } + + return -1; +} + +static int dso__inject_build_id(struct dso *self, struct perf_tool *tool, + struct machine *machine) +{ + u16 misc = PERF_RECORD_MISC_USER; + int err; + + if (dso__read_build_id(self) < 0) { + pr_debug("no build_id found for %s\n", self->long_name); + return -1; + } + + if (self->kernel) + misc = PERF_RECORD_MISC_KERNEL; + + err = perf_event__synthesize_build_id(tool, self, misc, perf_event__repipe, + machine); + if (err) { + pr_err("Can't synthesize build_id event for %s\n", self->long_name); + return -1; + } + + return 0; +} + +static int perf_event__inject_buildid(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel __maybe_unused, + struct machine *machine) +{ + struct addr_location al; + struct thread *thread; + u8 cpumode; + + cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; + + thread = machine__findnew_thread(machine, event->ip.pid); + if (thread == NULL) { + pr_err("problem processing %d event, skipping it.\n", + event->header.type); + goto repipe; + } + + thread__find_addr_map(thread, machine, cpumode, MAP__FUNCTION, + event->ip.ip, &al); + + if (al.map != NULL) { + if (!al.map->dso->hit) { + al.map->dso->hit = 1; + if (map__load(al.map, NULL) >= 0) { + dso__inject_build_id(al.map->dso, tool, machine); + /* + * If this fails, too bad, let the other side + * account this as unresolved. + */ + } else { +#ifdef LIBELF_SUPPORT + pr_warning("no symbols found in %s, maybe " + "install a debug package?\n", + al.map->dso->long_name); +#endif + } + } + } + +repipe: + perf_event__repipe(tool, event, sample, machine); + return 0; +} + +static int perf_inject__sched_process_exit(struct perf_tool *tool, + union perf_event *event __maybe_unused, + struct perf_sample *sample, + struct perf_evsel *evsel __maybe_unused, + struct machine *machine __maybe_unused) +{ + struct perf_inject *inject = container_of(tool, struct perf_inject, tool); + struct event_entry *ent; + + list_for_each_entry(ent, &inject->samples, node) { + if (sample->tid == ent->tid) { + list_del_init(&ent->node); + free(ent); + break; + } + } + + return 0; +} + +static int perf_inject__sched_switch(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct perf_inject *inject = container_of(tool, struct perf_inject, tool); + struct event_entry *ent; + + perf_inject__sched_process_exit(tool, event, sample, evsel, machine); + + ent = malloc(event->header.size + sizeof(struct event_entry)); + if (ent == NULL) { + color_fprintf(stderr, PERF_COLOR_RED, + "Not enough memory to process sched switch event!"); + return -1; + } + + ent->tid = sample->tid; + memcpy(&ent->event, event, event->header.size); + list_add(&ent->node, &inject->samples); + return 0; +} + +static int perf_inject__sched_stat(struct perf_tool *tool, + union perf_event *event __maybe_unused, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct event_entry *ent; + union perf_event *event_sw; + struct perf_sample sample_sw; + struct perf_inject *inject = container_of(tool, struct perf_inject, tool); + u32 pid = perf_evsel__intval(evsel, sample, "pid"); + + list_for_each_entry(ent, &inject->samples, node) { + if (pid == ent->tid) + goto found; + } + + return 0; +found: + event_sw = &ent->event[0]; + perf_evsel__parse_sample(evsel, event_sw, &sample_sw); + + sample_sw.period = sample->period; + sample_sw.time = sample->time; + perf_event__synthesize_sample(event_sw, evsel->attr.sample_type, + &sample_sw, false); + build_id__mark_dso_hit(tool, event_sw, &sample_sw, evsel, machine); + return perf_event__repipe(tool, event_sw, &sample_sw, machine); +} + +extern volatile int session_done; + +static void sig_handler(int sig __maybe_unused) +{ + session_done = 1; +} + +static int perf_evsel__check_stype(struct perf_evsel *evsel, + u64 sample_type, const char *sample_msg) +{ + struct perf_event_attr *attr = &evsel->attr; + const char *name = perf_evsel__name(evsel); + + if (!(attr->sample_type & sample_type)) { + pr_err("Samples for %s event do not have %s attribute set.", + name, sample_msg); + return -EINVAL; + } + + return 0; +} + +static int __cmd_inject(struct perf_inject *inject) +{ + struct perf_session *session; + int ret = -EINVAL; + + signal(SIGINT, sig_handler); + + if (inject->build_ids || inject->sched_stat) { + inject->tool.mmap = perf_event__repipe_mmap; + inject->tool.fork = perf_event__repipe_fork; + inject->tool.tracing_data = perf_event__repipe_tracing_data; + } + + session = perf_session__new(inject->input_name, O_RDONLY, false, true, &inject->tool); + if (session == NULL) + return -ENOMEM; + + if (inject->build_ids) { + inject->tool.sample = perf_event__inject_buildid; + } else if (inject->sched_stat) { + struct perf_evsel *evsel; + + inject->tool.ordered_samples = true; + + list_for_each_entry(evsel, &session->evlist->entries, node) { + const char *name = perf_evsel__name(evsel); + + if (!strcmp(name, "sched:sched_switch")) { + if (perf_evsel__check_stype(evsel, PERF_SAMPLE_TID, "TID")) + return -EINVAL; + + evsel->handler.func = perf_inject__sched_switch; + } else if (!strcmp(name, "sched:sched_process_exit")) + evsel->handler.func = perf_inject__sched_process_exit; + else if (!strncmp(name, "sched:sched_stat_", 17)) + evsel->handler.func = perf_inject__sched_stat; + } + } + + if (!inject->pipe_output) + lseek(inject->output, session->header.data_offset, SEEK_SET); + + ret = perf_session__process_events(session, &inject->tool); + + if (!inject->pipe_output) { + session->header.data_size = inject->bytes_written; + perf_session__write_header(session, session->evlist, inject->output, true); + } + + perf_session__delete(session); + + return ret; +} + +int cmd_inject(int argc, const char **argv, const char *prefix __maybe_unused) +{ + struct perf_inject inject = { + .tool = { + .sample = perf_event__repipe_sample, + .mmap = perf_event__repipe, + .comm = perf_event__repipe, + .fork = perf_event__repipe, + .exit = perf_event__repipe, + .lost = perf_event__repipe, + .read = perf_event__repipe_sample, + .throttle = perf_event__repipe, + .unthrottle = perf_event__repipe, + .attr = perf_event__repipe_attr, + .event_type = perf_event__repipe_event_type_synth, + .tracing_data = perf_event__repipe_tracing_data_synth, + .build_id = perf_event__repipe_op2_synth, + }, + .input_name = "-", + .samples = LIST_HEAD_INIT(inject.samples), + }; + const char *output_name = "-"; + const struct option options[] = { + OPT_BOOLEAN('b', "build-ids", &inject.build_ids, + "Inject build-ids into the output stream"), + OPT_STRING('i', "input", &inject.input_name, "file", + "input file name"), + OPT_STRING('o', "output", &output_name, "file", + "output file name"), + OPT_BOOLEAN('s', "sched-stat", &inject.sched_stat, + "Merge sched-stat and sched-switch for getting events " + "where and how long tasks slept"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show build ids, etc)"), + OPT_END() + }; + const char * const inject_usage[] = { + "perf inject [<options>]", + NULL + }; + + argc = parse_options(argc, argv, options, inject_usage, 0); + + /* + * Any (unrecognized) arguments left? + */ + if (argc) + usage_with_options(inject_usage, options); + + if (!strcmp(output_name, "-")) { + inject.pipe_output = 1; + inject.output = STDOUT_FILENO; + } else { + inject.output = open(output_name, O_CREAT | O_WRONLY | O_TRUNC, + S_IRUSR | S_IWUSR); + if (inject.output < 0) { + perror("failed to create output file"); + return -1; + } + } + + if (symbol__init() < 0) + return -1; + + return __cmd_inject(&inject); +} diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c new file mode 100644 index 00000000..46878dac --- /dev/null +++ b/tools/perf/builtin-kmem.c @@ -0,0 +1,781 @@ +#include "builtin.h" +#include "perf.h" + +#include "util/evlist.h" +#include "util/evsel.h" +#include "util/util.h" +#include "util/cache.h" +#include "util/symbol.h" +#include "util/thread.h" +#include "util/header.h" +#include "util/session.h" +#include "util/tool.h" + +#include "util/parse-options.h" +#include "util/trace-event.h" + +#include "util/debug.h" + +#include <linux/rbtree.h> +#include <linux/string.h> + +struct alloc_stat; +typedef int (*sort_fn_t)(struct alloc_stat *, struct alloc_stat *); + +static int alloc_flag; +static int caller_flag; + +static int alloc_lines = -1; +static int caller_lines = -1; + +static bool raw_ip; + +static int *cpunode_map; +static int max_cpu_num; + +struct alloc_stat { + u64 call_site; + u64 ptr; + u64 bytes_req; + u64 bytes_alloc; + u32 hit; + u32 pingpong; + + short alloc_cpu; + + struct rb_node node; +}; + +static struct rb_root root_alloc_stat; +static struct rb_root root_alloc_sorted; +static struct rb_root root_caller_stat; +static struct rb_root root_caller_sorted; + +static unsigned long total_requested, total_allocated; +static unsigned long nr_allocs, nr_cross_allocs; + +#define PATH_SYS_NODE "/sys/devices/system/node" + +static int init_cpunode_map(void) +{ + FILE *fp; + int i, err = -1; + + fp = fopen("/sys/devices/system/cpu/kernel_max", "r"); + if (!fp) { + max_cpu_num = 4096; + return 0; + } + + if (fscanf(fp, "%d", &max_cpu_num) < 1) { + pr_err("Failed to read 'kernel_max' from sysfs"); + goto out_close; + } + + max_cpu_num++; + + cpunode_map = calloc(max_cpu_num, sizeof(int)); + if (!cpunode_map) { + pr_err("%s: calloc failed\n", __func__); + goto out_close; + } + + for (i = 0; i < max_cpu_num; i++) + cpunode_map[i] = -1; + + err = 0; +out_close: + fclose(fp); + return err; +} + +static int setup_cpunode_map(void) +{ + struct dirent *dent1, *dent2; + DIR *dir1, *dir2; + unsigned int cpu, mem; + char buf[PATH_MAX]; + + if (init_cpunode_map()) + return -1; + + dir1 = opendir(PATH_SYS_NODE); + if (!dir1) + return -1; + + while ((dent1 = readdir(dir1)) != NULL) { + if (dent1->d_type != DT_DIR || + sscanf(dent1->d_name, "node%u", &mem) < 1) + continue; + + snprintf(buf, PATH_MAX, "%s/%s", PATH_SYS_NODE, dent1->d_name); + dir2 = opendir(buf); + if (!dir2) + continue; + while ((dent2 = readdir(dir2)) != NULL) { + if (dent2->d_type != DT_LNK || + sscanf(dent2->d_name, "cpu%u", &cpu) < 1) + continue; + cpunode_map[cpu] = mem; + } + closedir(dir2); + } + closedir(dir1); + return 0; +} + +static int insert_alloc_stat(unsigned long call_site, unsigned long ptr, + int bytes_req, int bytes_alloc, int cpu) +{ + struct rb_node **node = &root_alloc_stat.rb_node; + struct rb_node *parent = NULL; + struct alloc_stat *data = NULL; + + while (*node) { + parent = *node; + data = rb_entry(*node, struct alloc_stat, node); + + if (ptr > data->ptr) + node = &(*node)->rb_right; + else if (ptr < data->ptr) + node = &(*node)->rb_left; + else + break; + } + + if (data && data->ptr == ptr) { + data->hit++; + data->bytes_req += bytes_req; + data->bytes_alloc += bytes_alloc; + } else { + data = malloc(sizeof(*data)); + if (!data) { + pr_err("%s: malloc failed\n", __func__); + return -1; + } + data->ptr = ptr; + data->pingpong = 0; + data->hit = 1; + data->bytes_req = bytes_req; + data->bytes_alloc = bytes_alloc; + + rb_link_node(&data->node, parent, node); + rb_insert_color(&data->node, &root_alloc_stat); + } + data->call_site = call_site; + data->alloc_cpu = cpu; + return 0; +} + +static int insert_caller_stat(unsigned long call_site, + int bytes_req, int bytes_alloc) +{ + struct rb_node **node = &root_caller_stat.rb_node; + struct rb_node *parent = NULL; + struct alloc_stat *data = NULL; + + while (*node) { + parent = *node; + data = rb_entry(*node, struct alloc_stat, node); + + if (call_site > data->call_site) + node = &(*node)->rb_right; + else if (call_site < data->call_site) + node = &(*node)->rb_left; + else + break; + } + + if (data && data->call_site == call_site) { + data->hit++; + data->bytes_req += bytes_req; + data->bytes_alloc += bytes_alloc; + } else { + data = malloc(sizeof(*data)); + if (!data) { + pr_err("%s: malloc failed\n", __func__); + return -1; + } + data->call_site = call_site; + data->pingpong = 0; + data->hit = 1; + data->bytes_req = bytes_req; + data->bytes_alloc = bytes_alloc; + + rb_link_node(&data->node, parent, node); + rb_insert_color(&data->node, &root_caller_stat); + } + + return 0; +} + +static int perf_evsel__process_alloc_event(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + unsigned long ptr = perf_evsel__intval(evsel, sample, "ptr"), + call_site = perf_evsel__intval(evsel, sample, "call_site"); + int bytes_req = perf_evsel__intval(evsel, sample, "bytes_req"), + bytes_alloc = perf_evsel__intval(evsel, sample, "bytes_alloc"); + + if (insert_alloc_stat(call_site, ptr, bytes_req, bytes_alloc, sample->cpu) || + insert_caller_stat(call_site, bytes_req, bytes_alloc)) + return -1; + + total_requested += bytes_req; + total_allocated += bytes_alloc; + + nr_allocs++; + return 0; +} + +static int perf_evsel__process_alloc_node_event(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + int ret = perf_evsel__process_alloc_event(evsel, sample); + + if (!ret) { + int node1 = cpunode_map[sample->cpu], + node2 = perf_evsel__intval(evsel, sample, "node"); + + if (node1 != node2) + nr_cross_allocs++; + } + + return ret; +} + +static int ptr_cmp(struct alloc_stat *, struct alloc_stat *); +static int callsite_cmp(struct alloc_stat *, struct alloc_stat *); + +static struct alloc_stat *search_alloc_stat(unsigned long ptr, + unsigned long call_site, + struct rb_root *root, + sort_fn_t sort_fn) +{ + struct rb_node *node = root->rb_node; + struct alloc_stat key = { .ptr = ptr, .call_site = call_site }; + + while (node) { + struct alloc_stat *data; + int cmp; + + data = rb_entry(node, struct alloc_stat, node); + + cmp = sort_fn(&key, data); + if (cmp < 0) + node = node->rb_left; + else if (cmp > 0) + node = node->rb_right; + else + return data; + } + return NULL; +} + +static int perf_evsel__process_free_event(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + unsigned long ptr = perf_evsel__intval(evsel, sample, "ptr"); + struct alloc_stat *s_alloc, *s_caller; + + s_alloc = search_alloc_stat(ptr, 0, &root_alloc_stat, ptr_cmp); + if (!s_alloc) + return 0; + + if ((short)sample->cpu != s_alloc->alloc_cpu) { + s_alloc->pingpong++; + + s_caller = search_alloc_stat(0, s_alloc->call_site, + &root_caller_stat, callsite_cmp); + if (!s_caller) + return -1; + s_caller->pingpong++; + } + s_alloc->alloc_cpu = -1; + + return 0; +} + +typedef int (*tracepoint_handler)(struct perf_evsel *evsel, + struct perf_sample *sample); + +static int process_sample_event(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct thread *thread = machine__findnew_thread(machine, event->ip.pid); + + if (thread == NULL) { + pr_debug("problem processing %d event, skipping it.\n", + event->header.type); + return -1; + } + + dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid); + + if (evsel->handler.func != NULL) { + tracepoint_handler f = evsel->handler.func; + return f(evsel, sample); + } + + return 0; +} + +static struct perf_tool perf_kmem = { + .sample = process_sample_event, + .comm = perf_event__process_comm, + .ordered_samples = true, +}; + +static double fragmentation(unsigned long n_req, unsigned long n_alloc) +{ + if (n_alloc == 0) + return 0.0; + else + return 100.0 - (100.0 * n_req / n_alloc); +} + +static void __print_result(struct rb_root *root, struct perf_session *session, + int n_lines, int is_caller) +{ + struct rb_node *next; + struct machine *machine = &session->machines.host; + + printf("%.102s\n", graph_dotted_line); + printf(" %-34s |", is_caller ? "Callsite": "Alloc Ptr"); + printf(" Total_alloc/Per | Total_req/Per | Hit | Ping-pong | Frag\n"); + printf("%.102s\n", graph_dotted_line); + + next = rb_first(root); + + while (next && n_lines--) { + struct alloc_stat *data = rb_entry(next, struct alloc_stat, + node); + struct symbol *sym = NULL; + struct map *map; + char buf[BUFSIZ]; + u64 addr; + + if (is_caller) { + addr = data->call_site; + if (!raw_ip) + sym = machine__find_kernel_function(machine, addr, &map, NULL); + } else + addr = data->ptr; + + if (sym != NULL) + snprintf(buf, sizeof(buf), "%s+%" PRIx64 "", sym->name, + addr - map->unmap_ip(map, sym->start)); + else + snprintf(buf, sizeof(buf), "%#" PRIx64 "", addr); + printf(" %-34s |", buf); + + printf(" %9llu/%-5lu | %9llu/%-5lu | %8lu | %8lu | %6.3f%%\n", + (unsigned long long)data->bytes_alloc, + (unsigned long)data->bytes_alloc / data->hit, + (unsigned long long)data->bytes_req, + (unsigned long)data->bytes_req / data->hit, + (unsigned long)data->hit, + (unsigned long)data->pingpong, + fragmentation(data->bytes_req, data->bytes_alloc)); + + next = rb_next(next); + } + + if (n_lines == -1) + printf(" ... | ... | ... | ... | ... | ... \n"); + + printf("%.102s\n", graph_dotted_line); +} + +static void print_summary(void) +{ + printf("\nSUMMARY\n=======\n"); + printf("Total bytes requested: %lu\n", total_requested); + printf("Total bytes allocated: %lu\n", total_allocated); + printf("Total bytes wasted on internal fragmentation: %lu\n", + total_allocated - total_requested); + printf("Internal fragmentation: %f%%\n", + fragmentation(total_requested, total_allocated)); + printf("Cross CPU allocations: %lu/%lu\n", nr_cross_allocs, nr_allocs); +} + +static void print_result(struct perf_session *session) +{ + if (caller_flag) + __print_result(&root_caller_sorted, session, caller_lines, 1); + if (alloc_flag) + __print_result(&root_alloc_sorted, session, alloc_lines, 0); + print_summary(); +} + +struct sort_dimension { + const char name[20]; + sort_fn_t cmp; + struct list_head list; +}; + +static LIST_HEAD(caller_sort); +static LIST_HEAD(alloc_sort); + +static void sort_insert(struct rb_root *root, struct alloc_stat *data, + struct list_head *sort_list) +{ + struct rb_node **new = &(root->rb_node); + struct rb_node *parent = NULL; + struct sort_dimension *sort; + + while (*new) { + struct alloc_stat *this; + int cmp = 0; + + this = rb_entry(*new, struct alloc_stat, node); + parent = *new; + + list_for_each_entry(sort, sort_list, list) { + cmp = sort->cmp(data, this); + if (cmp) + break; + } + + if (cmp > 0) + new = &((*new)->rb_left); + else + new = &((*new)->rb_right); + } + + rb_link_node(&data->node, parent, new); + rb_insert_color(&data->node, root); +} + +static void __sort_result(struct rb_root *root, struct rb_root *root_sorted, + struct list_head *sort_list) +{ + struct rb_node *node; + struct alloc_stat *data; + + for (;;) { + node = rb_first(root); + if (!node) + break; + + rb_erase(node, root); + data = rb_entry(node, struct alloc_stat, node); + sort_insert(root_sorted, data, sort_list); + } +} + +static void sort_result(void) +{ + __sort_result(&root_alloc_stat, &root_alloc_sorted, &alloc_sort); + __sort_result(&root_caller_stat, &root_caller_sorted, &caller_sort); +} + +static int __cmd_kmem(void) +{ + int err = -EINVAL; + struct perf_session *session; + const struct perf_evsel_str_handler kmem_tracepoints[] = { + { "kmem:kmalloc", perf_evsel__process_alloc_event, }, + { "kmem:kmem_cache_alloc", perf_evsel__process_alloc_event, }, + { "kmem:kmalloc_node", perf_evsel__process_alloc_node_event, }, + { "kmem:kmem_cache_alloc_node", perf_evsel__process_alloc_node_event, }, + { "kmem:kfree", perf_evsel__process_free_event, }, + { "kmem:kmem_cache_free", perf_evsel__process_free_event, }, + }; + + session = perf_session__new(input_name, O_RDONLY, 0, false, &perf_kmem); + if (session == NULL) + return -ENOMEM; + + if (perf_session__create_kernel_maps(session) < 0) + goto out_delete; + + if (!perf_session__has_traces(session, "kmem record")) + goto out_delete; + + if (perf_session__set_tracepoints_handlers(session, kmem_tracepoints)) { + pr_err("Initializing perf session tracepoint handlers failed\n"); + return -1; + } + + setup_pager(); + err = perf_session__process_events(session, &perf_kmem); + if (err != 0) + goto out_delete; + sort_result(); + print_result(session); +out_delete: + perf_session__delete(session); + return err; +} + +static int ptr_cmp(struct alloc_stat *l, struct alloc_stat *r) +{ + if (l->ptr < r->ptr) + return -1; + else if (l->ptr > r->ptr) + return 1; + return 0; +} + +static struct sort_dimension ptr_sort_dimension = { + .name = "ptr", + .cmp = ptr_cmp, +}; + +static int callsite_cmp(struct alloc_stat *l, struct alloc_stat *r) +{ + if (l->call_site < r->call_site) + return -1; + else if (l->call_site > r->call_site) + return 1; + return 0; +} + +static struct sort_dimension callsite_sort_dimension = { + .name = "callsite", + .cmp = callsite_cmp, +}; + +static int hit_cmp(struct alloc_stat *l, struct alloc_stat *r) +{ + if (l->hit < r->hit) + return -1; + else if (l->hit > r->hit) + return 1; + return 0; +} + +static struct sort_dimension hit_sort_dimension = { + .name = "hit", + .cmp = hit_cmp, +}; + +static int bytes_cmp(struct alloc_stat *l, struct alloc_stat *r) +{ + if (l->bytes_alloc < r->bytes_alloc) + return -1; + else if (l->bytes_alloc > r->bytes_alloc) + return 1; + return 0; +} + +static struct sort_dimension bytes_sort_dimension = { + .name = "bytes", + .cmp = bytes_cmp, +}; + +static int frag_cmp(struct alloc_stat *l, struct alloc_stat *r) +{ + double x, y; + + x = fragmentation(l->bytes_req, l->bytes_alloc); + y = fragmentation(r->bytes_req, r->bytes_alloc); + + if (x < y) + return -1; + else if (x > y) + return 1; + return 0; +} + +static struct sort_dimension frag_sort_dimension = { + .name = "frag", + .cmp = frag_cmp, +}; + +static int pingpong_cmp(struct alloc_stat *l, struct alloc_stat *r) +{ + if (l->pingpong < r->pingpong) + return -1; + else if (l->pingpong > r->pingpong) + return 1; + return 0; +} + +static struct sort_dimension pingpong_sort_dimension = { + .name = "pingpong", + .cmp = pingpong_cmp, +}; + +static struct sort_dimension *avail_sorts[] = { + &ptr_sort_dimension, + &callsite_sort_dimension, + &hit_sort_dimension, + &bytes_sort_dimension, + &frag_sort_dimension, + &pingpong_sort_dimension, +}; + +#define NUM_AVAIL_SORTS ((int)ARRAY_SIZE(avail_sorts)) + +static int sort_dimension__add(const char *tok, struct list_head *list) +{ + struct sort_dimension *sort; + int i; + + for (i = 0; i < NUM_AVAIL_SORTS; i++) { + if (!strcmp(avail_sorts[i]->name, tok)) { + sort = memdup(avail_sorts[i], sizeof(*avail_sorts[i])); + if (!sort) { + pr_err("%s: memdup failed\n", __func__); + return -1; + } + list_add_tail(&sort->list, list); + return 0; + } + } + + return -1; +} + +static int setup_sorting(struct list_head *sort_list, const char *arg) +{ + char *tok; + char *str = strdup(arg); + + if (!str) { + pr_err("%s: strdup failed\n", __func__); + return -1; + } + + while (true) { + tok = strsep(&str, ","); + if (!tok) + break; + if (sort_dimension__add(tok, sort_list) < 0) { + error("Unknown --sort key: '%s'", tok); + free(str); + return -1; + } + } + + free(str); + return 0; +} + +static int parse_sort_opt(const struct option *opt __maybe_unused, + const char *arg, int unset __maybe_unused) +{ + if (!arg) + return -1; + + if (caller_flag > alloc_flag) + return setup_sorting(&caller_sort, arg); + else + return setup_sorting(&alloc_sort, arg); + + return 0; +} + +static int parse_caller_opt(const struct option *opt __maybe_unused, + const char *arg __maybe_unused, + int unset __maybe_unused) +{ + caller_flag = (alloc_flag + 1); + return 0; +} + +static int parse_alloc_opt(const struct option *opt __maybe_unused, + const char *arg __maybe_unused, + int unset __maybe_unused) +{ + alloc_flag = (caller_flag + 1); + return 0; +} + +static int parse_line_opt(const struct option *opt __maybe_unused, + const char *arg, int unset __maybe_unused) +{ + int lines; + + if (!arg) + return -1; + + lines = strtoul(arg, NULL, 10); + + if (caller_flag > alloc_flag) + caller_lines = lines; + else + alloc_lines = lines; + + return 0; +} + +static int __cmd_record(int argc, const char **argv) +{ + const char * const record_args[] = { + "record", "-a", "-R", "-f", "-c", "1", + "-e", "kmem:kmalloc", + "-e", "kmem:kmalloc_node", + "-e", "kmem:kfree", + "-e", "kmem:kmem_cache_alloc", + "-e", "kmem:kmem_cache_alloc_node", + "-e", "kmem:kmem_cache_free", + }; + unsigned int rec_argc, i, j; + const char **rec_argv; + + rec_argc = ARRAY_SIZE(record_args) + argc - 1; + rec_argv = calloc(rec_argc + 1, sizeof(char *)); + + if (rec_argv == NULL) + return -ENOMEM; + + for (i = 0; i < ARRAY_SIZE(record_args); i++) + rec_argv[i] = strdup(record_args[i]); + + for (j = 1; j < (unsigned int)argc; j++, i++) + rec_argv[i] = argv[j]; + + return cmd_record(i, rec_argv, NULL); +} + +int cmd_kmem(int argc, const char **argv, const char *prefix __maybe_unused) +{ + const char * const default_sort_order = "frag,hit,bytes"; + const struct option kmem_options[] = { + OPT_STRING('i', "input", &input_name, "file", "input file name"), + OPT_CALLBACK_NOOPT(0, "caller", NULL, NULL, + "show per-callsite statistics", parse_caller_opt), + OPT_CALLBACK_NOOPT(0, "alloc", NULL, NULL, + "show per-allocation statistics", parse_alloc_opt), + OPT_CALLBACK('s', "sort", NULL, "key[,key2...]", + "sort by keys: ptr, call_site, bytes, hit, pingpong, frag", + parse_sort_opt), + OPT_CALLBACK('l', "line", NULL, "num", "show n lines", parse_line_opt), + OPT_BOOLEAN(0, "raw-ip", &raw_ip, "show raw ip instead of symbol"), + OPT_END() + }; + const char * const kmem_usage[] = { + "perf kmem [<options>] {record|stat}", + NULL + }; + argc = parse_options(argc, argv, kmem_options, kmem_usage, 0); + + if (!argc) + usage_with_options(kmem_usage, kmem_options); + + symbol__init(); + + if (!strncmp(argv[0], "rec", 3)) { + return __cmd_record(argc, argv); + } else if (!strcmp(argv[0], "stat")) { + if (setup_cpunode_map()) + return -1; + + if (list_empty(&caller_sort)) + setup_sorting(&caller_sort, default_sort_order); + if (list_empty(&alloc_sort)) + setup_sorting(&alloc_sort, default_sort_order); + + return __cmd_kmem(); + } else + usage_with_options(kmem_usage, kmem_options); + + return 0; +} + diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c new file mode 100644 index 00000000..37a769d7 --- /dev/null +++ b/tools/perf/builtin-kvm.c @@ -0,0 +1,1047 @@ +#include "builtin.h" +#include "perf.h" + +#include "util/evsel.h" +#include "util/util.h" +#include "util/cache.h" +#include "util/symbol.h" +#include "util/thread.h" +#include "util/header.h" +#include "util/session.h" + +#include "util/parse-options.h" +#include "util/trace-event.h" +#include "util/debug.h" +#include "util/debugfs.h" +#include "util/tool.h" +#include "util/stat.h" + +#include <sys/prctl.h> + +#include <semaphore.h> +#include <pthread.h> +#include <math.h> + +#if defined(__i386__) || defined(__x86_64__) +#include <asm/svm.h> +#include <asm/vmx.h> +#include <asm/kvm.h> + +struct event_key { + #define INVALID_KEY (~0ULL) + u64 key; + int info; +}; + +struct kvm_event_stats { + u64 time; + struct stats stats; +}; + +struct kvm_event { + struct list_head hash_entry; + struct rb_node rb; + + struct event_key key; + + struct kvm_event_stats total; + + #define DEFAULT_VCPU_NUM 8 + int max_vcpu; + struct kvm_event_stats *vcpu; +}; + +typedef int (*key_cmp_fun)(struct kvm_event*, struct kvm_event*, int); + +struct kvm_event_key { + const char *name; + key_cmp_fun key; +}; + + +struct perf_kvm_stat; + +struct kvm_events_ops { + bool (*is_begin_event)(struct perf_evsel *evsel, + struct perf_sample *sample, + struct event_key *key); + bool (*is_end_event)(struct perf_evsel *evsel, + struct perf_sample *sample, struct event_key *key); + void (*decode_key)(struct perf_kvm_stat *kvm, struct event_key *key, + char decode[20]); + const char *name; +}; + +struct exit_reasons_table { + unsigned long exit_code; + const char *reason; +}; + +#define EVENTS_BITS 12 +#define EVENTS_CACHE_SIZE (1UL << EVENTS_BITS) + +struct perf_kvm_stat { + struct perf_tool tool; + struct perf_session *session; + + const char *file_name; + const char *report_event; + const char *sort_key; + int trace_vcpu; + + struct exit_reasons_table *exit_reasons; + int exit_reasons_size; + const char *exit_reasons_isa; + + struct kvm_events_ops *events_ops; + key_cmp_fun compare; + struct list_head kvm_events_cache[EVENTS_CACHE_SIZE]; + u64 total_time; + u64 total_count; + + struct rb_root result; +}; + + +static void exit_event_get_key(struct perf_evsel *evsel, + struct perf_sample *sample, + struct event_key *key) +{ + key->info = 0; + key->key = perf_evsel__intval(evsel, sample, "exit_reason"); +} + +static bool kvm_exit_event(struct perf_evsel *evsel) +{ + return !strcmp(evsel->name, "kvm:kvm_exit"); +} + +static bool exit_event_begin(struct perf_evsel *evsel, + struct perf_sample *sample, struct event_key *key) +{ + if (kvm_exit_event(evsel)) { + exit_event_get_key(evsel, sample, key); + return true; + } + + return false; +} + +static bool kvm_entry_event(struct perf_evsel *evsel) +{ + return !strcmp(evsel->name, "kvm:kvm_entry"); +} + +static bool exit_event_end(struct perf_evsel *evsel, + struct perf_sample *sample __maybe_unused, + struct event_key *key __maybe_unused) +{ + return kvm_entry_event(evsel); +} + +static struct exit_reasons_table vmx_exit_reasons[] = { + VMX_EXIT_REASONS +}; + +static struct exit_reasons_table svm_exit_reasons[] = { + SVM_EXIT_REASONS +}; + +static const char *get_exit_reason(struct perf_kvm_stat *kvm, u64 exit_code) +{ + int i = kvm->exit_reasons_size; + struct exit_reasons_table *tbl = kvm->exit_reasons; + + while (i--) { + if (tbl->exit_code == exit_code) + return tbl->reason; + tbl++; + } + + pr_err("unknown kvm exit code:%lld on %s\n", + (unsigned long long)exit_code, kvm->exit_reasons_isa); + return "UNKNOWN"; +} + +static void exit_event_decode_key(struct perf_kvm_stat *kvm, + struct event_key *key, + char decode[20]) +{ + const char *exit_reason = get_exit_reason(kvm, key->key); + + scnprintf(decode, 20, "%s", exit_reason); +} + +static struct kvm_events_ops exit_events = { + .is_begin_event = exit_event_begin, + .is_end_event = exit_event_end, + .decode_key = exit_event_decode_key, + .name = "VM-EXIT" +}; + +/* + * For the mmio events, we treat: + * the time of MMIO write: kvm_mmio(KVM_TRACE_MMIO_WRITE...) -> kvm_entry + * the time of MMIO read: kvm_exit -> kvm_mmio(KVM_TRACE_MMIO_READ...). + */ +static void mmio_event_get_key(struct perf_evsel *evsel, struct perf_sample *sample, + struct event_key *key) +{ + key->key = perf_evsel__intval(evsel, sample, "gpa"); + key->info = perf_evsel__intval(evsel, sample, "type"); +} + +#define KVM_TRACE_MMIO_READ_UNSATISFIED 0 +#define KVM_TRACE_MMIO_READ 1 +#define KVM_TRACE_MMIO_WRITE 2 + +static bool mmio_event_begin(struct perf_evsel *evsel, + struct perf_sample *sample, struct event_key *key) +{ + /* MMIO read begin event in kernel. */ + if (kvm_exit_event(evsel)) + return true; + + /* MMIO write begin event in kernel. */ + if (!strcmp(evsel->name, "kvm:kvm_mmio") && + perf_evsel__intval(evsel, sample, "type") == KVM_TRACE_MMIO_WRITE) { + mmio_event_get_key(evsel, sample, key); + return true; + } + + return false; +} + +static bool mmio_event_end(struct perf_evsel *evsel, struct perf_sample *sample, + struct event_key *key) +{ + /* MMIO write end event in kernel. */ + if (kvm_entry_event(evsel)) + return true; + + /* MMIO read end event in kernel.*/ + if (!strcmp(evsel->name, "kvm:kvm_mmio") && + perf_evsel__intval(evsel, sample, "type") == KVM_TRACE_MMIO_READ) { + mmio_event_get_key(evsel, sample, key); + return true; + } + + return false; +} + +static void mmio_event_decode_key(struct perf_kvm_stat *kvm __maybe_unused, + struct event_key *key, + char decode[20]) +{ + scnprintf(decode, 20, "%#lx:%s", (unsigned long)key->key, + key->info == KVM_TRACE_MMIO_WRITE ? "W" : "R"); +} + +static struct kvm_events_ops mmio_events = { + .is_begin_event = mmio_event_begin, + .is_end_event = mmio_event_end, + .decode_key = mmio_event_decode_key, + .name = "MMIO Access" +}; + + /* The time of emulation pio access is from kvm_pio to kvm_entry. */ +static void ioport_event_get_key(struct perf_evsel *evsel, + struct perf_sample *sample, + struct event_key *key) +{ + key->key = perf_evsel__intval(evsel, sample, "port"); + key->info = perf_evsel__intval(evsel, sample, "rw"); +} + +static bool ioport_event_begin(struct perf_evsel *evsel, + struct perf_sample *sample, + struct event_key *key) +{ + if (!strcmp(evsel->name, "kvm:kvm_pio")) { + ioport_event_get_key(evsel, sample, key); + return true; + } + + return false; +} + +static bool ioport_event_end(struct perf_evsel *evsel, + struct perf_sample *sample __maybe_unused, + struct event_key *key __maybe_unused) +{ + return kvm_entry_event(evsel); +} + +static void ioport_event_decode_key(struct perf_kvm_stat *kvm __maybe_unused, + struct event_key *key, + char decode[20]) +{ + scnprintf(decode, 20, "%#llx:%s", (unsigned long long)key->key, + key->info ? "POUT" : "PIN"); +} + +static struct kvm_events_ops ioport_events = { + .is_begin_event = ioport_event_begin, + .is_end_event = ioport_event_end, + .decode_key = ioport_event_decode_key, + .name = "IO Port Access" +}; + +static bool register_kvm_events_ops(struct perf_kvm_stat *kvm) +{ + bool ret = true; + + if (!strcmp(kvm->report_event, "vmexit")) + kvm->events_ops = &exit_events; + else if (!strcmp(kvm->report_event, "mmio")) + kvm->events_ops = &mmio_events; + else if (!strcmp(kvm->report_event, "ioport")) + kvm->events_ops = &ioport_events; + else { + pr_err("Unknown report event:%s\n", kvm->report_event); + ret = false; + } + + return ret; +} + +struct vcpu_event_record { + int vcpu_id; + u64 start_time; + struct kvm_event *last_event; +}; + + +static void init_kvm_event_record(struct perf_kvm_stat *kvm) +{ + unsigned int i; + + for (i = 0; i < EVENTS_CACHE_SIZE; i++) + INIT_LIST_HEAD(&kvm->kvm_events_cache[i]); +} + +static int kvm_events_hash_fn(u64 key) +{ + return key & (EVENTS_CACHE_SIZE - 1); +} + +static bool kvm_event_expand(struct kvm_event *event, int vcpu_id) +{ + int old_max_vcpu = event->max_vcpu; + + if (vcpu_id < event->max_vcpu) + return true; + + while (event->max_vcpu <= vcpu_id) + event->max_vcpu += DEFAULT_VCPU_NUM; + + event->vcpu = realloc(event->vcpu, + event->max_vcpu * sizeof(*event->vcpu)); + if (!event->vcpu) { + pr_err("Not enough memory\n"); + return false; + } + + memset(event->vcpu + old_max_vcpu, 0, + (event->max_vcpu - old_max_vcpu) * sizeof(*event->vcpu)); + return true; +} + +static struct kvm_event *kvm_alloc_init_event(struct event_key *key) +{ + struct kvm_event *event; + + event = zalloc(sizeof(*event)); + if (!event) { + pr_err("Not enough memory\n"); + return NULL; + } + + event->key = *key; + return event; +} + +static struct kvm_event *find_create_kvm_event(struct perf_kvm_stat *kvm, + struct event_key *key) +{ + struct kvm_event *event; + struct list_head *head; + + BUG_ON(key->key == INVALID_KEY); + + head = &kvm->kvm_events_cache[kvm_events_hash_fn(key->key)]; + list_for_each_entry(event, head, hash_entry) { + if (event->key.key == key->key && event->key.info == key->info) + return event; + } + + event = kvm_alloc_init_event(key); + if (!event) + return NULL; + + list_add(&event->hash_entry, head); + return event; +} + +static bool handle_begin_event(struct perf_kvm_stat *kvm, + struct vcpu_event_record *vcpu_record, + struct event_key *key, u64 timestamp) +{ + struct kvm_event *event = NULL; + + if (key->key != INVALID_KEY) + event = find_create_kvm_event(kvm, key); + + vcpu_record->last_event = event; + vcpu_record->start_time = timestamp; + return true; +} + +static void +kvm_update_event_stats(struct kvm_event_stats *kvm_stats, u64 time_diff) +{ + kvm_stats->time += time_diff; + update_stats(&kvm_stats->stats, time_diff); +} + +static double kvm_event_rel_stddev(int vcpu_id, struct kvm_event *event) +{ + struct kvm_event_stats *kvm_stats = &event->total; + + if (vcpu_id != -1) + kvm_stats = &event->vcpu[vcpu_id]; + + return rel_stddev_stats(stddev_stats(&kvm_stats->stats), + avg_stats(&kvm_stats->stats)); +} + +static bool update_kvm_event(struct kvm_event *event, int vcpu_id, + u64 time_diff) +{ + if (vcpu_id == -1) { + kvm_update_event_stats(&event->total, time_diff); + return true; + } + + if (!kvm_event_expand(event, vcpu_id)) + return false; + + kvm_update_event_stats(&event->vcpu[vcpu_id], time_diff); + return true; +} + +static bool handle_end_event(struct perf_kvm_stat *kvm, + struct vcpu_event_record *vcpu_record, + struct event_key *key, + u64 timestamp) +{ + struct kvm_event *event; + u64 time_begin, time_diff; + int vcpu; + + if (kvm->trace_vcpu == -1) + vcpu = -1; + else + vcpu = vcpu_record->vcpu_id; + + event = vcpu_record->last_event; + time_begin = vcpu_record->start_time; + + /* The begin event is not caught. */ + if (!time_begin) + return true; + + /* + * In some case, the 'begin event' only records the start timestamp, + * the actual event is recognized in the 'end event' (e.g. mmio-event). + */ + + /* Both begin and end events did not get the key. */ + if (!event && key->key == INVALID_KEY) + return true; + + if (!event) + event = find_create_kvm_event(kvm, key); + + if (!event) + return false; + + vcpu_record->last_event = NULL; + vcpu_record->start_time = 0; + + BUG_ON(timestamp < time_begin); + + time_diff = timestamp - time_begin; + return update_kvm_event(event, vcpu, time_diff); +} + +static +struct vcpu_event_record *per_vcpu_record(struct thread *thread, + struct perf_evsel *evsel, + struct perf_sample *sample) +{ + /* Only kvm_entry records vcpu id. */ + if (!thread->priv && kvm_entry_event(evsel)) { + struct vcpu_event_record *vcpu_record; + + vcpu_record = zalloc(sizeof(*vcpu_record)); + if (!vcpu_record) { + pr_err("%s: Not enough memory\n", __func__); + return NULL; + } + + vcpu_record->vcpu_id = perf_evsel__intval(evsel, sample, "vcpu_id"); + thread->priv = vcpu_record; + } + + return thread->priv; +} + +static bool handle_kvm_event(struct perf_kvm_stat *kvm, + struct thread *thread, + struct perf_evsel *evsel, + struct perf_sample *sample) +{ + struct vcpu_event_record *vcpu_record; + struct event_key key = {.key = INVALID_KEY}; + + vcpu_record = per_vcpu_record(thread, evsel, sample); + if (!vcpu_record) + return true; + + /* only process events for vcpus user cares about */ + if ((kvm->trace_vcpu != -1) && + (kvm->trace_vcpu != vcpu_record->vcpu_id)) + return true; + + if (kvm->events_ops->is_begin_event(evsel, sample, &key)) + return handle_begin_event(kvm, vcpu_record, &key, sample->time); + + if (kvm->events_ops->is_end_event(evsel, sample, &key)) + return handle_end_event(kvm, vcpu_record, &key, sample->time); + + return true; +} + +#define GET_EVENT_KEY(func, field) \ +static u64 get_event_ ##func(struct kvm_event *event, int vcpu) \ +{ \ + if (vcpu == -1) \ + return event->total.field; \ + \ + if (vcpu >= event->max_vcpu) \ + return 0; \ + \ + return event->vcpu[vcpu].field; \ +} + +#define COMPARE_EVENT_KEY(func, field) \ +GET_EVENT_KEY(func, field) \ +static int compare_kvm_event_ ## func(struct kvm_event *one, \ + struct kvm_event *two, int vcpu)\ +{ \ + return get_event_ ##func(one, vcpu) > \ + get_event_ ##func(two, vcpu); \ +} + +GET_EVENT_KEY(time, time); +COMPARE_EVENT_KEY(count, stats.n); +COMPARE_EVENT_KEY(mean, stats.mean); + +#define DEF_SORT_NAME_KEY(name, compare_key) \ + { #name, compare_kvm_event_ ## compare_key } + +static struct kvm_event_key keys[] = { + DEF_SORT_NAME_KEY(sample, count), + DEF_SORT_NAME_KEY(time, mean), + { NULL, NULL } +}; + +static bool select_key(struct perf_kvm_stat *kvm) +{ + int i; + + for (i = 0; keys[i].name; i++) { + if (!strcmp(keys[i].name, kvm->sort_key)) { + kvm->compare = keys[i].key; + return true; + } + } + + pr_err("Unknown compare key:%s\n", kvm->sort_key); + return false; +} + +static void insert_to_result(struct rb_root *result, struct kvm_event *event, + key_cmp_fun bigger, int vcpu) +{ + struct rb_node **rb = &result->rb_node; + struct rb_node *parent = NULL; + struct kvm_event *p; + + while (*rb) { + p = container_of(*rb, struct kvm_event, rb); + parent = *rb; + + if (bigger(event, p, vcpu)) + rb = &(*rb)->rb_left; + else + rb = &(*rb)->rb_right; + } + + rb_link_node(&event->rb, parent, rb); + rb_insert_color(&event->rb, result); +} + +static void +update_total_count(struct perf_kvm_stat *kvm, struct kvm_event *event) +{ + int vcpu = kvm->trace_vcpu; + + kvm->total_count += get_event_count(event, vcpu); + kvm->total_time += get_event_time(event, vcpu); +} + +static bool event_is_valid(struct kvm_event *event, int vcpu) +{ + return !!get_event_count(event, vcpu); +} + +static void sort_result(struct perf_kvm_stat *kvm) +{ + unsigned int i; + int vcpu = kvm->trace_vcpu; + struct kvm_event *event; + + for (i = 0; i < EVENTS_CACHE_SIZE; i++) { + list_for_each_entry(event, &kvm->kvm_events_cache[i], hash_entry) { + if (event_is_valid(event, vcpu)) { + update_total_count(kvm, event); + insert_to_result(&kvm->result, event, + kvm->compare, vcpu); + } + } + } +} + +/* returns left most element of result, and erase it */ +static struct kvm_event *pop_from_result(struct rb_root *result) +{ + struct rb_node *node = rb_first(result); + + if (!node) + return NULL; + + rb_erase(node, result); + return container_of(node, struct kvm_event, rb); +} + +static void print_vcpu_info(int vcpu) +{ + pr_info("Analyze events for "); + + if (vcpu == -1) + pr_info("all VCPUs:\n\n"); + else + pr_info("VCPU %d:\n\n", vcpu); +} + +static void print_result(struct perf_kvm_stat *kvm) +{ + char decode[20]; + struct kvm_event *event; + int vcpu = kvm->trace_vcpu; + + pr_info("\n\n"); + print_vcpu_info(vcpu); + pr_info("%20s ", kvm->events_ops->name); + pr_info("%10s ", "Samples"); + pr_info("%9s ", "Samples%"); + + pr_info("%9s ", "Time%"); + pr_info("%16s ", "Avg time"); + pr_info("\n\n"); + + while ((event = pop_from_result(&kvm->result))) { + u64 ecount, etime; + + ecount = get_event_count(event, vcpu); + etime = get_event_time(event, vcpu); + + kvm->events_ops->decode_key(kvm, &event->key, decode); + pr_info("%20s ", decode); + pr_info("%10llu ", (unsigned long long)ecount); + pr_info("%8.2f%% ", (double)ecount / kvm->total_count * 100); + pr_info("%8.2f%% ", (double)etime / kvm->total_time * 100); + pr_info("%9.2fus ( +-%7.2f%% )", (double)etime / ecount/1e3, + kvm_event_rel_stddev(vcpu, event)); + pr_info("\n"); + } + + pr_info("\nTotal Samples:%" PRIu64 ", Total events handled time:%.2fus.\n\n", + kvm->total_count, kvm->total_time / 1e3); +} + +static int process_sample_event(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct thread *thread = machine__findnew_thread(machine, sample->tid); + struct perf_kvm_stat *kvm = container_of(tool, struct perf_kvm_stat, + tool); + + if (thread == NULL) { + pr_debug("problem processing %d event, skipping it.\n", + event->header.type); + return -1; + } + + if (!handle_kvm_event(kvm, thread, evsel, sample)) + return -1; + + return 0; +} + +static int get_cpu_isa(struct perf_session *session) +{ + char *cpuid = session->header.env.cpuid; + int isa; + + if (strstr(cpuid, "Intel")) + isa = 1; + else if (strstr(cpuid, "AMD")) + isa = 0; + else { + pr_err("CPU %s is not supported.\n", cpuid); + isa = -ENOTSUP; + } + + return isa; +} + +static int read_events(struct perf_kvm_stat *kvm) +{ + int ret; + + struct perf_tool eops = { + .sample = process_sample_event, + .comm = perf_event__process_comm, + .ordered_samples = true, + }; + + kvm->tool = eops; + kvm->session = perf_session__new(kvm->file_name, O_RDONLY, 0, false, + &kvm->tool); + if (!kvm->session) { + pr_err("Initializing perf session failed\n"); + return -EINVAL; + } + + if (!perf_session__has_traces(kvm->session, "kvm record")) + return -EINVAL; + + /* + * Do not use 'isa' recorded in kvm_exit tracepoint since it is not + * traced in the old kernel. + */ + ret = get_cpu_isa(kvm->session); + + if (ret < 0) + return ret; + + if (ret == 1) { + kvm->exit_reasons = vmx_exit_reasons; + kvm->exit_reasons_size = ARRAY_SIZE(vmx_exit_reasons); + kvm->exit_reasons_isa = "VMX"; + } + + return perf_session__process_events(kvm->session, &kvm->tool); +} + +static bool verify_vcpu(int vcpu) +{ + if (vcpu != -1 && vcpu < 0) { + pr_err("Invalid vcpu:%d.\n", vcpu); + return false; + } + + return true; +} + +static int kvm_events_report_vcpu(struct perf_kvm_stat *kvm) +{ + int ret = -EINVAL; + int vcpu = kvm->trace_vcpu; + + if (!verify_vcpu(vcpu)) + goto exit; + + if (!select_key(kvm)) + goto exit; + + if (!register_kvm_events_ops(kvm)) + goto exit; + + init_kvm_event_record(kvm); + setup_pager(); + + ret = read_events(kvm); + if (ret) + goto exit; + + sort_result(kvm); + print_result(kvm); + +exit: + return ret; +} + +static const char * const record_args[] = { + "record", + "-R", + "-f", + "-m", "1024", + "-c", "1", + "-e", "kvm:kvm_entry", + "-e", "kvm:kvm_exit", + "-e", "kvm:kvm_mmio", + "-e", "kvm:kvm_pio", +}; + +#define STRDUP_FAIL_EXIT(s) \ + ({ char *_p; \ + _p = strdup(s); \ + if (!_p) \ + return -ENOMEM; \ + _p; \ + }) + +static int +kvm_events_record(struct perf_kvm_stat *kvm, int argc, const char **argv) +{ + unsigned int rec_argc, i, j; + const char **rec_argv; + + rec_argc = ARRAY_SIZE(record_args) + argc + 2; + rec_argv = calloc(rec_argc + 1, sizeof(char *)); + + if (rec_argv == NULL) + return -ENOMEM; + + for (i = 0; i < ARRAY_SIZE(record_args); i++) + rec_argv[i] = STRDUP_FAIL_EXIT(record_args[i]); + + rec_argv[i++] = STRDUP_FAIL_EXIT("-o"); + rec_argv[i++] = STRDUP_FAIL_EXIT(kvm->file_name); + + for (j = 1; j < (unsigned int)argc; j++, i++) + rec_argv[i] = argv[j]; + + return cmd_record(i, rec_argv, NULL); +} + +static int +kvm_events_report(struct perf_kvm_stat *kvm, int argc, const char **argv) +{ + const struct option kvm_events_report_options[] = { + OPT_STRING(0, "event", &kvm->report_event, "report event", + "event for reporting: vmexit, mmio, ioport"), + OPT_INTEGER(0, "vcpu", &kvm->trace_vcpu, + "vcpu id to report"), + OPT_STRING('k', "key", &kvm->sort_key, "sort-key", + "key for sorting: sample(sort by samples number)" + " time (sort by avg time)"), + OPT_END() + }; + + const char * const kvm_events_report_usage[] = { + "perf kvm stat report [<options>]", + NULL + }; + + symbol__init(); + + if (argc) { + argc = parse_options(argc, argv, + kvm_events_report_options, + kvm_events_report_usage, 0); + if (argc) + usage_with_options(kvm_events_report_usage, + kvm_events_report_options); + } + + return kvm_events_report_vcpu(kvm); +} + +static void print_kvm_stat_usage(void) +{ + printf("Usage: perf kvm stat <command>\n\n"); + + printf("# Available commands:\n"); + printf("\trecord: record kvm events\n"); + printf("\treport: report statistical data of kvm events\n"); + + printf("\nOtherwise, it is the alias of 'perf stat':\n"); +} + +static int kvm_cmd_stat(const char *file_name, int argc, const char **argv) +{ + struct perf_kvm_stat kvm = { + .file_name = file_name, + + .trace_vcpu = -1, + .report_event = "vmexit", + .sort_key = "sample", + + .exit_reasons = svm_exit_reasons, + .exit_reasons_size = ARRAY_SIZE(svm_exit_reasons), + .exit_reasons_isa = "SVM", + }; + + if (argc == 1) { + print_kvm_stat_usage(); + goto perf_stat; + } + + if (!strncmp(argv[1], "rec", 3)) + return kvm_events_record(&kvm, argc - 1, argv + 1); + + if (!strncmp(argv[1], "rep", 3)) + return kvm_events_report(&kvm, argc - 1 , argv + 1); + +perf_stat: + return cmd_stat(argc, argv, NULL); +} +#endif + +static int __cmd_record(const char *file_name, int argc, const char **argv) +{ + int rec_argc, i = 0, j; + const char **rec_argv; + + rec_argc = argc + 2; + rec_argv = calloc(rec_argc + 1, sizeof(char *)); + rec_argv[i++] = strdup("record"); + rec_argv[i++] = strdup("-o"); + rec_argv[i++] = strdup(file_name); + for (j = 1; j < argc; j++, i++) + rec_argv[i] = argv[j]; + + BUG_ON(i != rec_argc); + + return cmd_record(i, rec_argv, NULL); +} + +static int __cmd_report(const char *file_name, int argc, const char **argv) +{ + int rec_argc, i = 0, j; + const char **rec_argv; + + rec_argc = argc + 2; + rec_argv = calloc(rec_argc + 1, sizeof(char *)); + rec_argv[i++] = strdup("report"); + rec_argv[i++] = strdup("-i"); + rec_argv[i++] = strdup(file_name); + for (j = 1; j < argc; j++, i++) + rec_argv[i] = argv[j]; + + BUG_ON(i != rec_argc); + + return cmd_report(i, rec_argv, NULL); +} + +static int +__cmd_buildid_list(const char *file_name, int argc, const char **argv) +{ + int rec_argc, i = 0, j; + const char **rec_argv; + + rec_argc = argc + 2; + rec_argv = calloc(rec_argc + 1, sizeof(char *)); + rec_argv[i++] = strdup("buildid-list"); + rec_argv[i++] = strdup("-i"); + rec_argv[i++] = strdup(file_name); + for (j = 1; j < argc; j++, i++) + rec_argv[i] = argv[j]; + + BUG_ON(i != rec_argc); + + return cmd_buildid_list(i, rec_argv, NULL); +} + +int cmd_kvm(int argc, const char **argv, const char *prefix __maybe_unused) +{ + const char *file_name = NULL; + const struct option kvm_options[] = { + OPT_STRING('i', "input", &file_name, "file", + "Input file name"), + OPT_STRING('o', "output", &file_name, "file", + "Output file name"), + OPT_BOOLEAN(0, "guest", &perf_guest, + "Collect guest os data"), + OPT_BOOLEAN(0, "host", &perf_host, + "Collect host os data"), + OPT_STRING(0, "guestmount", &symbol_conf.guestmount, "directory", + "guest mount directory under which every guest os" + " instance has a subdir"), + OPT_STRING(0, "guestvmlinux", &symbol_conf.default_guest_vmlinux_name, + "file", "file saving guest os vmlinux"), + OPT_STRING(0, "guestkallsyms", &symbol_conf.default_guest_kallsyms, + "file", "file saving guest os /proc/kallsyms"), + OPT_STRING(0, "guestmodules", &symbol_conf.default_guest_modules, + "file", "file saving guest os /proc/modules"), + OPT_END() + }; + + + const char * const kvm_usage[] = { + "perf kvm [<options>] {top|record|report|diff|buildid-list|stat}", + NULL + }; + + perf_host = 0; + perf_guest = 1; + + argc = parse_options(argc, argv, kvm_options, kvm_usage, + PARSE_OPT_STOP_AT_NON_OPTION); + if (!argc) + usage_with_options(kvm_usage, kvm_options); + + if (!perf_host) + perf_guest = 1; + + if (!file_name) { + if (perf_host && !perf_guest) + file_name = strdup("perf.data.host"); + else if (!perf_host && perf_guest) + file_name = strdup("perf.data.guest"); + else + file_name = strdup("perf.data.kvm"); + + if (!file_name) { + pr_err("Failed to allocate memory for filename\n"); + return -ENOMEM; + } + } + + if (!strncmp(argv[0], "rec", 3)) + return __cmd_record(file_name, argc, argv); + else if (!strncmp(argv[0], "rep", 3)) + return __cmd_report(file_name, argc, argv); + else if (!strncmp(argv[0], "diff", 4)) + return cmd_diff(argc, argv, NULL); + else if (!strncmp(argv[0], "top", 3)) + return cmd_top(argc, argv, NULL); + else if (!strncmp(argv[0], "buildid-list", 12)) + return __cmd_buildid_list(file_name, argc, argv); +#if defined(__i386__) || defined(__x86_64__) + else if (!strncmp(argv[0], "stat", 4)) + return kvm_cmd_stat(file_name, argc, argv); +#endif + else + usage_with_options(kvm_usage, kvm_options); + + return 0; +} diff --git a/tools/perf/builtin-list.c b/tools/perf/builtin-list.c new file mode 100644 index 00000000..1948eceb --- /dev/null +++ b/tools/perf/builtin-list.c @@ -0,0 +1,62 @@ +/* + * 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> + * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com> + */ +#include "builtin.h" + +#include "perf.h" + +#include "util/parse-events.h" +#include "util/cache.h" + +int cmd_list(int argc, const char **argv, const char *prefix __maybe_unused) +{ + setup_pager(); + + if (argc == 1) + print_events(NULL, false); + else { + int i; + + for (i = 1; i < argc; ++i) { + if (i > 2) + putchar('\n'); + if (strncmp(argv[i], "tracepoint", 10) == 0) + print_tracepoint_events(NULL, NULL, false); + else if (strcmp(argv[i], "hw") == 0 || + strcmp(argv[i], "hardware") == 0) + print_events_type(PERF_TYPE_HARDWARE); + else if (strcmp(argv[i], "sw") == 0 || + strcmp(argv[i], "software") == 0) + print_events_type(PERF_TYPE_SOFTWARE); + else if (strcmp(argv[i], "cache") == 0 || + strcmp(argv[i], "hwcache") == 0) + print_hwcache_events(NULL, false); + else if (strcmp(argv[i], "--raw-dump") == 0) + print_events(NULL, true); + else { + char *sep = strchr(argv[i], ':'), *s; + int sep_idx; + + if (sep == NULL) { + print_events(argv[i], false); + continue; + } + sep_idx = sep - argv[i]; + s = strdup(argv[i]); + if (s == NULL) + return -1; + + s[sep_idx] = '\0'; + print_tracepoint_events(s, s + sep_idx + 1, false); + free(s); + } + } + } + return 0; +} diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c new file mode 100644 index 00000000..42583006 --- /dev/null +++ b/tools/perf/builtin-lock.c @@ -0,0 +1,997 @@ +#include "builtin.h" +#include "perf.h" + +#include "util/evlist.h" +#include "util/evsel.h" +#include "util/util.h" +#include "util/cache.h" +#include "util/symbol.h" +#include "util/thread.h" +#include "util/header.h" + +#include "util/parse-options.h" +#include "util/trace-event.h" + +#include "util/debug.h" +#include "util/session.h" +#include "util/tool.h" + +#include <sys/types.h> +#include <sys/prctl.h> +#include <semaphore.h> +#include <pthread.h> +#include <math.h> +#include <limits.h> + +#include <linux/list.h> +#include <linux/hash.h> + +static struct perf_session *session; + +/* based on kernel/lockdep.c */ +#define LOCKHASH_BITS 12 +#define LOCKHASH_SIZE (1UL << LOCKHASH_BITS) + +static struct list_head lockhash_table[LOCKHASH_SIZE]; + +#define __lockhashfn(key) hash_long((unsigned long)key, LOCKHASH_BITS) +#define lockhashentry(key) (lockhash_table + __lockhashfn((key))) + +struct lock_stat { + struct list_head hash_entry; + struct rb_node rb; /* used for sorting */ + + /* + * FIXME: perf_evsel__intval() returns u64, + * so address of lockdep_map should be dealed as 64bit. + * Is there more better solution? + */ + void *addr; /* address of lockdep_map, used as ID */ + char *name; /* for strcpy(), we cannot use const */ + + unsigned int nr_acquire; + unsigned int nr_acquired; + unsigned int nr_contended; + unsigned int nr_release; + + unsigned int nr_readlock; + unsigned int nr_trylock; + /* these times are in nano sec. */ + u64 wait_time_total; + u64 wait_time_min; + u64 wait_time_max; + + int discard; /* flag of blacklist */ +}; + +/* + * States of lock_seq_stat + * + * UNINITIALIZED is required for detecting first event of acquire. + * As the nature of lock events, there is no guarantee + * that the first event for the locks are acquire, + * it can be acquired, contended or release. + */ +#define SEQ_STATE_UNINITIALIZED 0 /* initial state */ +#define SEQ_STATE_RELEASED 1 +#define SEQ_STATE_ACQUIRING 2 +#define SEQ_STATE_ACQUIRED 3 +#define SEQ_STATE_READ_ACQUIRED 4 +#define SEQ_STATE_CONTENDED 5 + +/* + * MAX_LOCK_DEPTH + * Imported from include/linux/sched.h. + * Should this be synchronized? + */ +#define MAX_LOCK_DEPTH 48 + +/* + * struct lock_seq_stat: + * Place to put on state of one lock sequence + * 1) acquire -> acquired -> release + * 2) acquire -> contended -> acquired -> release + * 3) acquire (with read or try) -> release + * 4) Are there other patterns? + */ +struct lock_seq_stat { + struct list_head list; + int state; + u64 prev_event_time; + void *addr; + + int read_count; +}; + +struct thread_stat { + struct rb_node rb; + + u32 tid; + struct list_head seq_list; +}; + +static struct rb_root thread_stats; + +static struct thread_stat *thread_stat_find(u32 tid) +{ + struct rb_node *node; + struct thread_stat *st; + + node = thread_stats.rb_node; + while (node) { + st = container_of(node, struct thread_stat, rb); + if (st->tid == tid) + return st; + else if (tid < st->tid) + node = node->rb_left; + else + node = node->rb_right; + } + + return NULL; +} + +static void thread_stat_insert(struct thread_stat *new) +{ + struct rb_node **rb = &thread_stats.rb_node; + struct rb_node *parent = NULL; + struct thread_stat *p; + + while (*rb) { + p = container_of(*rb, struct thread_stat, rb); + parent = *rb; + + if (new->tid < p->tid) + rb = &(*rb)->rb_left; + else if (new->tid > p->tid) + rb = &(*rb)->rb_right; + else + BUG_ON("inserting invalid thread_stat\n"); + } + + rb_link_node(&new->rb, parent, rb); + rb_insert_color(&new->rb, &thread_stats); +} + +static struct thread_stat *thread_stat_findnew_after_first(u32 tid) +{ + struct thread_stat *st; + + st = thread_stat_find(tid); + if (st) + return st; + + st = zalloc(sizeof(struct thread_stat)); + if (!st) { + pr_err("memory allocation failed\n"); + return NULL; + } + + st->tid = tid; + INIT_LIST_HEAD(&st->seq_list); + + thread_stat_insert(st); + + return st; +} + +static struct thread_stat *thread_stat_findnew_first(u32 tid); +static struct thread_stat *(*thread_stat_findnew)(u32 tid) = + thread_stat_findnew_first; + +static struct thread_stat *thread_stat_findnew_first(u32 tid) +{ + struct thread_stat *st; + + st = zalloc(sizeof(struct thread_stat)); + if (!st) { + pr_err("memory allocation failed\n"); + return NULL; + } + st->tid = tid; + INIT_LIST_HEAD(&st->seq_list); + + rb_link_node(&st->rb, NULL, &thread_stats.rb_node); + rb_insert_color(&st->rb, &thread_stats); + + thread_stat_findnew = thread_stat_findnew_after_first; + return st; +} + +/* build simple key function one is bigger than two */ +#define SINGLE_KEY(member) \ + static int lock_stat_key_ ## member(struct lock_stat *one, \ + struct lock_stat *two) \ + { \ + return one->member > two->member; \ + } + +SINGLE_KEY(nr_acquired) +SINGLE_KEY(nr_contended) +SINGLE_KEY(wait_time_total) +SINGLE_KEY(wait_time_max) + +static int lock_stat_key_wait_time_min(struct lock_stat *one, + struct lock_stat *two) +{ + u64 s1 = one->wait_time_min; + u64 s2 = two->wait_time_min; + if (s1 == ULLONG_MAX) + s1 = 0; + if (s2 == ULLONG_MAX) + s2 = 0; + return s1 > s2; +} + +struct lock_key { + /* + * name: the value for specify by user + * this should be simpler than raw name of member + * e.g. nr_acquired -> acquired, wait_time_total -> wait_total + */ + const char *name; + int (*key)(struct lock_stat*, struct lock_stat*); +}; + +static const char *sort_key = "acquired"; + +static int (*compare)(struct lock_stat *, struct lock_stat *); + +static struct rb_root result; /* place to store sorted data */ + +#define DEF_KEY_LOCK(name, fn_suffix) \ + { #name, lock_stat_key_ ## fn_suffix } +struct lock_key keys[] = { + DEF_KEY_LOCK(acquired, nr_acquired), + DEF_KEY_LOCK(contended, nr_contended), + DEF_KEY_LOCK(wait_total, wait_time_total), + DEF_KEY_LOCK(wait_min, wait_time_min), + DEF_KEY_LOCK(wait_max, wait_time_max), + + /* extra comparisons much complicated should be here */ + + { NULL, NULL } +}; + +static int select_key(void) +{ + int i; + + for (i = 0; keys[i].name; i++) { + if (!strcmp(keys[i].name, sort_key)) { + compare = keys[i].key; + return 0; + } + } + + pr_err("Unknown compare key: %s\n", sort_key); + + return -1; +} + +static void insert_to_result(struct lock_stat *st, + int (*bigger)(struct lock_stat *, struct lock_stat *)) +{ + struct rb_node **rb = &result.rb_node; + struct rb_node *parent = NULL; + struct lock_stat *p; + + while (*rb) { + p = container_of(*rb, struct lock_stat, rb); + parent = *rb; + + if (bigger(st, p)) + rb = &(*rb)->rb_left; + else + rb = &(*rb)->rb_right; + } + + rb_link_node(&st->rb, parent, rb); + rb_insert_color(&st->rb, &result); +} + +/* returns left most element of result, and erase it */ +static struct lock_stat *pop_from_result(void) +{ + struct rb_node *node = result.rb_node; + + if (!node) + return NULL; + + while (node->rb_left) + node = node->rb_left; + + rb_erase(node, &result); + return container_of(node, struct lock_stat, rb); +} + +static struct lock_stat *lock_stat_findnew(void *addr, const char *name) +{ + struct list_head *entry = lockhashentry(addr); + struct lock_stat *ret, *new; + + list_for_each_entry(ret, entry, hash_entry) { + if (ret->addr == addr) + return ret; + } + + new = zalloc(sizeof(struct lock_stat)); + if (!new) + goto alloc_failed; + + new->addr = addr; + new->name = zalloc(sizeof(char) * strlen(name) + 1); + if (!new->name) + goto alloc_failed; + strcpy(new->name, name); + + new->wait_time_min = ULLONG_MAX; + + list_add(&new->hash_entry, entry); + return new; + +alloc_failed: + pr_err("memory allocation failed\n"); + return NULL; +} + +struct trace_lock_handler { + int (*acquire_event)(struct perf_evsel *evsel, + struct perf_sample *sample); + + int (*acquired_event)(struct perf_evsel *evsel, + struct perf_sample *sample); + + int (*contended_event)(struct perf_evsel *evsel, + struct perf_sample *sample); + + int (*release_event)(struct perf_evsel *evsel, + struct perf_sample *sample); +}; + +static struct lock_seq_stat *get_seq(struct thread_stat *ts, void *addr) +{ + struct lock_seq_stat *seq; + + list_for_each_entry(seq, &ts->seq_list, list) { + if (seq->addr == addr) + return seq; + } + + seq = zalloc(sizeof(struct lock_seq_stat)); + if (!seq) { + pr_err("memory allocation failed\n"); + return NULL; + } + seq->state = SEQ_STATE_UNINITIALIZED; + seq->addr = addr; + + list_add(&seq->list, &ts->seq_list); + return seq; +} + +enum broken_state { + BROKEN_ACQUIRE, + BROKEN_ACQUIRED, + BROKEN_CONTENDED, + BROKEN_RELEASE, + BROKEN_MAX, +}; + +static int bad_hist[BROKEN_MAX]; + +enum acquire_flags { + TRY_LOCK = 1, + READ_LOCK = 2, +}; + +static int report_lock_acquire_event(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + void *addr; + struct lock_stat *ls; + struct thread_stat *ts; + struct lock_seq_stat *seq; + const char *name = perf_evsel__strval(evsel, sample, "name"); + u64 tmp = perf_evsel__intval(evsel, sample, "lockdep_addr"); + int flag = perf_evsel__intval(evsel, sample, "flag"); + + memcpy(&addr, &tmp, sizeof(void *)); + + ls = lock_stat_findnew(addr, name); + if (!ls) + return -1; + if (ls->discard) + return 0; + + ts = thread_stat_findnew(sample->tid); + if (!ts) + return -1; + + seq = get_seq(ts, addr); + if (!seq) + return -1; + + switch (seq->state) { + case SEQ_STATE_UNINITIALIZED: + case SEQ_STATE_RELEASED: + if (!flag) { + seq->state = SEQ_STATE_ACQUIRING; + } else { + if (flag & TRY_LOCK) + ls->nr_trylock++; + if (flag & READ_LOCK) + ls->nr_readlock++; + seq->state = SEQ_STATE_READ_ACQUIRED; + seq->read_count = 1; + ls->nr_acquired++; + } + break; + case SEQ_STATE_READ_ACQUIRED: + if (flag & READ_LOCK) { + seq->read_count++; + ls->nr_acquired++; + goto end; + } else { + goto broken; + } + break; + case SEQ_STATE_ACQUIRED: + case SEQ_STATE_ACQUIRING: + case SEQ_STATE_CONTENDED: +broken: + /* broken lock sequence, discard it */ + ls->discard = 1; + bad_hist[BROKEN_ACQUIRE]++; + list_del(&seq->list); + free(seq); + goto end; + break; + default: + BUG_ON("Unknown state of lock sequence found!\n"); + break; + } + + ls->nr_acquire++; + seq->prev_event_time = sample->time; +end: + return 0; +} + +static int report_lock_acquired_event(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + void *addr; + struct lock_stat *ls; + struct thread_stat *ts; + struct lock_seq_stat *seq; + u64 contended_term; + const char *name = perf_evsel__strval(evsel, sample, "name"); + u64 tmp = perf_evsel__intval(evsel, sample, "lockdep_addr"); + + memcpy(&addr, &tmp, sizeof(void *)); + + ls = lock_stat_findnew(addr, name); + if (!ls) + return -1; + if (ls->discard) + return 0; + + ts = thread_stat_findnew(sample->tid); + if (!ts) + return -1; + + seq = get_seq(ts, addr); + if (!seq) + return -1; + + switch (seq->state) { + case SEQ_STATE_UNINITIALIZED: + /* orphan event, do nothing */ + return 0; + case SEQ_STATE_ACQUIRING: + break; + case SEQ_STATE_CONTENDED: + contended_term = sample->time - seq->prev_event_time; + ls->wait_time_total += contended_term; + if (contended_term < ls->wait_time_min) + ls->wait_time_min = contended_term; + if (ls->wait_time_max < contended_term) + ls->wait_time_max = contended_term; + break; + case SEQ_STATE_RELEASED: + case SEQ_STATE_ACQUIRED: + case SEQ_STATE_READ_ACQUIRED: + /* broken lock sequence, discard it */ + ls->discard = 1; + bad_hist[BROKEN_ACQUIRED]++; + list_del(&seq->list); + free(seq); + goto end; + break; + + default: + BUG_ON("Unknown state of lock sequence found!\n"); + break; + } + + seq->state = SEQ_STATE_ACQUIRED; + ls->nr_acquired++; + seq->prev_event_time = sample->time; +end: + return 0; +} + +static int report_lock_contended_event(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + void *addr; + struct lock_stat *ls; + struct thread_stat *ts; + struct lock_seq_stat *seq; + const char *name = perf_evsel__strval(evsel, sample, "name"); + u64 tmp = perf_evsel__intval(evsel, sample, "lockdep_addr"); + + memcpy(&addr, &tmp, sizeof(void *)); + + ls = lock_stat_findnew(addr, name); + if (!ls) + return -1; + if (ls->discard) + return 0; + + ts = thread_stat_findnew(sample->tid); + if (!ts) + return -1; + + seq = get_seq(ts, addr); + if (!seq) + return -1; + + switch (seq->state) { + case SEQ_STATE_UNINITIALIZED: + /* orphan event, do nothing */ + return 0; + case SEQ_STATE_ACQUIRING: + break; + case SEQ_STATE_RELEASED: + case SEQ_STATE_ACQUIRED: + case SEQ_STATE_READ_ACQUIRED: + case SEQ_STATE_CONTENDED: + /* broken lock sequence, discard it */ + ls->discard = 1; + bad_hist[BROKEN_CONTENDED]++; + list_del(&seq->list); + free(seq); + goto end; + break; + default: + BUG_ON("Unknown state of lock sequence found!\n"); + break; + } + + seq->state = SEQ_STATE_CONTENDED; + ls->nr_contended++; + seq->prev_event_time = sample->time; +end: + return 0; +} + +static int report_lock_release_event(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + void *addr; + struct lock_stat *ls; + struct thread_stat *ts; + struct lock_seq_stat *seq; + const char *name = perf_evsel__strval(evsel, sample, "name"); + u64 tmp = perf_evsel__intval(evsel, sample, "lockdep_addr"); + + memcpy(&addr, &tmp, sizeof(void *)); + + ls = lock_stat_findnew(addr, name); + if (!ls) + return -1; + if (ls->discard) + return 0; + + ts = thread_stat_findnew(sample->tid); + if (!ts) + return -1; + + seq = get_seq(ts, addr); + if (!seq) + return -1; + + switch (seq->state) { + case SEQ_STATE_UNINITIALIZED: + goto end; + break; + case SEQ_STATE_ACQUIRED: + break; + case SEQ_STATE_READ_ACQUIRED: + seq->read_count--; + BUG_ON(seq->read_count < 0); + if (!seq->read_count) { + ls->nr_release++; + goto end; + } + break; + case SEQ_STATE_ACQUIRING: + case SEQ_STATE_CONTENDED: + case SEQ_STATE_RELEASED: + /* broken lock sequence, discard it */ + ls->discard = 1; + bad_hist[BROKEN_RELEASE]++; + goto free_seq; + break; + default: + BUG_ON("Unknown state of lock sequence found!\n"); + break; + } + + ls->nr_release++; +free_seq: + list_del(&seq->list); + free(seq); +end: + return 0; +} + +/* lock oriented handlers */ +/* TODO: handlers for CPU oriented, thread oriented */ +static struct trace_lock_handler report_lock_ops = { + .acquire_event = report_lock_acquire_event, + .acquired_event = report_lock_acquired_event, + .contended_event = report_lock_contended_event, + .release_event = report_lock_release_event, +}; + +static struct trace_lock_handler *trace_handler; + +static int perf_evsel__process_lock_acquire(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + if (trace_handler->acquire_event) + return trace_handler->acquire_event(evsel, sample); + return 0; +} + +static int perf_evsel__process_lock_acquired(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + if (trace_handler->acquired_event) + return trace_handler->acquired_event(evsel, sample); + return 0; +} + +static int perf_evsel__process_lock_contended(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + if (trace_handler->contended_event) + return trace_handler->contended_event(evsel, sample); + return 0; +} + +static int perf_evsel__process_lock_release(struct perf_evsel *evsel, + struct perf_sample *sample) +{ + if (trace_handler->release_event) + return trace_handler->release_event(evsel, sample); + return 0; +} + +static void print_bad_events(int bad, int total) +{ + /* Output for debug, this have to be removed */ + int i; + const char *name[4] = + { "acquire", "acquired", "contended", "release" }; + + pr_info("\n=== output for debug===\n\n"); + pr_info("bad: %d, total: %d\n", bad, total); + pr_info("bad rate: %f %%\n", (double)bad / (double)total * 100); + pr_info("histogram of events caused bad sequence\n"); + for (i = 0; i < BROKEN_MAX; i++) + pr_info(" %10s: %d\n", name[i], bad_hist[i]); +} + +/* TODO: various way to print, coloring, nano or milli sec */ +static void print_result(void) +{ + struct lock_stat *st; + char cut_name[20]; + int bad, total; + + pr_info("%20s ", "Name"); + pr_info("%10s ", "acquired"); + pr_info("%10s ", "contended"); + + pr_info("%15s ", "total wait (ns)"); + pr_info("%15s ", "max wait (ns)"); + pr_info("%15s ", "min wait (ns)"); + + pr_info("\n\n"); + + bad = total = 0; + while ((st = pop_from_result())) { + total++; + if (st->discard) { + bad++; + continue; + } + bzero(cut_name, 20); + + if (strlen(st->name) < 16) { + /* output raw name */ + pr_info("%20s ", st->name); + } else { + strncpy(cut_name, st->name, 16); + cut_name[16] = '.'; + cut_name[17] = '.'; + cut_name[18] = '.'; + cut_name[19] = '\0'; + /* cut off name for saving output style */ + pr_info("%20s ", cut_name); + } + + pr_info("%10u ", st->nr_acquired); + pr_info("%10u ", st->nr_contended); + + pr_info("%15" PRIu64 " ", st->wait_time_total); + pr_info("%15" PRIu64 " ", st->wait_time_max); + pr_info("%15" PRIu64 " ", st->wait_time_min == ULLONG_MAX ? + 0 : st->wait_time_min); + pr_info("\n"); + } + + print_bad_events(bad, total); +} + +static bool info_threads, info_map; + +static void dump_threads(void) +{ + struct thread_stat *st; + struct rb_node *node; + struct thread *t; + + pr_info("%10s: comm\n", "Thread ID"); + + node = rb_first(&thread_stats); + while (node) { + st = container_of(node, struct thread_stat, rb); + t = perf_session__findnew(session, st->tid); + pr_info("%10d: %s\n", st->tid, t->comm); + node = rb_next(node); + }; +} + +static void dump_map(void) +{ + unsigned int i; + struct lock_stat *st; + + pr_info("Address of instance: name of class\n"); + for (i = 0; i < LOCKHASH_SIZE; i++) { + list_for_each_entry(st, &lockhash_table[i], hash_entry) { + pr_info(" %p: %s\n", st->addr, st->name); + } + } +} + +static int dump_info(void) +{ + int rc = 0; + + if (info_threads) + dump_threads(); + else if (info_map) + dump_map(); + else { + rc = -1; + pr_err("Unknown type of information\n"); + } + + return rc; +} + +typedef int (*tracepoint_handler)(struct perf_evsel *evsel, + struct perf_sample *sample); + +static int process_sample_event(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct thread *thread = machine__findnew_thread(machine, sample->tid); + + if (thread == NULL) { + pr_debug("problem processing %d event, skipping it.\n", + event->header.type); + return -1; + } + + if (evsel->handler.func != NULL) { + tracepoint_handler f = evsel->handler.func; + return f(evsel, sample); + } + + return 0; +} + +static const struct perf_evsel_str_handler lock_tracepoints[] = { + { "lock:lock_acquire", perf_evsel__process_lock_acquire, }, /* CONFIG_LOCKDEP */ + { "lock:lock_acquired", perf_evsel__process_lock_acquired, }, /* CONFIG_LOCKDEP, CONFIG_LOCK_STAT */ + { "lock:lock_contended", perf_evsel__process_lock_contended, }, /* CONFIG_LOCKDEP, CONFIG_LOCK_STAT */ + { "lock:lock_release", perf_evsel__process_lock_release, }, /* CONFIG_LOCKDEP */ +}; + +static int read_events(void) +{ + struct perf_tool eops = { + .sample = process_sample_event, + .comm = perf_event__process_comm, + .ordered_samples = true, + }; + session = perf_session__new(input_name, O_RDONLY, 0, false, &eops); + if (!session) { + pr_err("Initializing perf session failed\n"); + return -1; + } + + if (perf_session__set_tracepoints_handlers(session, lock_tracepoints)) { + pr_err("Initializing perf session tracepoint handlers failed\n"); + return -1; + } + + return perf_session__process_events(session, &eops); +} + +static void sort_result(void) +{ + unsigned int i; + struct lock_stat *st; + + for (i = 0; i < LOCKHASH_SIZE; i++) { + list_for_each_entry(st, &lockhash_table[i], hash_entry) { + insert_to_result(st, compare); + } + } +} + +static int __cmd_report(void) +{ + setup_pager(); + + if ((select_key() != 0) || + (read_events() != 0)) + return -1; + + sort_result(); + print_result(); + + return 0; +} + +static int __cmd_record(int argc, const char **argv) +{ + const char *record_args[] = { + "record", "-R", "-f", "-m", "1024", "-c", "1", + }; + unsigned int rec_argc, i, j; + const char **rec_argv; + + for (i = 0; i < ARRAY_SIZE(lock_tracepoints); i++) { + if (!is_valid_tracepoint(lock_tracepoints[i].name)) { + pr_err("tracepoint %s is not enabled. " + "Are CONFIG_LOCKDEP and CONFIG_LOCK_STAT enabled?\n", + lock_tracepoints[i].name); + return 1; + } + } + + rec_argc = ARRAY_SIZE(record_args) + argc - 1; + /* factor of 2 is for -e in front of each tracepoint */ + rec_argc += 2 * ARRAY_SIZE(lock_tracepoints); + + rec_argv = calloc(rec_argc + 1, sizeof(char *)); + if (rec_argv == NULL) + return -ENOMEM; + + for (i = 0; i < ARRAY_SIZE(record_args); i++) + rec_argv[i] = strdup(record_args[i]); + + for (j = 0; j < ARRAY_SIZE(lock_tracepoints); j++) { + rec_argv[i++] = "-e"; + rec_argv[i++] = strdup(lock_tracepoints[j].name); + } + + for (j = 1; j < (unsigned int)argc; j++, i++) + rec_argv[i] = argv[j]; + + BUG_ON(i != rec_argc); + + return cmd_record(i, rec_argv, NULL); +} + +int cmd_lock(int argc, const char **argv, const char *prefix __maybe_unused) +{ + const struct option info_options[] = { + OPT_BOOLEAN('t', "threads", &info_threads, + "dump thread list in perf.data"), + OPT_BOOLEAN('m', "map", &info_map, + "map of lock instances (address:name table)"), + OPT_END() + }; + const struct option lock_options[] = { + OPT_STRING('i', "input", &input_name, "file", "input file name"), + OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), + OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"), + OPT_END() + }; + const struct option report_options[] = { + OPT_STRING('k', "key", &sort_key, "acquired", + "key for sorting (acquired / contended / wait_total / wait_max / wait_min)"), + /* TODO: type */ + OPT_END() + }; + const char * const info_usage[] = { + "perf lock info [<options>]", + NULL + }; + const char * const lock_usage[] = { + "perf lock [<options>] {record|report|script|info}", + NULL + }; + const char * const report_usage[] = { + "perf lock report [<options>]", + NULL + }; + unsigned int i; + int rc = 0; + + symbol__init(); + for (i = 0; i < LOCKHASH_SIZE; i++) + INIT_LIST_HEAD(lockhash_table + i); + + argc = parse_options(argc, argv, lock_options, lock_usage, + PARSE_OPT_STOP_AT_NON_OPTION); + if (!argc) + usage_with_options(lock_usage, lock_options); + + if (!strncmp(argv[0], "rec", 3)) { + return __cmd_record(argc, argv); + } else if (!strncmp(argv[0], "report", 6)) { + trace_handler = &report_lock_ops; + if (argc) { + argc = parse_options(argc, argv, + report_options, report_usage, 0); + if (argc) + usage_with_options(report_usage, report_options); + } + __cmd_report(); + } else if (!strcmp(argv[0], "script")) { + /* Aliased to 'perf script' */ + return cmd_script(argc, argv, prefix); + } else if (!strcmp(argv[0], "info")) { + if (argc) { + argc = parse_options(argc, argv, + info_options, info_usage, 0); + if (argc) + usage_with_options(info_usage, info_options); + } + /* recycling report_lock_ops */ + trace_handler = &report_lock_ops; + setup_pager(); + if (read_events() != 0) + rc = -1; + else + rc = dump_info(); + } else { + usage_with_options(lock_usage, lock_options); + } + + return rc; +} diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c new file mode 100644 index 00000000..de38a034 --- /dev/null +++ b/tools/perf/builtin-probe.c @@ -0,0 +1,470 @@ +/* + * builtin-probe.c + * + * Builtin probe command: Set up probe events by C expression + * + * Written by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ +#include <sys/utsname.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <errno.h> +#include <stdio.h> +#include <unistd.h> +#include <stdlib.h> +#include <string.h> + +#include "perf.h" +#include "builtin.h" +#include "util/util.h" +#include "util/strlist.h" +#include "util/strfilter.h" +#include "util/symbol.h" +#include "util/debug.h" +#include "util/debugfs.h" +#include "util/parse-options.h" +#include "util/probe-finder.h" +#include "util/probe-event.h" + +#define DEFAULT_VAR_FILTER "!__k???tab_* & !__crc_*" +#define DEFAULT_FUNC_FILTER "!_*" + +/* Session management structure */ +static struct { + bool list_events; + bool force_add; + bool show_lines; + bool show_vars; + bool show_ext_vars; + bool show_funcs; + bool mod_events; + bool uprobes; + int nevents; + struct perf_probe_event events[MAX_PROBES]; + struct strlist *dellist; + struct line_range line_range; + const char *target; + int max_probe_points; + struct strfilter *filter; +} params; + +/* Parse an event definition. Note that any error must die. */ +static int parse_probe_event(const char *str) +{ + struct perf_probe_event *pev = ¶ms.events[params.nevents]; + int ret; + + pr_debug("probe-definition(%d): %s\n", params.nevents, str); + if (++params.nevents == MAX_PROBES) { + pr_err("Too many probes (> %d) were specified.", MAX_PROBES); + return -1; + } + + pev->uprobes = params.uprobes; + + /* Parse a perf-probe command into event */ + ret = parse_perf_probe_command(str, pev); + pr_debug("%d arguments\n", pev->nargs); + + return ret; +} + +static int set_target(const char *ptr) +{ + int found = 0; + const char *buf; + + /* + * The first argument after options can be an absolute path + * to an executable / library or kernel module. + * + * TODO: Support relative path, and $PATH, $LD_LIBRARY_PATH, + * short module name. + */ + if (!params.target && ptr && *ptr == '/') { + params.target = ptr; + found = 1; + buf = ptr + (strlen(ptr) - 3); + + if (strcmp(buf, ".ko")) + params.uprobes = true; + + } + + return found; +} + +static int parse_probe_event_argv(int argc, const char **argv) +{ + int i, len, ret, found_target; + char *buf; + + found_target = set_target(argv[0]); + if (found_target && argc == 1) + return 0; + + /* Bind up rest arguments */ + len = 0; + for (i = 0; i < argc; i++) { + if (i == 0 && found_target) + continue; + + len += strlen(argv[i]) + 1; + } + buf = zalloc(len + 1); + if (buf == NULL) + return -ENOMEM; + len = 0; + for (i = 0; i < argc; i++) { + if (i == 0 && found_target) + continue; + + len += sprintf(&buf[len], "%s ", argv[i]); + } + params.mod_events = true; + ret = parse_probe_event(buf); + free(buf); + return ret; +} + +static int opt_add_probe_event(const struct option *opt __maybe_unused, + const char *str, int unset __maybe_unused) +{ + if (str) { + params.mod_events = true; + return parse_probe_event(str); + } else + return 0; +} + +static int opt_del_probe_event(const struct option *opt __maybe_unused, + const char *str, int unset __maybe_unused) +{ + if (str) { + params.mod_events = true; + if (!params.dellist) + params.dellist = strlist__new(true, NULL); + strlist__add(params.dellist, str); + } + return 0; +} + +static int opt_set_target(const struct option *opt, const char *str, + int unset __maybe_unused) +{ + int ret = -ENOENT; + + if (str && !params.target) { + if (!strcmp(opt->long_name, "exec")) + params.uprobes = true; +#ifdef DWARF_SUPPORT + else if (!strcmp(opt->long_name, "module")) + params.uprobes = false; +#endif + else + return ret; + + params.target = str; + ret = 0; + } + + return ret; +} + +#ifdef DWARF_SUPPORT +static int opt_show_lines(const struct option *opt __maybe_unused, + const char *str, int unset __maybe_unused) +{ + int ret = 0; + + if (!str) + return 0; + + if (params.show_lines) { + pr_warning("Warning: more than one --line options are" + " detected. Only the first one is valid.\n"); + return 0; + } + + params.show_lines = true; + ret = parse_line_range_desc(str, ¶ms.line_range); + INIT_LIST_HEAD(¶ms.line_range.line_list); + + return ret; +} + +static int opt_show_vars(const struct option *opt __maybe_unused, + const char *str, int unset __maybe_unused) +{ + struct perf_probe_event *pev = ¶ms.events[params.nevents]; + int ret; + + if (!str) + return 0; + + ret = parse_probe_event(str); + if (!ret && pev->nargs != 0) { + pr_err(" Error: '--vars' doesn't accept arguments.\n"); + return -EINVAL; + } + params.show_vars = true; + + return ret; +} +#endif + +static int opt_set_filter(const struct option *opt __maybe_unused, + const char *str, int unset __maybe_unused) +{ + const char *err; + + if (str) { + pr_debug2("Set filter: %s\n", str); + if (params.filter) + strfilter__delete(params.filter); + params.filter = strfilter__new(str, &err); + if (!params.filter) { + pr_err("Filter parse error at %td.\n", err - str + 1); + pr_err("Source: \"%s\"\n", str); + pr_err(" %*c\n", (int)(err - str + 1), '^'); + return -EINVAL; + } + } + + return 0; +} + +int cmd_probe(int argc, const char **argv, const char *prefix __maybe_unused) +{ + const char * const probe_usage[] = { + "perf probe [<options>] 'PROBEDEF' ['PROBEDEF' ...]", + "perf probe [<options>] --add 'PROBEDEF' [--add 'PROBEDEF' ...]", + "perf probe [<options>] --del '[GROUP:]EVENT' ...", + "perf probe --list", +#ifdef DWARF_SUPPORT + "perf probe [<options>] --line 'LINEDESC'", + "perf probe [<options>] --vars 'PROBEPOINT'", +#endif + NULL +}; + const struct option options[] = { + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show parsed arguments, etc)"), + OPT_BOOLEAN('l', "list", ¶ms.list_events, + "list up current probe events"), + OPT_CALLBACK('d', "del", NULL, "[GROUP:]EVENT", "delete a probe event.", + opt_del_probe_event), + OPT_CALLBACK('a', "add", NULL, +#ifdef DWARF_SUPPORT + "[EVENT=]FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT" + " [[NAME=]ARG ...]", +#else + "[EVENT=]FUNC[+OFF|%return] [[NAME=]ARG ...]", +#endif + "probe point definition, where\n" + "\t\tGROUP:\tGroup name (optional)\n" + "\t\tEVENT:\tEvent name\n" + "\t\tFUNC:\tFunction name\n" + "\t\tOFF:\tOffset from function entry (in byte)\n" + "\t\t%return:\tPut the probe at function return\n" +#ifdef DWARF_SUPPORT + "\t\tSRC:\tSource code path\n" + "\t\tRL:\tRelative line number from function entry.\n" + "\t\tAL:\tAbsolute line number in file.\n" + "\t\tPT:\tLazy expression of line code.\n" + "\t\tARG:\tProbe argument (local variable name or\n" + "\t\t\tkprobe-tracer argument format.)\n", +#else + "\t\tARG:\tProbe argument (kprobe-tracer argument format.)\n", +#endif + opt_add_probe_event), + OPT_BOOLEAN('f', "force", ¶ms.force_add, "forcibly add events" + " with existing name"), +#ifdef DWARF_SUPPORT + OPT_CALLBACK('L', "line", NULL, + "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]", + "Show source code lines.", opt_show_lines), + OPT_CALLBACK('V', "vars", NULL, + "FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT", + "Show accessible variables on PROBEDEF", opt_show_vars), + OPT_BOOLEAN('\0', "externs", ¶ms.show_ext_vars, + "Show external variables too (with --vars only)"), + OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name, + "file", "vmlinux pathname"), + OPT_STRING('s', "source", &symbol_conf.source_prefix, + "directory", "path to kernel source"), + OPT_CALLBACK('m', "module", NULL, "modname|path", + "target module name (for online) or path (for offline)", + opt_set_target), +#endif + OPT__DRY_RUN(&probe_event_dry_run), + OPT_INTEGER('\0', "max-probes", ¶ms.max_probe_points, + "Set how many probe points can be found for a probe."), + OPT_BOOLEAN('F', "funcs", ¶ms.show_funcs, + "Show potential probe-able functions."), + OPT_CALLBACK('\0', "filter", NULL, + "[!]FILTER", "Set a filter (with --vars/funcs only)\n" + "\t\t\t(default: \"" DEFAULT_VAR_FILTER "\" for --vars,\n" + "\t\t\t \"" DEFAULT_FUNC_FILTER "\" for --funcs)", + opt_set_filter), + OPT_CALLBACK('x', "exec", NULL, "executable|path", + "target executable name or path", opt_set_target), + OPT_END() + }; + int ret; + + argc = parse_options(argc, argv, options, probe_usage, + PARSE_OPT_STOP_AT_NON_OPTION); + if (argc > 0) { + if (strcmp(argv[0], "-") == 0) { + pr_warning(" Error: '-' is not supported.\n"); + usage_with_options(probe_usage, options); + } + ret = parse_probe_event_argv(argc, argv); + if (ret < 0) { + pr_err(" Error: Parse Error. (%d)\n", ret); + return ret; + } + } + + if (params.max_probe_points == 0) + params.max_probe_points = MAX_PROBES; + + if ((!params.nevents && !params.dellist && !params.list_events && + !params.show_lines && !params.show_funcs)) + usage_with_options(probe_usage, options); + + /* + * Only consider the user's kernel image path if given. + */ + symbol_conf.try_vmlinux_path = (symbol_conf.vmlinux_name == NULL); + + if (params.list_events) { + if (params.mod_events) { + pr_err(" Error: Don't use --list with --add/--del.\n"); + usage_with_options(probe_usage, options); + } + if (params.show_lines) { + pr_err(" Error: Don't use --list with --line.\n"); + usage_with_options(probe_usage, options); + } + if (params.show_vars) { + pr_err(" Error: Don't use --list with --vars.\n"); + usage_with_options(probe_usage, options); + } + if (params.show_funcs) { + pr_err(" Error: Don't use --list with --funcs.\n"); + usage_with_options(probe_usage, options); + } + if (params.uprobes) { + pr_warning(" Error: Don't use --list with --exec.\n"); + usage_with_options(probe_usage, options); + } + ret = show_perf_probe_events(); + if (ret < 0) + pr_err(" Error: Failed to show event list. (%d)\n", + ret); + return ret; + } + if (params.show_funcs) { + if (params.nevents != 0 || params.dellist) { + pr_err(" Error: Don't use --funcs with" + " --add/--del.\n"); + usage_with_options(probe_usage, options); + } + if (params.show_lines) { + pr_err(" Error: Don't use --funcs with --line.\n"); + usage_with_options(probe_usage, options); + } + if (params.show_vars) { + pr_err(" Error: Don't use --funcs with --vars.\n"); + usage_with_options(probe_usage, options); + } + if (!params.filter) + params.filter = strfilter__new(DEFAULT_FUNC_FILTER, + NULL); + ret = show_available_funcs(params.target, params.filter, + params.uprobes); + strfilter__delete(params.filter); + if (ret < 0) + pr_err(" Error: Failed to show functions." + " (%d)\n", ret); + return ret; + } + +#ifdef DWARF_SUPPORT + if (params.show_lines && !params.uprobes) { + if (params.mod_events) { + pr_err(" Error: Don't use --line with" + " --add/--del.\n"); + usage_with_options(probe_usage, options); + } + if (params.show_vars) { + pr_err(" Error: Don't use --line with --vars.\n"); + usage_with_options(probe_usage, options); + } + + ret = show_line_range(¶ms.line_range, params.target); + if (ret < 0) + pr_err(" Error: Failed to show lines. (%d)\n", ret); + return ret; + } + if (params.show_vars) { + if (params.mod_events) { + pr_err(" Error: Don't use --vars with" + " --add/--del.\n"); + usage_with_options(probe_usage, options); + } + if (!params.filter) + params.filter = strfilter__new(DEFAULT_VAR_FILTER, + NULL); + + ret = show_available_vars(params.events, params.nevents, + params.max_probe_points, + params.target, + params.filter, + params.show_ext_vars); + strfilter__delete(params.filter); + if (ret < 0) + pr_err(" Error: Failed to show vars. (%d)\n", ret); + return ret; + } +#endif + + if (params.dellist) { + ret = del_perf_probe_events(params.dellist); + strlist__delete(params.dellist); + if (ret < 0) { + pr_err(" Error: Failed to delete events. (%d)\n", ret); + return ret; + } + } + + if (params.nevents) { + ret = add_perf_probe_events(params.events, params.nevents, + params.max_probe_points, + params.target, + params.force_add); + if (ret < 0) { + pr_err(" Error: Failed to add events. (%d)\n", ret); + return ret; + } + } + return 0; +} diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c new file mode 100644 index 00000000..774c9071 --- /dev/null +++ b/tools/perf/builtin-record.c @@ -0,0 +1,1063 @@ +/* + * 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. + */ +#define _FILE_OFFSET_BITS 64 + +#include "builtin.h" + +#include "perf.h" + +#include "util/build-id.h" +#include "util/util.h" +#include "util/parse-options.h" +#include "util/parse-events.h" + +#include "util/header.h" +#include "util/event.h" +#include "util/evlist.h" +#include "util/evsel.h" +#include "util/debug.h" +#include "util/session.h" +#include "util/tool.h" +#include "util/symbol.h" +#include "util/cpumap.h" +#include "util/thread_map.h" + +#include <unistd.h> +#include <sched.h> +#include <sys/mman.h> + +#ifndef HAVE_ON_EXIT +#ifndef ATEXIT_MAX +#define ATEXIT_MAX 32 +#endif +static int __on_exit_count = 0; +typedef void (*on_exit_func_t) (int, void *); +static on_exit_func_t __on_exit_funcs[ATEXIT_MAX]; +static void *__on_exit_args[ATEXIT_MAX]; +static int __exitcode = 0; +static void __handle_on_exit_funcs(void); +static int on_exit(on_exit_func_t function, void *arg); +#define exit(x) (exit)(__exitcode = (x)) + +static int on_exit(on_exit_func_t function, void *arg) +{ + if (__on_exit_count == ATEXIT_MAX) + return -ENOMEM; + else if (__on_exit_count == 0) + atexit(__handle_on_exit_funcs); + __on_exit_funcs[__on_exit_count] = function; + __on_exit_args[__on_exit_count++] = arg; + return 0; +} + +static void __handle_on_exit_funcs(void) +{ + int i; + for (i = 0; i < __on_exit_count; i++) + __on_exit_funcs[i] (__exitcode, __on_exit_args[i]); +} +#endif + +enum write_mode_t { + WRITE_FORCE, + WRITE_APPEND +}; + +struct perf_record { + struct perf_tool tool; + struct perf_record_opts opts; + u64 bytes_written; + const char *output_name; + struct perf_evlist *evlist; + struct perf_session *session; + const char *progname; + int output; + unsigned int page_size; + int realtime_prio; + enum write_mode_t write_mode; + bool no_buildid; + bool no_buildid_cache; + bool force; + bool file_new; + bool append_file; + long samples; + off_t post_processing_offset; +}; + +static void advance_output(struct perf_record *rec, size_t size) +{ + rec->bytes_written += size; +} + +static int write_output(struct perf_record *rec, void *buf, size_t size) +{ + while (size) { + int ret = write(rec->output, buf, size); + + if (ret < 0) { + pr_err("failed to write\n"); + return -1; + } + + size -= ret; + buf += ret; + + rec->bytes_written += ret; + } + + return 0; +} + +static int process_synthesized_event(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine __maybe_unused) +{ + struct perf_record *rec = container_of(tool, struct perf_record, tool); + if (write_output(rec, event, event->header.size) < 0) + return -1; + + return 0; +} + +static int perf_record__mmap_read(struct perf_record *rec, + struct perf_mmap *md) +{ + unsigned int head = perf_mmap__read_head(md); + unsigned int old = md->prev; + unsigned char *data = md->base + rec->page_size; + unsigned long size; + void *buf; + int rc = 0; + + if (old == head) + return 0; + + rec->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; + + if (write_output(rec, buf, size) < 0) { + rc = -1; + goto out; + } + } + + buf = &data[old & md->mask]; + size = head - old; + old += size; + + if (write_output(rec, buf, size) < 0) { + rc = -1; + goto out; + } + + md->prev = old; + perf_mmap__write_tail(md, old); + +out: + return rc; +} + +static volatile int done = 0; +static volatile int signr = -1; +static volatile int child_finished = 0; + +static void sig_handler(int sig) +{ + if (sig == SIGCHLD) + child_finished = 1; + + done = 1; + signr = sig; +} + +static void perf_record__sig_exit(int exit_status __maybe_unused, void *arg) +{ + struct perf_record *rec = arg; + int status; + + if (rec->evlist->workload.pid > 0) { + if (!child_finished) + kill(rec->evlist->workload.pid, SIGTERM); + + wait(&status); + if (WIFSIGNALED(status)) + psignal(WTERMSIG(status), rec->progname); + } + + if (signr == -1 || signr == SIGUSR1) + return; + + signal(signr, SIG_DFL); + kill(getpid(), signr); +} + +static bool perf_evlist__equal(struct perf_evlist *evlist, + struct perf_evlist *other) +{ + struct perf_evsel *pos, *pair; + + if (evlist->nr_entries != other->nr_entries) + return false; + + pair = perf_evlist__first(other); + + list_for_each_entry(pos, &evlist->entries, node) { + if (memcmp(&pos->attr, &pair->attr, sizeof(pos->attr) != 0)) + return false; + pair = perf_evsel__next(pair); + } + + return true; +} + +static int perf_record__open(struct perf_record *rec) +{ + char msg[512]; + struct perf_evsel *pos; + struct perf_evlist *evlist = rec->evlist; + struct perf_session *session = rec->session; + struct perf_record_opts *opts = &rec->opts; + int rc = 0; + + perf_evlist__config(evlist, opts); + + list_for_each_entry(pos, &evlist->entries, node) { +try_again: + if (perf_evsel__open(pos, evlist->cpus, evlist->threads) < 0) { + if (perf_evsel__fallback(pos, errno, msg, sizeof(msg))) { + if (verbose) + ui__warning("%s\n", msg); + goto try_again; + } + + rc = -errno; + perf_evsel__open_strerror(pos, &opts->target, + errno, msg, sizeof(msg)); + ui__error("%s\n", msg); + goto out; + } + } + + if (perf_evlist__apply_filters(evlist)) { + error("failed to set filter with %d (%s)\n", errno, + strerror(errno)); + rc = -1; + goto out; + } + + if (perf_evlist__mmap(evlist, opts->mmap_pages, false) < 0) { + if (errno == EPERM) { + pr_err("Permission error mapping pages.\n" + "Consider increasing " + "/proc/sys/kernel/perf_event_mlock_kb,\n" + "or try again with a smaller value of -m/--mmap_pages.\n" + "(current value: %d)\n", opts->mmap_pages); + rc = -errno; + } else if (!is_power_of_2(opts->mmap_pages) && + (opts->mmap_pages != UINT_MAX)) { + pr_err("--mmap_pages/-m value must be a power of two."); + rc = -EINVAL; + } else { + pr_err("failed to mmap with %d (%s)\n", errno, strerror(errno)); + rc = -errno; + } + goto out; + } + + if (rec->file_new) + session->evlist = evlist; + else { + if (!perf_evlist__equal(session->evlist, evlist)) { + fprintf(stderr, "incompatible append\n"); + rc = -1; + goto out; + } + } + + perf_session__set_id_hdr_size(session); +out: + return rc; +} + +static int process_buildids(struct perf_record *rec) +{ + u64 size = lseek(rec->output, 0, SEEK_CUR); + + if (size == 0) + return 0; + + rec->session->fd = rec->output; + return __perf_session__process_events(rec->session, rec->post_processing_offset, + size - rec->post_processing_offset, + size, &build_id__mark_dso_hit_ops); +} + +static void perf_record__exit(int status, void *arg) +{ + struct perf_record *rec = arg; + + if (status != 0) + return; + + if (!rec->opts.pipe_output) { + rec->session->header.data_size += rec->bytes_written; + + if (!rec->no_buildid) + process_buildids(rec); + perf_session__write_header(rec->session, rec->evlist, + rec->output, true); + perf_session__delete(rec->session); + perf_evlist__delete(rec->evlist); + symbol__exit(); + } +} + +static void perf_event__synthesize_guest_os(struct machine *machine, void *data) +{ + int err; + struct perf_tool *tool = data; + /* + *As for guest kernel when processing subcommand record&report, + *we arrange module mmap prior to guest kernel mmap and trigger + *a preload dso because default guest module symbols are loaded + *from guest kallsyms instead of /lib/modules/XXX/XXX. This + *method is used to avoid symbol missing when the first addr is + *in module instead of in guest kernel. + */ + err = perf_event__synthesize_modules(tool, process_synthesized_event, + machine); + if (err < 0) + pr_err("Couldn't record guest kernel [%d]'s reference" + " relocation symbol.\n", machine->pid); + + /* + * We use _stext for guest kernel because guest kernel's /proc/kallsyms + * have no _text sometimes. + */ + err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event, + machine, "_text"); + if (err < 0) + err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event, + machine, "_stext"); + if (err < 0) + pr_err("Couldn't record guest kernel [%d]'s reference" + " relocation symbol.\n", machine->pid); +} + +static struct perf_event_header finished_round_event = { + .size = sizeof(struct perf_event_header), + .type = PERF_RECORD_FINISHED_ROUND, +}; + +static int perf_record__mmap_read_all(struct perf_record *rec) +{ + int i; + int rc = 0; + + for (i = 0; i < rec->evlist->nr_mmaps; i++) { + if (rec->evlist->mmap[i].base) { + if (perf_record__mmap_read(rec, &rec->evlist->mmap[i]) != 0) { + rc = -1; + goto out; + } + } + } + + if (perf_header__has_feat(&rec->session->header, HEADER_TRACING_DATA)) + rc = write_output(rec, &finished_round_event, + sizeof(finished_round_event)); + +out: + return rc; +} + +static int __cmd_record(struct perf_record *rec, int argc, const char **argv) +{ + struct stat st; + int flags; + int err, output, feat; + unsigned long waking = 0; + const bool forks = argc > 0; + struct machine *machine; + struct perf_tool *tool = &rec->tool; + struct perf_record_opts *opts = &rec->opts; + struct perf_evlist *evsel_list = rec->evlist; + const char *output_name = rec->output_name; + struct perf_session *session; + bool disabled = false; + + rec->progname = argv[0]; + + rec->page_size = sysconf(_SC_PAGE_SIZE); + + on_exit(perf_record__sig_exit, rec); + signal(SIGCHLD, sig_handler); + signal(SIGINT, sig_handler); + signal(SIGUSR1, sig_handler); + + if (!output_name) { + if (!fstat(STDOUT_FILENO, &st) && S_ISFIFO(st.st_mode)) + opts->pipe_output = true; + else + rec->output_name = output_name = "perf.data"; + } + if (output_name) { + if (!strcmp(output_name, "-")) + opts->pipe_output = true; + else if (!stat(output_name, &st) && st.st_size) { + if (rec->write_mode == WRITE_FORCE) { + char oldname[PATH_MAX]; + snprintf(oldname, sizeof(oldname), "%s.old", + output_name); + unlink(oldname); + rename(output_name, oldname); + } + } else if (rec->write_mode == WRITE_APPEND) { + rec->write_mode = WRITE_FORCE; + } + } + + flags = O_CREAT|O_RDWR; + if (rec->write_mode == WRITE_APPEND) + rec->file_new = 0; + else + flags |= O_TRUNC; + + if (opts->pipe_output) + output = STDOUT_FILENO; + else + output = open(output_name, flags, S_IRUSR | S_IWUSR); + if (output < 0) { + perror("failed to create output file"); + return -1; + } + + rec->output = output; + + session = perf_session__new(output_name, O_WRONLY, + rec->write_mode == WRITE_FORCE, false, NULL); + if (session == NULL) { + pr_err("Not enough memory for reading perf file header\n"); + return -1; + } + + rec->session = session; + + for (feat = HEADER_FIRST_FEATURE; feat < HEADER_LAST_FEATURE; feat++) + perf_header__set_feat(&session->header, feat); + + if (rec->no_buildid) + perf_header__clear_feat(&session->header, HEADER_BUILD_ID); + + if (!have_tracepoints(&evsel_list->entries)) + perf_header__clear_feat(&session->header, HEADER_TRACING_DATA); + + if (!rec->opts.branch_stack) + perf_header__clear_feat(&session->header, HEADER_BRANCH_STACK); + + if (!rec->file_new) { + err = perf_session__read_header(session, output); + if (err < 0) + goto out_delete_session; + } + + if (forks) { + err = perf_evlist__prepare_workload(evsel_list, opts, argv); + if (err < 0) { + pr_err("Couldn't run the workload!\n"); + goto out_delete_session; + } + } + + if (perf_record__open(rec) != 0) { + err = -1; + goto out_delete_session; + } + + if (!evsel_list->nr_groups) + perf_header__clear_feat(&session->header, HEADER_GROUP_DESC); + + /* + * perf_session__delete(session) will be called at perf_record__exit() + */ + on_exit(perf_record__exit, rec); + + if (opts->pipe_output) { + err = perf_header__write_pipe(output); + if (err < 0) + goto out_delete_session; + } else if (rec->file_new) { + err = perf_session__write_header(session, evsel_list, + output, false); + if (err < 0) + goto out_delete_session; + } + + if (!rec->no_buildid + && !perf_header__has_feat(&session->header, HEADER_BUILD_ID)) { + pr_err("Couldn't generate buildids. " + "Use --no-buildid to profile anyway.\n"); + err = -1; + goto out_delete_session; + } + + rec->post_processing_offset = lseek(output, 0, SEEK_CUR); + + machine = &session->machines.host; + + if (opts->pipe_output) { + err = perf_event__synthesize_attrs(tool, session, + process_synthesized_event); + if (err < 0) { + pr_err("Couldn't synthesize attrs.\n"); + goto out_delete_session; + } + + err = perf_event__synthesize_event_types(tool, process_synthesized_event, + machine); + if (err < 0) { + pr_err("Couldn't synthesize event_types.\n"); + goto out_delete_session; + } + + if (have_tracepoints(&evsel_list->entries)) { + /* + * FIXME err <= 0 here actually means that + * there were no tracepoints so its not really + * an error, just that we don't need to + * synthesize anything. We really have to + * return this more properly and also + * propagate errors that now are calling die() + */ + err = perf_event__synthesize_tracing_data(tool, output, evsel_list, + process_synthesized_event); + if (err <= 0) { + pr_err("Couldn't record tracing data.\n"); + goto out_delete_session; + } + advance_output(rec, err); + } + } + + err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event, + machine, "_text"); + if (err < 0) + err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event, + machine, "_stext"); + if (err < 0) + pr_err("Couldn't record kernel reference relocation symbol\n" + "Symbol resolution may be skewed if relocation was used (e.g. kexec).\n" + "Check /proc/kallsyms permission or run as root.\n"); + + err = perf_event__synthesize_modules(tool, process_synthesized_event, + machine); + if (err < 0) + pr_err("Couldn't record kernel module information.\n" + "Symbol resolution may be skewed if relocation was used (e.g. kexec).\n" + "Check /proc/modules permission or run as root.\n"); + + if (perf_guest) { + machines__process_guests(&session->machines, + perf_event__synthesize_guest_os, tool); + } + + if (!opts->target.system_wide) + err = perf_event__synthesize_thread_map(tool, evsel_list->threads, + process_synthesized_event, + machine); + else + err = perf_event__synthesize_threads(tool, process_synthesized_event, + machine); + + if (err != 0) + goto out_delete_session; + + if (rec->realtime_prio) { + struct sched_param param; + + param.sched_priority = rec->realtime_prio; + if (sched_setscheduler(0, SCHED_FIFO, ¶m)) { + pr_err("Could not set realtime priority.\n"); + err = -1; + goto out_delete_session; + } + } + + /* + * When perf is starting the traced process, all the events + * (apart from group members) have enable_on_exec=1 set, + * so don't spoil it by prematurely enabling them. + */ + if (!perf_target__none(&opts->target)) + perf_evlist__enable(evsel_list); + + /* + * Let the child rip + */ + if (forks) + perf_evlist__start_workload(evsel_list); + + for (;;) { + int hits = rec->samples; + + if (perf_record__mmap_read_all(rec) < 0) { + err = -1; + goto out_delete_session; + } + + if (hits == rec->samples) { + if (done) + break; + err = poll(evsel_list->pollfd, evsel_list->nr_fds, -1); + waking++; + } + + /* + * When perf is starting the traced process, at the end events + * die with the process and we wait for that. Thus no need to + * disable events in this case. + */ + if (done && !disabled && !perf_target__none(&opts->target)) { + perf_evlist__disable(evsel_list); + disabled = true; + } + } + + if (quiet || signr == SIGUSR1) + return 0; + + fprintf(stderr, "[ perf record: Woken up %ld times to write data ]\n", waking); + + /* + * Approximate RIP event size: 24 bytes. + */ + fprintf(stderr, + "[ perf record: Captured and wrote %.3f MB %s (~%" PRIu64 " samples) ]\n", + (double)rec->bytes_written / 1024.0 / 1024.0, + output_name, + rec->bytes_written / 24); + + return 0; + +out_delete_session: + perf_session__delete(session); + return err; +} + +#define BRANCH_OPT(n, m) \ + { .name = n, .mode = (m) } + +#define BRANCH_END { .name = NULL } + +struct branch_mode { + const char *name; + int mode; +}; + +static const struct branch_mode branch_modes[] = { + BRANCH_OPT("u", PERF_SAMPLE_BRANCH_USER), + BRANCH_OPT("k", PERF_SAMPLE_BRANCH_KERNEL), + BRANCH_OPT("hv", PERF_SAMPLE_BRANCH_HV), + BRANCH_OPT("any", PERF_SAMPLE_BRANCH_ANY), + BRANCH_OPT("any_call", PERF_SAMPLE_BRANCH_ANY_CALL), + BRANCH_OPT("any_ret", PERF_SAMPLE_BRANCH_ANY_RETURN), + BRANCH_OPT("ind_call", PERF_SAMPLE_BRANCH_IND_CALL), + BRANCH_END +}; + +static int +parse_branch_stack(const struct option *opt, const char *str, int unset) +{ +#define ONLY_PLM \ + (PERF_SAMPLE_BRANCH_USER |\ + PERF_SAMPLE_BRANCH_KERNEL |\ + PERF_SAMPLE_BRANCH_HV) + + uint64_t *mode = (uint64_t *)opt->value; + const struct branch_mode *br; + char *s, *os = NULL, *p; + int ret = -1; + + if (unset) + return 0; + + /* + * cannot set it twice, -b + --branch-filter for instance + */ + if (*mode) + return -1; + + /* str may be NULL in case no arg is passed to -b */ + if (str) { + /* because str is read-only */ + s = os = strdup(str); + if (!s) + return -1; + + for (;;) { + p = strchr(s, ','); + if (p) + *p = '\0'; + + for (br = branch_modes; br->name; br++) { + if (!strcasecmp(s, br->name)) + break; + } + if (!br->name) { + ui__warning("unknown branch filter %s," + " check man page\n", s); + goto error; + } + + *mode |= br->mode; + + if (!p) + break; + + s = p + 1; + } + } + ret = 0; + + /* default to any branch */ + if ((*mode & ~ONLY_PLM) == 0) { + *mode = PERF_SAMPLE_BRANCH_ANY; + } +error: + free(os); + return ret; +} + +#ifdef LIBUNWIND_SUPPORT +static int get_stack_size(char *str, unsigned long *_size) +{ + char *endptr; + unsigned long size; + unsigned long max_size = round_down(USHRT_MAX, sizeof(u64)); + + size = strtoul(str, &endptr, 0); + + do { + if (*endptr) + break; + + size = round_up(size, sizeof(u64)); + if (!size || size > max_size) + break; + + *_size = size; + return 0; + + } while (0); + + pr_err("callchain: Incorrect stack dump size (max %ld): %s\n", + max_size, str); + return -1; +} +#endif /* LIBUNWIND_SUPPORT */ + +int record_parse_callchain_opt(const struct option *opt, + const char *arg, int unset) +{ + struct perf_record_opts *opts = opt->value; + char *tok, *name, *saveptr = NULL; + char *buf; + int ret = -1; + + /* --no-call-graph */ + if (unset) + return 0; + + /* We specified default option if none is provided. */ + BUG_ON(!arg); + + /* We need buffer that we know we can write to. */ + buf = malloc(strlen(arg) + 1); + if (!buf) + return -ENOMEM; + + strcpy(buf, arg); + + tok = strtok_r((char *)buf, ",", &saveptr); + name = tok ? : (char *)buf; + + do { + /* Framepointer style */ + if (!strncmp(name, "fp", sizeof("fp"))) { + if (!strtok_r(NULL, ",", &saveptr)) { + opts->call_graph = CALLCHAIN_FP; + ret = 0; + } else + pr_err("callchain: No more arguments " + "needed for -g fp\n"); + break; + +#ifdef LIBUNWIND_SUPPORT + /* Dwarf style */ + } else if (!strncmp(name, "dwarf", sizeof("dwarf"))) { + const unsigned long default_stack_dump_size = 8192; + + ret = 0; + opts->call_graph = CALLCHAIN_DWARF; + opts->stack_dump_size = default_stack_dump_size; + + tok = strtok_r(NULL, ",", &saveptr); + if (tok) { + unsigned long size = 0; + + ret = get_stack_size(tok, &size); + opts->stack_dump_size = size; + } + + if (!ret) + pr_debug("callchain: stack dump size %d\n", + opts->stack_dump_size); +#endif /* LIBUNWIND_SUPPORT */ + } else { + pr_err("callchain: Unknown -g option " + "value: %s\n", arg); + break; + } + + } while (0); + + free(buf); + + if (!ret) + pr_debug("callchain: type %d\n", opts->call_graph); + + return ret; +} + +static const char * const record_usage[] = { + "perf record [<options>] [<command>]", + "perf record [<options>] -- <command> [<options>]", + NULL +}; + +/* + * XXX Ideally would be local to cmd_record() and passed to a perf_record__new + * because we need to have access to it in perf_record__exit, that is called + * after cmd_record() exits, but since record_options need to be accessible to + * builtin-script, leave it here. + * + * At least we don't ouch it in all the other functions here directly. + * + * Just say no to tons of global variables, sigh. + */ +static struct perf_record record = { + .opts = { + .mmap_pages = UINT_MAX, + .user_freq = UINT_MAX, + .user_interval = ULLONG_MAX, + .freq = 4000, + .target = { + .uses_mmap = true, + }, + }, + .write_mode = WRITE_FORCE, + .file_new = true, +}; + +#define CALLCHAIN_HELP "do call-graph (stack chain/backtrace) recording: " + +#ifdef LIBUNWIND_SUPPORT +const char record_callchain_help[] = CALLCHAIN_HELP "[fp] dwarf"; +#else +const char record_callchain_help[] = CALLCHAIN_HELP "[fp]"; +#endif + +/* + * XXX Will stay a global variable till we fix builtin-script.c to stop messing + * with it and switch to use the library functions in perf_evlist that came + * from builtin-record.c, i.e. use perf_record_opts, + * perf_evlist__prepare_workload, etc instead of fork+exec'in 'perf record', + * using pipes, etc. + */ +const struct option record_options[] = { + OPT_CALLBACK('e', "event", &record.evlist, "event", + "event selector. use 'perf list' to list available events", + parse_events_option), + OPT_CALLBACK(0, "filter", &record.evlist, "filter", + "event filter", parse_filter), + OPT_STRING('p', "pid", &record.opts.target.pid, "pid", + "record events on existing process id"), + OPT_STRING('t', "tid", &record.opts.target.tid, "tid", + "record events on existing thread id"), + OPT_INTEGER('r', "realtime", &record.realtime_prio, + "collect data with this RT SCHED_FIFO priority"), + OPT_BOOLEAN('D', "no-delay", &record.opts.no_delay, + "collect data without buffering"), + OPT_BOOLEAN('R', "raw-samples", &record.opts.raw_samples, + "collect raw sample records from all opened counters"), + OPT_BOOLEAN('a', "all-cpus", &record.opts.target.system_wide, + "system-wide collection from all CPUs"), + OPT_BOOLEAN('A', "append", &record.append_file, + "append to the output file to do incremental profiling"), + OPT_STRING('C', "cpu", &record.opts.target.cpu_list, "cpu", + "list of cpus to monitor"), + OPT_BOOLEAN('f', "force", &record.force, + "overwrite existing data file (deprecated)"), + OPT_U64('c', "count", &record.opts.user_interval, "event period to sample"), + OPT_STRING('o', "output", &record.output_name, "file", + "output file name"), + OPT_BOOLEAN('i', "no-inherit", &record.opts.no_inherit, + "child tasks do not inherit counters"), + OPT_UINTEGER('F', "freq", &record.opts.user_freq, "profile at this frequency"), + OPT_UINTEGER('m', "mmap-pages", &record.opts.mmap_pages, + "number of mmap data pages"), + OPT_BOOLEAN(0, "group", &record.opts.group, + "put the counters into a counter group"), + OPT_CALLBACK_DEFAULT('g', "call-graph", &record.opts, + "mode[,dump_size]", record_callchain_help, + &record_parse_callchain_opt, "fp"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show counter open errors, etc)"), + OPT_BOOLEAN('q', "quiet", &quiet, "don't print any message"), + OPT_BOOLEAN('s', "stat", &record.opts.inherit_stat, + "per thread counts"), + OPT_BOOLEAN('d', "data", &record.opts.sample_address, + "Sample addresses"), + OPT_BOOLEAN('T', "timestamp", &record.opts.sample_time, "Sample timestamps"), + OPT_BOOLEAN('P', "period", &record.opts.period, "Sample period"), + OPT_BOOLEAN('n', "no-samples", &record.opts.no_samples, + "don't sample"), + OPT_BOOLEAN('N', "no-buildid-cache", &record.no_buildid_cache, + "do not update the buildid cache"), + OPT_BOOLEAN('B', "no-buildid", &record.no_buildid, + "do not collect buildids in perf.data"), + OPT_CALLBACK('G', "cgroup", &record.evlist, "name", + "monitor event in cgroup name only", + parse_cgroups), + OPT_STRING('u', "uid", &record.opts.target.uid_str, "user", + "user to profile"), + + OPT_CALLBACK_NOOPT('b', "branch-any", &record.opts.branch_stack, + "branch any", "sample any taken branches", + parse_branch_stack), + + OPT_CALLBACK('j', "branch-filter", &record.opts.branch_stack, + "branch filter mask", "branch stack filter modes", + parse_branch_stack), + OPT_END() +}; + +int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused) +{ + int err = -ENOMEM; + struct perf_evsel *pos; + struct perf_evlist *evsel_list; + struct perf_record *rec = &record; + char errbuf[BUFSIZ]; + + evsel_list = perf_evlist__new(NULL, NULL); + if (evsel_list == NULL) + return -ENOMEM; + + rec->evlist = evsel_list; + + argc = parse_options(argc, argv, record_options, record_usage, + PARSE_OPT_STOP_AT_NON_OPTION); + if (!argc && perf_target__none(&rec->opts.target)) + usage_with_options(record_usage, record_options); + + if (rec->force && rec->append_file) { + ui__error("Can't overwrite and append at the same time." + " You need to choose between -f and -A"); + usage_with_options(record_usage, record_options); + } else if (rec->append_file) { + rec->write_mode = WRITE_APPEND; + } else { + rec->write_mode = WRITE_FORCE; + } + + if (nr_cgroups && !rec->opts.target.system_wide) { + ui__error("cgroup monitoring only available in" + " system-wide mode\n"); + usage_with_options(record_usage, record_options); + } + + symbol__init(); + + if (symbol_conf.kptr_restrict) + pr_warning( +"WARNING: Kernel address maps (/proc/{kallsyms,modules}) are restricted,\n" +"check /proc/sys/kernel/kptr_restrict.\n\n" +"Samples in kernel functions may not be resolved if a suitable vmlinux\n" +"file is not found in the buildid cache or in the vmlinux path.\n\n" +"Samples in kernel modules won't be resolved at all.\n\n" +"If some relocation was applied (e.g. kexec) symbols may be misresolved\n" +"even with a suitable vmlinux or kallsyms file.\n\n"); + + if (rec->no_buildid_cache || rec->no_buildid) + disable_buildid_cache(); + + if (evsel_list->nr_entries == 0 && + perf_evlist__add_default(evsel_list) < 0) { + pr_err("Not enough memory for event selector list\n"); + goto out_symbol_exit; + } + + err = perf_target__validate(&rec->opts.target); + if (err) { + perf_target__strerror(&rec->opts.target, err, errbuf, BUFSIZ); + ui__warning("%s", errbuf); + } + + err = perf_target__parse_uid(&rec->opts.target); + if (err) { + int saved_errno = errno; + + perf_target__strerror(&rec->opts.target, err, errbuf, BUFSIZ); + ui__error("%s", errbuf); + + err = -saved_errno; + goto out_free_fd; + } + + err = -ENOMEM; + if (perf_evlist__create_maps(evsel_list, &rec->opts.target) < 0) + usage_with_options(record_usage, record_options); + + list_for_each_entry(pos, &evsel_list->entries, node) { + if (perf_header__push_event(pos->attr.config, perf_evsel__name(pos))) + goto out_free_fd; + } + + if (rec->opts.user_interval != ULLONG_MAX) + rec->opts.default_interval = rec->opts.user_interval; + if (rec->opts.user_freq != UINT_MAX) + rec->opts.freq = rec->opts.user_freq; + + /* + * User specified count overrides default frequency. + */ + if (rec->opts.default_interval) + rec->opts.freq = 0; + else if (rec->opts.freq) { + rec->opts.default_interval = rec->opts.freq; + } else { + ui__error("frequency and count are zero, aborting\n"); + err = -EINVAL; + goto out_free_fd; + } + + err = __cmd_record(&record, argc, argv); +out_free_fd: + perf_evlist__delete_maps(evsel_list); +out_symbol_exit: + symbol__exit(); + return err; +} diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c new file mode 100644 index 00000000..96b5a7fe --- /dev/null +++ b/tools/perf/builtin-report.c @@ -0,0 +1,832 @@ +/* + * 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/cache.h" + +#include "util/annotate.h" +#include "util/color.h" +#include <linux/list.h> +#include "util/cache.h" +#include <linux/rbtree.h> +#include "util/symbol.h" +#include "util/callchain.h" +#include "util/strlist.h" +#include "util/values.h" + +#include "perf.h" +#include "util/debug.h" +#include "util/evlist.h" +#include "util/evsel.h" +#include "util/header.h" +#include "util/session.h" +#include "util/tool.h" + +#include "util/parse-options.h" +#include "util/parse-events.h" + +#include "util/thread.h" +#include "util/sort.h" +#include "util/hist.h" +#include "arch/common.h" + +#include <linux/bitmap.h> + +struct perf_report { + struct perf_tool tool; + struct perf_session *session; + bool force, use_tui, use_gtk, use_stdio; + bool hide_unresolved; + bool dont_use_callchains; + bool show_full_info; + bool show_threads; + bool inverted_callchain; + struct perf_read_values show_threads_values; + const char *pretty_printing_style; + symbol_filter_t annotate_init; + const char *cpu_list; + const char *symbol_filter_str; + DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS); +}; + +static int perf_report_config(const char *var, const char *value, void *cb) +{ + if (!strcmp(var, "report.group")) { + symbol_conf.event_group = perf_config_bool(var, value); + return 0; + } + + return perf_default_config(var, value, cb); +} + +static int perf_report__add_branch_hist_entry(struct perf_tool *tool, + struct addr_location *al, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct perf_report *rep = container_of(tool, struct perf_report, tool); + struct symbol *parent = NULL; + int err = 0; + unsigned i; + struct hist_entry *he; + struct branch_info *bi, *bx; + + if ((sort__has_parent || symbol_conf.use_callchain) + && sample->callchain) { + err = machine__resolve_callchain(machine, evsel, al->thread, + sample, &parent); + if (err) + return err; + } + + bi = machine__resolve_bstack(machine, al->thread, + sample->branch_stack); + if (!bi) + return -ENOMEM; + + for (i = 0; i < sample->branch_stack->nr; i++) { + if (rep->hide_unresolved && !(bi[i].from.sym && bi[i].to.sym)) + continue; + /* + * The report shows the percentage of total branches captured + * and not events sampled. Thus we use a pseudo period of 1. + */ + he = __hists__add_branch_entry(&evsel->hists, al, parent, + &bi[i], 1); + if (he) { + struct annotation *notes; + err = -ENOMEM; + bx = he->branch_info; + if (bx->from.sym && use_browser == 1 && sort__has_sym) { + notes = symbol__annotation(bx->from.sym); + if (!notes->src + && symbol__alloc_hist(bx->from.sym) < 0) + goto out; + + err = symbol__inc_addr_samples(bx->from.sym, + bx->from.map, + evsel->idx, + bx->from.al_addr); + if (err) + goto out; + } + + if (bx->to.sym && use_browser == 1 && sort__has_sym) { + notes = symbol__annotation(bx->to.sym); + if (!notes->src + && symbol__alloc_hist(bx->to.sym) < 0) + goto out; + + err = symbol__inc_addr_samples(bx->to.sym, + bx->to.map, + evsel->idx, + bx->to.al_addr); + if (err) + goto out; + } + evsel->hists.stats.total_period += 1; + hists__inc_nr_events(&evsel->hists, PERF_RECORD_SAMPLE); + err = 0; + } else + return -ENOMEM; + } +out: + return err; +} + +static int perf_evsel__add_hist_entry(struct perf_evsel *evsel, + struct addr_location *al, + struct perf_sample *sample, + struct machine *machine) +{ + struct symbol *parent = NULL; + int err = 0; + struct hist_entry *he; + + if ((sort__has_parent || symbol_conf.use_callchain) && sample->callchain) { + err = machine__resolve_callchain(machine, evsel, al->thread, + sample, &parent); + if (err) + return err; + } + + he = __hists__add_entry(&evsel->hists, al, parent, sample->period); + if (he == NULL) + return -ENOMEM; + + if (symbol_conf.use_callchain) { + err = callchain_append(he->callchain, + &callchain_cursor, + sample->period); + if (err) + return err; + } + /* + * Only in the newt browser we are doing integrated annotation, + * so we don't allocated the extra space needed because the stdio + * code will not use it. + */ + if (he->ms.sym != NULL && use_browser == 1 && sort__has_sym) { + struct annotation *notes = symbol__annotation(he->ms.sym); + + assert(evsel != NULL); + + err = -ENOMEM; + if (notes->src == NULL && symbol__alloc_hist(he->ms.sym) < 0) + goto out; + + err = hist_entry__inc_addr_samples(he, evsel->idx, al->addr); + } + + evsel->hists.stats.total_period += sample->period; + hists__inc_nr_events(&evsel->hists, PERF_RECORD_SAMPLE); +out: + return err; +} + + +static int process_sample_event(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct perf_report *rep = container_of(tool, struct perf_report, tool); + struct addr_location al; + + if (perf_event__preprocess_sample(event, machine, &al, sample, + rep->annotate_init) < 0) { + fprintf(stderr, "problem processing %d event, skipping it.\n", + event->header.type); + return -1; + } + + if (al.filtered || (rep->hide_unresolved && al.sym == NULL)) + return 0; + + if (rep->cpu_list && !test_bit(sample->cpu, rep->cpu_bitmap)) + return 0; + + if (sort__branch_mode == 1) { + if (perf_report__add_branch_hist_entry(tool, &al, sample, + evsel, machine)) { + pr_debug("problem adding lbr entry, skipping event\n"); + return -1; + } + } else { + if (al.map != NULL) + al.map->dso->hit = 1; + + if (perf_evsel__add_hist_entry(evsel, &al, sample, machine)) { + pr_debug("problem incrementing symbol period, skipping event\n"); + return -1; + } + } + return 0; +} + +static int process_read_event(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct perf_evsel *evsel, + struct machine *machine __maybe_unused) +{ + struct perf_report *rep = container_of(tool, struct perf_report, tool); + + if (rep->show_threads) { + const char *name = evsel ? perf_evsel__name(evsel) : "unknown"; + perf_read_values_add_value(&rep->show_threads_values, + event->read.pid, event->read.tid, + event->read.id, + name, + event->read.value); + } + + dump_printf(": %d %d %s %" PRIu64 "\n", event->read.pid, event->read.tid, + evsel ? perf_evsel__name(evsel) : "FAIL", + event->read.value); + + return 0; +} + +/* For pipe mode, sample_type is not currently set */ +static int perf_report__setup_sample_type(struct perf_report *rep) +{ + struct perf_session *self = rep->session; + u64 sample_type = perf_evlist__sample_type(self->evlist); + + if (!self->fd_pipe && !(sample_type & PERF_SAMPLE_CALLCHAIN)) { + if (sort__has_parent) { + ui__error("Selected --sort parent, but no " + "callchain data. Did you call " + "'perf record' without -g?\n"); + return -EINVAL; + } + if (symbol_conf.use_callchain) { + ui__error("Selected -g but no callchain data. Did " + "you call 'perf record' without -g?\n"); + return -1; + } + } else if (!rep->dont_use_callchains && + callchain_param.mode != CHAIN_NONE && + !symbol_conf.use_callchain) { + symbol_conf.use_callchain = true; + if (callchain_register_param(&callchain_param) < 0) { + ui__error("Can't register callchain params.\n"); + return -EINVAL; + } + } + + if (sort__branch_mode == 1) { + if (!self->fd_pipe && + !(sample_type & PERF_SAMPLE_BRANCH_STACK)) { + ui__error("Selected -b but no branch data. " + "Did you call perf record without -b?\n"); + return -1; + } + } + + return 0; +} + +extern volatile int session_done; + +static void sig_handler(int sig __maybe_unused) +{ + session_done = 1; +} + +static size_t hists__fprintf_nr_sample_events(struct hists *self, + const char *evname, FILE *fp) +{ + size_t ret; + char unit; + unsigned long nr_samples = self->stats.nr_events[PERF_RECORD_SAMPLE]; + u64 nr_events = self->stats.total_period; + struct perf_evsel *evsel = hists_to_evsel(self); + char buf[512]; + size_t size = sizeof(buf); + + if (symbol_conf.event_group && evsel->nr_members > 1) { + struct perf_evsel *pos; + + perf_evsel__group_desc(evsel, buf, size); + evname = buf; + + for_each_group_member(pos, evsel) { + nr_samples += pos->hists.stats.nr_events[PERF_RECORD_SAMPLE]; + nr_events += pos->hists.stats.total_period; + } + } + + nr_samples = convert_unit(nr_samples, &unit); + ret = fprintf(fp, "# Samples: %lu%c", nr_samples, unit); + if (evname != NULL) + ret += fprintf(fp, " of event '%s'", evname); + + ret += fprintf(fp, "\n# Event count (approx.): %" PRIu64, nr_events); + return ret + fprintf(fp, "\n#\n"); +} + +static int perf_evlist__tty_browse_hists(struct perf_evlist *evlist, + struct perf_report *rep, + const char *help) +{ + struct perf_evsel *pos; + + list_for_each_entry(pos, &evlist->entries, node) { + struct hists *hists = &pos->hists; + const char *evname = perf_evsel__name(pos); + + if (symbol_conf.event_group && + !perf_evsel__is_group_leader(pos)) + continue; + + hists__fprintf_nr_sample_events(hists, evname, stdout); + hists__fprintf(hists, true, 0, 0, stdout); + fprintf(stdout, "\n\n"); + } + + if (sort_order == default_sort_order && + parent_pattern == default_parent_pattern) { + fprintf(stdout, "#\n# (%s)\n#\n", help); + + if (rep->show_threads) { + bool style = !strcmp(rep->pretty_printing_style, "raw"); + perf_read_values_display(stdout, &rep->show_threads_values, + style); + perf_read_values_destroy(&rep->show_threads_values); + } + } + + return 0; +} + +static int __cmd_report(struct perf_report *rep) +{ + int ret = -EINVAL; + u64 nr_samples; + struct perf_session *session = rep->session; + struct perf_evsel *pos; + struct map *kernel_map; + struct kmap *kernel_kmap; + const char *help = "For a higher level overview, try: perf report --sort comm,dso"; + + signal(SIGINT, sig_handler); + + if (rep->cpu_list) { + ret = perf_session__cpu_bitmap(session, rep->cpu_list, + rep->cpu_bitmap); + if (ret) + goto out_delete; + } + + if (use_browser <= 0) + perf_session__fprintf_info(session, stdout, rep->show_full_info); + + if (rep->show_threads) + perf_read_values_init(&rep->show_threads_values); + + ret = perf_report__setup_sample_type(rep); + if (ret) + goto out_delete; + + ret = perf_session__process_events(session, &rep->tool); + if (ret) + goto out_delete; + + kernel_map = session->machines.host.vmlinux_maps[MAP__FUNCTION]; + kernel_kmap = map__kmap(kernel_map); + if (kernel_map == NULL || + (kernel_map->dso->hit && + (kernel_kmap->ref_reloc_sym == NULL || + kernel_kmap->ref_reloc_sym->addr == 0))) { + const char *desc = + "As no suitable kallsyms nor vmlinux was found, kernel samples\n" + "can't be resolved."; + + if (kernel_map) { + const struct dso *kdso = kernel_map->dso; + if (!RB_EMPTY_ROOT(&kdso->symbols[MAP__FUNCTION])) { + desc = "If some relocation was applied (e.g. " + "kexec) symbols may be misresolved."; + } + } + + ui__warning( +"Kernel address maps (/proc/{kallsyms,modules}) were restricted.\n\n" +"Check /proc/sys/kernel/kptr_restrict before running 'perf record'.\n\n%s\n\n" +"Samples in kernel modules can't be resolved as well.\n\n", + desc); + } + + if (verbose > 3) + perf_session__fprintf(session, stdout); + + if (verbose > 2) + perf_session__fprintf_dsos(session, stdout); + + if (dump_trace) { + perf_session__fprintf_nr_events(session, stdout); + goto out_delete; + } + + nr_samples = 0; + list_for_each_entry(pos, &session->evlist->entries, node) { + struct hists *hists = &pos->hists; + + if (pos->idx == 0) + hists->symbol_filter_str = rep->symbol_filter_str; + + hists__collapse_resort(hists); + nr_samples += hists->stats.nr_events[PERF_RECORD_SAMPLE]; + + /* Non-group events are considered as leader */ + if (symbol_conf.event_group && + !perf_evsel__is_group_leader(pos)) { + struct hists *leader_hists = &pos->leader->hists; + + hists__match(leader_hists, hists); + hists__link(leader_hists, hists); + } + } + + if (nr_samples == 0) { + ui__error("The %s file has no samples!\n", session->filename); + goto out_delete; + } + + list_for_each_entry(pos, &session->evlist->entries, node) + hists__output_resort(&pos->hists); + + if (use_browser > 0) { + if (use_browser == 1) { + ret = perf_evlist__tui_browse_hists(session->evlist, + help, + NULL, + &session->header.env); + /* + * Usually "ret" is the last pressed key, and we only + * care if the key notifies us to switch data file. + */ + if (ret != K_SWITCH_INPUT_DATA) + ret = 0; + + } else if (use_browser == 2) { + perf_evlist__gtk_browse_hists(session->evlist, help, + NULL); + } + } else + perf_evlist__tty_browse_hists(session->evlist, rep, help); + +out_delete: + /* + * Speed up the exit process, for large files this can + * take quite a while. + * + * XXX Enable this when using valgrind or if we ever + * librarize this command. + * + * Also experiment with obstacks to see how much speed + * up we'll get here. + * + * perf_session__delete(session); + */ + return ret; +} + +static int +parse_callchain_opt(const struct option *opt, const char *arg, int unset) +{ + struct perf_report *rep = (struct perf_report *)opt->value; + char *tok, *tok2; + char *endptr; + + /* + * --no-call-graph + */ + if (unset) { + rep->dont_use_callchains = true; + return 0; + } + + symbol_conf.use_callchain = true; + + if (!arg) + return 0; + + tok = strtok((char *)arg, ","); + if (!tok) + return -1; + + /* get the output mode */ + if (!strncmp(tok, "graph", strlen(arg))) + callchain_param.mode = CHAIN_GRAPH_ABS; + + else if (!strncmp(tok, "flat", strlen(arg))) + callchain_param.mode = CHAIN_FLAT; + + else if (!strncmp(tok, "fractal", strlen(arg))) + callchain_param.mode = CHAIN_GRAPH_REL; + + else if (!strncmp(tok, "none", strlen(arg))) { + callchain_param.mode = CHAIN_NONE; + symbol_conf.use_callchain = false; + + return 0; + } + + else + return -1; + + /* get the min percentage */ + tok = strtok(NULL, ","); + if (!tok) + goto setup; + + callchain_param.min_percent = strtod(tok, &endptr); + if (tok == endptr) + return -1; + + /* get the print limit */ + tok2 = strtok(NULL, ","); + if (!tok2) + goto setup; + + if (tok2[0] != 'c') { + callchain_param.print_limit = strtoul(tok2, &endptr, 0); + tok2 = strtok(NULL, ","); + if (!tok2) + goto setup; + } + + /* get the call chain order */ + if (!strcmp(tok2, "caller")) + callchain_param.order = ORDER_CALLER; + else if (!strcmp(tok2, "callee")) + callchain_param.order = ORDER_CALLEE; + else + return -1; +setup: + if (callchain_register_param(&callchain_param) < 0) { + fprintf(stderr, "Can't register callchain params\n"); + return -1; + } + return 0; +} + +static int +parse_branch_mode(const struct option *opt __maybe_unused, + const char *str __maybe_unused, int unset) +{ + sort__branch_mode = !unset; + return 0; +} + +int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused) +{ + struct perf_session *session; + struct stat st; + bool has_br_stack = false; + int ret = -1; + char callchain_default_opt[] = "fractal,0.5,callee"; + const char * const report_usage[] = { + "perf report [<options>]", + NULL + }; + struct perf_report report = { + .tool = { + .sample = process_sample_event, + .mmap = perf_event__process_mmap, + .comm = perf_event__process_comm, + .exit = perf_event__process_exit, + .fork = perf_event__process_fork, + .lost = perf_event__process_lost, + .read = process_read_event, + .attr = perf_event__process_attr, + .event_type = perf_event__process_event_type, + .tracing_data = perf_event__process_tracing_data, + .build_id = perf_event__process_build_id, + .ordered_samples = true, + .ordering_requires_timestamps = true, + }, + .pretty_printing_style = "normal", + }; + const struct option options[] = { + OPT_STRING('i', "input", &input_name, "file", + "input file name"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show symbol address, etc)"), + OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, + "dump raw trace in ASCII"), + OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name, + "file", "vmlinux pathname"), + OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name, + "file", "kallsyms pathname"), + OPT_BOOLEAN('f', "force", &report.force, "don't complain, do it"), + OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules, + "load module symbols - WARNING: use only with -k and LIVE kernel"), + OPT_BOOLEAN('n', "show-nr-samples", &symbol_conf.show_nr_samples, + "Show a column with the number of samples"), + OPT_BOOLEAN('T', "threads", &report.show_threads, + "Show per-thread event counters"), + OPT_STRING(0, "pretty", &report.pretty_printing_style, "key", + "pretty printing style key: normal raw"), + OPT_BOOLEAN(0, "tui", &report.use_tui, "Use the TUI interface"), + OPT_BOOLEAN(0, "gtk", &report.use_gtk, "Use the GTK2 interface"), + OPT_BOOLEAN(0, "stdio", &report.use_stdio, + "Use the stdio interface"), + OPT_STRING('s', "sort", &sort_order, "key[,key2...]", + "sort by key(s): pid, comm, dso, symbol, parent, cpu, srcline," + " dso_to, dso_from, symbol_to, symbol_from, mispredict"), + OPT_BOOLEAN(0, "showcpuutilization", &symbol_conf.show_cpu_utilization, + "Show sample percentage for different cpu modes"), + OPT_STRING('p', "parent", &parent_pattern, "regex", + "regex filter to identify parent, see: '--sort parent'"), + OPT_BOOLEAN('x', "exclude-other", &symbol_conf.exclude_other, + "Only display entries with parent-match"), + OPT_CALLBACK_DEFAULT('g', "call-graph", &report, "output_type,min_percent[,print_limit],call_order", + "Display callchains using output_type (graph, flat, fractal, or none) , min percent threshold, optional print limit and callchain order. " + "Default: fractal,0.5,callee", &parse_callchain_opt, callchain_default_opt), + OPT_BOOLEAN('G', "inverted", &report.inverted_callchain, + "alias for inverted call graph"), + OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]", + "only consider symbols in these dsos"), + OPT_STRING('c', "comms", &symbol_conf.comm_list_str, "comm[,comm...]", + "only consider symbols in these comms"), + OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]", + "only consider these symbols"), + OPT_STRING(0, "symbol-filter", &report.symbol_filter_str, "filter", + "only show symbols that (partially) match with this filter"), + OPT_STRING('w', "column-widths", &symbol_conf.col_width_list_str, + "width[,width...]", + "don't try to adjust column width, use these fixed values"), + OPT_STRING('t', "field-separator", &symbol_conf.field_sep, "separator", + "separator for columns, no spaces will be added between " + "columns '.' is reserved."), + OPT_BOOLEAN('U', "hide-unresolved", &report.hide_unresolved, + "Only display entries resolved to a symbol"), + OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory", + "Look for files with symbols relative to this directory"), + OPT_STRING('C', "cpu", &report.cpu_list, "cpu", + "list of cpus to profile"), + OPT_BOOLEAN('I', "show-info", &report.show_full_info, + "Display extended information about perf.data file"), + OPT_BOOLEAN(0, "source", &symbol_conf.annotate_src, + "Interleave source code with assembly code (default)"), + OPT_BOOLEAN(0, "asm-raw", &symbol_conf.annotate_asm_raw, + "Display raw encoding of assembly instructions (default)"), + OPT_STRING('M', "disassembler-style", &disassembler_style, "disassembler style", + "Specify disassembler style (e.g. -M intel for intel syntax)"), + OPT_BOOLEAN(0, "show-total-period", &symbol_conf.show_total_period, + "Show a column with the sum of periods"), + OPT_BOOLEAN(0, "group", &symbol_conf.event_group, + "Show event group information together"), + OPT_CALLBACK_NOOPT('b', "branch-stack", &sort__branch_mode, "", + "use branch records for histogram filling", parse_branch_mode), + OPT_STRING(0, "objdump", &objdump_path, "path", + "objdump binary to use for disassembly and annotations"), + OPT_END() + }; + + perf_config(perf_report_config, NULL); + + argc = parse_options(argc, argv, options, report_usage, 0); + + if (report.use_stdio) + use_browser = 0; + else if (report.use_tui) + use_browser = 1; + else if (report.use_gtk) + use_browser = 2; + + if (report.inverted_callchain) + callchain_param.order = ORDER_CALLER; + + if (!input_name || !strlen(input_name)) { + if (!fstat(STDIN_FILENO, &st) && S_ISFIFO(st.st_mode)) + input_name = "-"; + else + input_name = "perf.data"; + } + + if (strcmp(input_name, "-") != 0) + setup_browser(true); + else { + use_browser = 0; + perf_hpp__column_enable(PERF_HPP__OVERHEAD); + perf_hpp__init(); + } + +repeat: + session = perf_session__new(input_name, O_RDONLY, + report.force, false, &report.tool); + if (session == NULL) + return -ENOMEM; + + report.session = session; + + has_br_stack = perf_header__has_feat(&session->header, + HEADER_BRANCH_STACK); + + if (sort__branch_mode == -1 && has_br_stack) + sort__branch_mode = 1; + + /* sort__branch_mode could be 0 if --no-branch-stack */ + if (sort__branch_mode == 1) { + /* + * if no sort_order is provided, then specify + * branch-mode specific order + */ + if (sort_order == default_sort_order) + sort_order = "comm,dso_from,symbol_from," + "dso_to,symbol_to"; + + } + + if (setup_sorting() < 0) + usage_with_options(report_usage, options); + + /* + * Only in the newt browser we are doing integrated annotation, + * so don't allocate extra space that won't be used in the stdio + * implementation. + */ + if (use_browser == 1 && sort__has_sym) { + symbol_conf.priv_size = sizeof(struct annotation); + report.annotate_init = symbol__annotate_init; + /* + * For searching by name on the "Browse map details". + * providing it only in verbose mode not to bloat too + * much struct symbol. + */ + if (verbose) { + /* + * XXX: Need to provide a less kludgy way to ask for + * more space per symbol, the u32 is for the index on + * the ui browser. + * See symbol__browser_index. + */ + symbol_conf.priv_size += sizeof(u32); + symbol_conf.sort_by_name = true; + } + } + + if (symbol__init() < 0) + goto error; + + if (parent_pattern != default_parent_pattern) { + if (sort_dimension__add("parent") < 0) + goto error; + + /* + * Only show the parent fields if we explicitly + * sort that way. If we only use parent machinery + * for filtering, we don't want it. + */ + if (!strstr(sort_order, "parent")) + sort_parent.elide = 1; + } else + symbol_conf.exclude_other = false; + + if (argc) { + /* + * Special case: if there's an argument left then assume that + * it's a symbol filter: + */ + if (argc > 1) + usage_with_options(report_usage, options); + + report.symbol_filter_str = argv[0]; + } + + sort_entry__setup_elide(&sort_comm, symbol_conf.comm_list, "comm", stdout); + + if (sort__branch_mode == 1) { + sort_entry__setup_elide(&sort_dso_from, symbol_conf.dso_from_list, "dso_from", stdout); + sort_entry__setup_elide(&sort_dso_to, symbol_conf.dso_to_list, "dso_to", stdout); + sort_entry__setup_elide(&sort_sym_from, symbol_conf.sym_from_list, "sym_from", stdout); + sort_entry__setup_elide(&sort_sym_to, symbol_conf.sym_to_list, "sym_to", stdout); + } else { + sort_entry__setup_elide(&sort_dso, symbol_conf.dso_list, "dso", stdout); + sort_entry__setup_elide(&sort_sym, symbol_conf.sym_list, "symbol", stdout); + } + + ret = __cmd_report(&report); + if (ret == K_SWITCH_INPUT_DATA) { + perf_session__delete(session); + goto repeat; + } else + ret = 0; + +error: + perf_session__delete(session); + return ret; +} diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c new file mode 100644 index 00000000..13822943 --- /dev/null +++ b/tools/perf/builtin-sched.c @@ -0,0 +1,1786 @@ +#include "builtin.h" +#include "perf.h" + +#include "util/util.h" +#include "util/evlist.h" +#include "util/cache.h" +#include "util/evsel.h" +#include "util/symbol.h" +#include "util/thread.h" +#include "util/header.h" +#include "util/session.h" +#include "util/tool.h" + +#include "util/parse-options.h" +#include "util/trace-event.h" + +#include "util/debug.h" + +#include <sys/prctl.h> +#include <sys/resource.h> + +#include <semaphore.h> +#include <pthread.h> +#include <math.h> + +#define PR_SET_NAME 15 /* Set process name */ +#define MAX_CPUS 4096 +#define COMM_LEN 20 +#define SYM_LEN 129 +#define MAX_PID 65536 + +struct sched_atom; + +struct task_desc { + unsigned long nr; + unsigned long pid; + char comm[COMM_LEN]; + + unsigned long nr_events; + unsigned long curr_event; + struct sched_atom **atoms; + + pthread_t thread; + sem_t sleep_sem; + + sem_t ready_for_work; + sem_t work_done_sem; + + u64 cpu_usage; +}; + +enum sched_event_type { + SCHED_EVENT_RUN, + SCHED_EVENT_SLEEP, + SCHED_EVENT_WAKEUP, + SCHED_EVENT_MIGRATION, +}; + +struct sched_atom { + enum sched_event_type type; + int specific_wait; + u64 timestamp; + u64 duration; + unsigned long nr; + sem_t *wait_sem; + struct task_desc *wakee; +}; + +#define TASK_STATE_TO_CHAR_STR "RSDTtZX" + +enum thread_state { + THREAD_SLEEPING = 0, + THREAD_WAIT_CPU, + THREAD_SCHED_IN, + THREAD_IGNORE +}; + +struct work_atom { + struct list_head list; + enum thread_state state; + u64 sched_out_time; + u64 wake_up_time; + u64 sched_in_time; + u64 runtime; +}; + +struct work_atoms { + struct list_head work_list; + struct thread *thread; + struct rb_node node; + u64 max_lat; + u64 max_lat_at; + u64 total_lat; + u64 nb_atoms; + u64 total_runtime; +}; + +typedef int (*sort_fn_t)(struct work_atoms *, struct work_atoms *); + +struct perf_sched; + +struct trace_sched_handler { + int (*switch_event)(struct perf_sched *sched, struct perf_evsel *evsel, + struct perf_sample *sample, struct machine *machine); + + int (*runtime_event)(struct perf_sched *sched, struct perf_evsel *evsel, + struct perf_sample *sample, struct machine *machine); + + int (*wakeup_event)(struct perf_sched *sched, struct perf_evsel *evsel, + struct perf_sample *sample, struct machine *machine); + + int (*fork_event)(struct perf_sched *sched, struct perf_evsel *evsel, + struct perf_sample *sample); + + int (*migrate_task_event)(struct perf_sched *sched, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine); +}; + +struct perf_sched { + struct perf_tool tool; + const char *sort_order; + unsigned long nr_tasks; + struct task_desc *pid_to_task[MAX_PID]; + struct task_desc **tasks; + const struct trace_sched_handler *tp_handler; + pthread_mutex_t start_work_mutex; + pthread_mutex_t work_done_wait_mutex; + int profile_cpu; +/* + * Track the current task - that way we can know whether there's any + * weird events, such as a task being switched away that is not current. + */ + int max_cpu; + u32 curr_pid[MAX_CPUS]; + struct thread *curr_thread[MAX_CPUS]; + char next_shortname1; + char next_shortname2; + unsigned int replay_repeat; + unsigned long nr_run_events; + unsigned long nr_sleep_events; + unsigned long nr_wakeup_events; + unsigned long nr_sleep_corrections; + unsigned long nr_run_events_optimized; + unsigned long targetless_wakeups; + unsigned long multitarget_wakeups; + unsigned long nr_runs; + unsigned long nr_timestamps; + unsigned long nr_unordered_timestamps; + unsigned long nr_state_machine_bugs; + unsigned long nr_context_switch_bugs; + unsigned long nr_events; + unsigned long nr_lost_chunks; + unsigned long nr_lost_events; + u64 run_measurement_overhead; + u64 sleep_measurement_overhead; + u64 start_time; + u64 cpu_usage; + u64 runavg_cpu_usage; + u64 parent_cpu_usage; + u64 runavg_parent_cpu_usage; + u64 sum_runtime; + u64 sum_fluct; + u64 run_avg; + u64 all_runtime; + u64 all_count; + u64 cpu_last_switched[MAX_CPUS]; + struct rb_root atom_root, sorted_atom_root; + struct list_head sort_list, cmp_pid; +}; + +static u64 get_nsecs(void) +{ + struct timespec ts; + + clock_gettime(CLOCK_MONOTONIC, &ts); + + return ts.tv_sec * 1000000000ULL + ts.tv_nsec; +} + +static void burn_nsecs(struct perf_sched *sched, u64 nsecs) +{ + u64 T0 = get_nsecs(), T1; + + do { + T1 = get_nsecs(); + } while (T1 + sched->run_measurement_overhead < T0 + nsecs); +} + +static void sleep_nsecs(u64 nsecs) +{ + struct timespec ts; + + ts.tv_nsec = nsecs % 999999999; + ts.tv_sec = nsecs / 999999999; + + nanosleep(&ts, NULL); +} + +static void calibrate_run_measurement_overhead(struct perf_sched *sched) +{ + u64 T0, T1, delta, min_delta = 1000000000ULL; + int i; + + for (i = 0; i < 10; i++) { + T0 = get_nsecs(); + burn_nsecs(sched, 0); + T1 = get_nsecs(); + delta = T1-T0; + min_delta = min(min_delta, delta); + } + sched->run_measurement_overhead = min_delta; + + printf("run measurement overhead: %" PRIu64 " nsecs\n", min_delta); +} + +static void calibrate_sleep_measurement_overhead(struct perf_sched *sched) +{ + u64 T0, T1, delta, min_delta = 1000000000ULL; + int i; + + for (i = 0; i < 10; i++) { + T0 = get_nsecs(); + sleep_nsecs(10000); + T1 = get_nsecs(); + delta = T1-T0; + min_delta = min(min_delta, delta); + } + min_delta -= 10000; + sched->sleep_measurement_overhead = min_delta; + + printf("sleep measurement overhead: %" PRIu64 " nsecs\n", min_delta); +} + +static struct sched_atom * +get_new_event(struct task_desc *task, u64 timestamp) +{ + struct sched_atom *event = zalloc(sizeof(*event)); + unsigned long idx = task->nr_events; + size_t size; + + event->timestamp = timestamp; + event->nr = idx; + + task->nr_events++; + size = sizeof(struct sched_atom *) * task->nr_events; + task->atoms = realloc(task->atoms, size); + BUG_ON(!task->atoms); + + task->atoms[idx] = event; + + return event; +} + +static struct sched_atom *last_event(struct task_desc *task) +{ + if (!task->nr_events) + return NULL; + + return task->atoms[task->nr_events - 1]; +} + +static void add_sched_event_run(struct perf_sched *sched, struct task_desc *task, + u64 timestamp, u64 duration) +{ + struct sched_atom *event, *curr_event = last_event(task); + + /* + * optimize an existing RUN event by merging this one + * to it: + */ + if (curr_event && curr_event->type == SCHED_EVENT_RUN) { + sched->nr_run_events_optimized++; + curr_event->duration += duration; + return; + } + + event = get_new_event(task, timestamp); + + event->type = SCHED_EVENT_RUN; + event->duration = duration; + + sched->nr_run_events++; +} + +static void add_sched_event_wakeup(struct perf_sched *sched, struct task_desc *task, + u64 timestamp, struct task_desc *wakee) +{ + struct sched_atom *event, *wakee_event; + + event = get_new_event(task, timestamp); + event->type = SCHED_EVENT_WAKEUP; + event->wakee = wakee; + + wakee_event = last_event(wakee); + if (!wakee_event || wakee_event->type != SCHED_EVENT_SLEEP) { + sched->targetless_wakeups++; + return; + } + if (wakee_event->wait_sem) { + sched->multitarget_wakeups++; + return; + } + + wakee_event->wait_sem = zalloc(sizeof(*wakee_event->wait_sem)); + sem_init(wakee_event->wait_sem, 0, 0); + wakee_event->specific_wait = 1; + event->wait_sem = wakee_event->wait_sem; + + sched->nr_wakeup_events++; +} + +static void add_sched_event_sleep(struct perf_sched *sched, struct task_desc *task, + u64 timestamp, u64 task_state __maybe_unused) +{ + struct sched_atom *event = get_new_event(task, timestamp); + + event->type = SCHED_EVENT_SLEEP; + + sched->nr_sleep_events++; +} + +static struct task_desc *register_pid(struct perf_sched *sched, + unsigned long pid, const char *comm) +{ + struct task_desc *task; + + BUG_ON(pid >= MAX_PID); + + task = sched->pid_to_task[pid]; + + if (task) + return task; + + task = zalloc(sizeof(*task)); + task->pid = pid; + task->nr = sched->nr_tasks; + strcpy(task->comm, comm); + /* + * every task starts in sleeping state - this gets ignored + * if there's no wakeup pointing to this sleep state: + */ + add_sched_event_sleep(sched, task, 0, 0); + + sched->pid_to_task[pid] = task; + sched->nr_tasks++; + sched->tasks = realloc(sched->tasks, sched->nr_tasks * sizeof(struct task_task *)); + BUG_ON(!sched->tasks); + sched->tasks[task->nr] = task; + + if (verbose) + printf("registered task #%ld, PID %ld (%s)\n", sched->nr_tasks, pid, comm); + + return task; +} + + +static void print_task_traces(struct perf_sched *sched) +{ + struct task_desc *task; + unsigned long i; + + for (i = 0; i < sched->nr_tasks; i++) { + task = sched->tasks[i]; + printf("task %6ld (%20s:%10ld), nr_events: %ld\n", + task->nr, task->comm, task->pid, task->nr_events); + } +} + +static void add_cross_task_wakeups(struct perf_sched *sched) +{ + struct task_desc *task1, *task2; + unsigned long i, j; + + for (i = 0; i < sched->nr_tasks; i++) { + task1 = sched->tasks[i]; + j = i + 1; + if (j == sched->nr_tasks) + j = 0; + task2 = sched->tasks[j]; + add_sched_event_wakeup(sched, task1, 0, task2); + } +} + +static void perf_sched__process_event(struct perf_sched *sched, + struct sched_atom *atom) +{ + int ret = 0; + + switch (atom->type) { + case SCHED_EVENT_RUN: + burn_nsecs(sched, atom->duration); + break; + case SCHED_EVENT_SLEEP: + if (atom->wait_sem) + ret = sem_wait(atom->wait_sem); + BUG_ON(ret); + break; + case SCHED_EVENT_WAKEUP: + if (atom->wait_sem) + ret = sem_post(atom->wait_sem); + BUG_ON(ret); + break; + case SCHED_EVENT_MIGRATION: + break; + default: + BUG_ON(1); + } +} + +static u64 get_cpu_usage_nsec_parent(void) +{ + struct rusage ru; + u64 sum; + int err; + + err = getrusage(RUSAGE_SELF, &ru); + BUG_ON(err); + + sum = ru.ru_utime.tv_sec*1e9 + ru.ru_utime.tv_usec*1e3; + sum += ru.ru_stime.tv_sec*1e9 + ru.ru_stime.tv_usec*1e3; + + return sum; +} + +static int self_open_counters(void) +{ + struct perf_event_attr attr; + int fd; + + memset(&attr, 0, sizeof(attr)); + + attr.type = PERF_TYPE_SOFTWARE; + attr.config = PERF_COUNT_SW_TASK_CLOCK; + + fd = sys_perf_event_open(&attr, 0, -1, -1, 0); + + if (fd < 0) + pr_err("Error: sys_perf_event_open() syscall returned " + "with %d (%s)\n", fd, strerror(errno)); + return fd; +} + +static u64 get_cpu_usage_nsec_self(int fd) +{ + u64 runtime; + int ret; + + ret = read(fd, &runtime, sizeof(runtime)); + BUG_ON(ret != sizeof(runtime)); + + return runtime; +} + +struct sched_thread_parms { + struct task_desc *task; + struct perf_sched *sched; +}; + +static void *thread_func(void *ctx) +{ + struct sched_thread_parms *parms = ctx; + struct task_desc *this_task = parms->task; + struct perf_sched *sched = parms->sched; + u64 cpu_usage_0, cpu_usage_1; + unsigned long i, ret; + char comm2[22]; + int fd; + + free(parms); + + sprintf(comm2, ":%s", this_task->comm); + prctl(PR_SET_NAME, comm2); + fd = self_open_counters(); + if (fd < 0) + return NULL; +again: + ret = sem_post(&this_task->ready_for_work); + BUG_ON(ret); + ret = pthread_mutex_lock(&sched->start_work_mutex); + BUG_ON(ret); + ret = pthread_mutex_unlock(&sched->start_work_mutex); + BUG_ON(ret); + + cpu_usage_0 = get_cpu_usage_nsec_self(fd); + + for (i = 0; i < this_task->nr_events; i++) { + this_task->curr_event = i; + perf_sched__process_event(sched, this_task->atoms[i]); + } + + cpu_usage_1 = get_cpu_usage_nsec_self(fd); + this_task->cpu_usage = cpu_usage_1 - cpu_usage_0; + ret = sem_post(&this_task->work_done_sem); + BUG_ON(ret); + + ret = pthread_mutex_lock(&sched->work_done_wait_mutex); + BUG_ON(ret); + ret = pthread_mutex_unlock(&sched->work_done_wait_mutex); + BUG_ON(ret); + + goto again; +} + +static void create_tasks(struct perf_sched *sched) +{ + struct task_desc *task; + pthread_attr_t attr; + unsigned long i; + int err; + + err = pthread_attr_init(&attr); + BUG_ON(err); + err = pthread_attr_setstacksize(&attr, + (size_t) max(16 * 1024, PTHREAD_STACK_MIN)); + BUG_ON(err); + err = pthread_mutex_lock(&sched->start_work_mutex); + BUG_ON(err); + err = pthread_mutex_lock(&sched->work_done_wait_mutex); + BUG_ON(err); + for (i = 0; i < sched->nr_tasks; i++) { + struct sched_thread_parms *parms = malloc(sizeof(*parms)); + BUG_ON(parms == NULL); + parms->task = task = sched->tasks[i]; + parms->sched = sched; + sem_init(&task->sleep_sem, 0, 0); + sem_init(&task->ready_for_work, 0, 0); + sem_init(&task->work_done_sem, 0, 0); + task->curr_event = 0; + err = pthread_create(&task->thread, &attr, thread_func, parms); + BUG_ON(err); + } +} + +static void wait_for_tasks(struct perf_sched *sched) +{ + u64 cpu_usage_0, cpu_usage_1; + struct task_desc *task; + unsigned long i, ret; + + sched->start_time = get_nsecs(); + sched->cpu_usage = 0; + pthread_mutex_unlock(&sched->work_done_wait_mutex); + + for (i = 0; i < sched->nr_tasks; i++) { + task = sched->tasks[i]; + ret = sem_wait(&task->ready_for_work); + BUG_ON(ret); + sem_init(&task->ready_for_work, 0, 0); + } + ret = pthread_mutex_lock(&sched->work_done_wait_mutex); + BUG_ON(ret); + + cpu_usage_0 = get_cpu_usage_nsec_parent(); + + pthread_mutex_unlock(&sched->start_work_mutex); + + for (i = 0; i < sched->nr_tasks; i++) { + task = sched->tasks[i]; + ret = sem_wait(&task->work_done_sem); + BUG_ON(ret); + sem_init(&task->work_done_sem, 0, 0); + sched->cpu_usage += task->cpu_usage; + task->cpu_usage = 0; + } + + cpu_usage_1 = get_cpu_usage_nsec_parent(); + if (!sched->runavg_cpu_usage) + sched->runavg_cpu_usage = sched->cpu_usage; + sched->runavg_cpu_usage = (sched->runavg_cpu_usage * 9 + sched->cpu_usage) / 10; + + sched->parent_cpu_usage = cpu_usage_1 - cpu_usage_0; + if (!sched->runavg_parent_cpu_usage) + sched->runavg_parent_cpu_usage = sched->parent_cpu_usage; + sched->runavg_parent_cpu_usage = (sched->runavg_parent_cpu_usage * 9 + + sched->parent_cpu_usage)/10; + + ret = pthread_mutex_lock(&sched->start_work_mutex); + BUG_ON(ret); + + for (i = 0; i < sched->nr_tasks; i++) { + task = sched->tasks[i]; + sem_init(&task->sleep_sem, 0, 0); + task->curr_event = 0; + } +} + +static void run_one_test(struct perf_sched *sched) +{ + u64 T0, T1, delta, avg_delta, fluct; + + T0 = get_nsecs(); + wait_for_tasks(sched); + T1 = get_nsecs(); + + delta = T1 - T0; + sched->sum_runtime += delta; + sched->nr_runs++; + + avg_delta = sched->sum_runtime / sched->nr_runs; + if (delta < avg_delta) + fluct = avg_delta - delta; + else + fluct = delta - avg_delta; + sched->sum_fluct += fluct; + if (!sched->run_avg) + sched->run_avg = delta; + sched->run_avg = (sched->run_avg * 9 + delta) / 10; + + printf("#%-3ld: %0.3f, ", sched->nr_runs, (double)delta / 1000000.0); + + printf("ravg: %0.2f, ", (double)sched->run_avg / 1e6); + + printf("cpu: %0.2f / %0.2f", + (double)sched->cpu_usage / 1e6, (double)sched->runavg_cpu_usage / 1e6); + +#if 0 + /* + * rusage statistics done by the parent, these are less + * accurate than the sched->sum_exec_runtime based statistics: + */ + printf(" [%0.2f / %0.2f]", + (double)sched->parent_cpu_usage/1e6, + (double)sched->runavg_parent_cpu_usage/1e6); +#endif + + printf("\n"); + + if (sched->nr_sleep_corrections) + printf(" (%ld sleep corrections)\n", sched->nr_sleep_corrections); + sched->nr_sleep_corrections = 0; +} + +static void test_calibrations(struct perf_sched *sched) +{ + u64 T0, T1; + + T0 = get_nsecs(); + burn_nsecs(sched, 1e6); + T1 = get_nsecs(); + + printf("the run test took %" PRIu64 " nsecs\n", T1 - T0); + + T0 = get_nsecs(); + sleep_nsecs(1e6); + T1 = get_nsecs(); + + printf("the sleep test took %" PRIu64 " nsecs\n", T1 - T0); +} + +static int +replay_wakeup_event(struct perf_sched *sched, + struct perf_evsel *evsel, struct perf_sample *sample, + struct machine *machine __maybe_unused) +{ + const char *comm = perf_evsel__strval(evsel, sample, "comm"); + const u32 pid = perf_evsel__intval(evsel, sample, "pid"); + struct task_desc *waker, *wakee; + + if (verbose) { + printf("sched_wakeup event %p\n", evsel); + + printf(" ... pid %d woke up %s/%d\n", sample->tid, comm, pid); + } + + waker = register_pid(sched, sample->tid, "<unknown>"); + wakee = register_pid(sched, pid, comm); + + add_sched_event_wakeup(sched, waker, sample->time, wakee); + return 0; +} + +static int replay_switch_event(struct perf_sched *sched, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine __maybe_unused) +{ + const char *prev_comm = perf_evsel__strval(evsel, sample, "prev_comm"), + *next_comm = perf_evsel__strval(evsel, sample, "next_comm"); + const u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"), + next_pid = perf_evsel__intval(evsel, sample, "next_pid"); + const u64 prev_state = perf_evsel__intval(evsel, sample, "prev_state"); + struct task_desc *prev, __maybe_unused *next; + u64 timestamp0, timestamp = sample->time; + int cpu = sample->cpu; + s64 delta; + + if (verbose) + printf("sched_switch event %p\n", evsel); + + if (cpu >= MAX_CPUS || cpu < 0) + return 0; + + timestamp0 = sched->cpu_last_switched[cpu]; + if (timestamp0) + delta = timestamp - timestamp0; + else + delta = 0; + + if (delta < 0) { + pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta); + return -1; + } + + pr_debug(" ... switch from %s/%d to %s/%d [ran %" PRIu64 " nsecs]\n", + prev_comm, prev_pid, next_comm, next_pid, delta); + + prev = register_pid(sched, prev_pid, prev_comm); + next = register_pid(sched, next_pid, next_comm); + + sched->cpu_last_switched[cpu] = timestamp; + + add_sched_event_run(sched, prev, timestamp, delta); + add_sched_event_sleep(sched, prev, timestamp, prev_state); + + return 0; +} + +static int replay_fork_event(struct perf_sched *sched, struct perf_evsel *evsel, + struct perf_sample *sample) +{ + const char *parent_comm = perf_evsel__strval(evsel, sample, "parent_comm"), + *child_comm = perf_evsel__strval(evsel, sample, "child_comm"); + const u32 parent_pid = perf_evsel__intval(evsel, sample, "parent_pid"), + child_pid = perf_evsel__intval(evsel, sample, "child_pid"); + + if (verbose) { + printf("sched_fork event %p\n", evsel); + printf("... parent: %s/%d\n", parent_comm, parent_pid); + printf("... child: %s/%d\n", child_comm, child_pid); + } + + register_pid(sched, parent_pid, parent_comm); + register_pid(sched, child_pid, child_comm); + return 0; +} + +struct sort_dimension { + const char *name; + sort_fn_t cmp; + struct list_head list; +}; + +static int +thread_lat_cmp(struct list_head *list, struct work_atoms *l, struct work_atoms *r) +{ + struct sort_dimension *sort; + int ret = 0; + + BUG_ON(list_empty(list)); + + list_for_each_entry(sort, list, list) { + ret = sort->cmp(l, r); + if (ret) + return ret; + } + + return ret; +} + +static struct work_atoms * +thread_atoms_search(struct rb_root *root, struct thread *thread, + struct list_head *sort_list) +{ + struct rb_node *node = root->rb_node; + struct work_atoms key = { .thread = thread }; + + while (node) { + struct work_atoms *atoms; + int cmp; + + atoms = container_of(node, struct work_atoms, node); + + cmp = thread_lat_cmp(sort_list, &key, atoms); + if (cmp > 0) + node = node->rb_left; + else if (cmp < 0) + node = node->rb_right; + else { + BUG_ON(thread != atoms->thread); + return atoms; + } + } + return NULL; +} + +static void +__thread_latency_insert(struct rb_root *root, struct work_atoms *data, + struct list_head *sort_list) +{ + struct rb_node **new = &(root->rb_node), *parent = NULL; + + while (*new) { + struct work_atoms *this; + int cmp; + + this = container_of(*new, struct work_atoms, node); + parent = *new; + + cmp = thread_lat_cmp(sort_list, data, this); + + if (cmp > 0) + new = &((*new)->rb_left); + else + new = &((*new)->rb_right); + } + + rb_link_node(&data->node, parent, new); + rb_insert_color(&data->node, root); +} + +static int thread_atoms_insert(struct perf_sched *sched, struct thread *thread) +{ + struct work_atoms *atoms = zalloc(sizeof(*atoms)); + if (!atoms) { + pr_err("No memory at %s\n", __func__); + return -1; + } + + atoms->thread = thread; + INIT_LIST_HEAD(&atoms->work_list); + __thread_latency_insert(&sched->atom_root, atoms, &sched->cmp_pid); + return 0; +} + +static int latency_fork_event(struct perf_sched *sched __maybe_unused, + struct perf_evsel *evsel __maybe_unused, + struct perf_sample *sample __maybe_unused) +{ + /* should insert the newcomer */ + return 0; +} + +static char sched_out_state(u64 prev_state) +{ + const char *str = TASK_STATE_TO_CHAR_STR; + + return str[prev_state]; +} + +static int +add_sched_out_event(struct work_atoms *atoms, + char run_state, + u64 timestamp) +{ + struct work_atom *atom = zalloc(sizeof(*atom)); + if (!atom) { + pr_err("Non memory at %s", __func__); + return -1; + } + + atom->sched_out_time = timestamp; + + if (run_state == 'R') { + atom->state = THREAD_WAIT_CPU; + atom->wake_up_time = atom->sched_out_time; + } + + list_add_tail(&atom->list, &atoms->work_list); + return 0; +} + +static void +add_runtime_event(struct work_atoms *atoms, u64 delta, + u64 timestamp __maybe_unused) +{ + struct work_atom *atom; + + BUG_ON(list_empty(&atoms->work_list)); + + atom = list_entry(atoms->work_list.prev, struct work_atom, list); + + atom->runtime += delta; + atoms->total_runtime += delta; +} + +static void +add_sched_in_event(struct work_atoms *atoms, u64 timestamp) +{ + struct work_atom *atom; + u64 delta; + + if (list_empty(&atoms->work_list)) + return; + + atom = list_entry(atoms->work_list.prev, struct work_atom, list); + + if (atom->state != THREAD_WAIT_CPU) + return; + + if (timestamp < atom->wake_up_time) { + atom->state = THREAD_IGNORE; + return; + } + + atom->state = THREAD_SCHED_IN; + atom->sched_in_time = timestamp; + + delta = atom->sched_in_time - atom->wake_up_time; + atoms->total_lat += delta; + if (delta > atoms->max_lat) { + atoms->max_lat = delta; + atoms->max_lat_at = timestamp; + } + atoms->nb_atoms++; +} + +static int latency_switch_event(struct perf_sched *sched, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine) +{ + const u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"), + next_pid = perf_evsel__intval(evsel, sample, "next_pid"); + const u64 prev_state = perf_evsel__intval(evsel, sample, "prev_state"); + struct work_atoms *out_events, *in_events; + struct thread *sched_out, *sched_in; + u64 timestamp0, timestamp = sample->time; + int cpu = sample->cpu; + s64 delta; + + BUG_ON(cpu >= MAX_CPUS || cpu < 0); + + timestamp0 = sched->cpu_last_switched[cpu]; + sched->cpu_last_switched[cpu] = timestamp; + if (timestamp0) + delta = timestamp - timestamp0; + else + delta = 0; + + if (delta < 0) { + pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta); + return -1; + } + + sched_out = machine__findnew_thread(machine, prev_pid); + sched_in = machine__findnew_thread(machine, next_pid); + + out_events = thread_atoms_search(&sched->atom_root, sched_out, &sched->cmp_pid); + if (!out_events) { + if (thread_atoms_insert(sched, sched_out)) + return -1; + out_events = thread_atoms_search(&sched->atom_root, sched_out, &sched->cmp_pid); + if (!out_events) { + pr_err("out-event: Internal tree error"); + return -1; + } + } + if (add_sched_out_event(out_events, sched_out_state(prev_state), timestamp)) + return -1; + + in_events = thread_atoms_search(&sched->atom_root, sched_in, &sched->cmp_pid); + if (!in_events) { + if (thread_atoms_insert(sched, sched_in)) + return -1; + in_events = thread_atoms_search(&sched->atom_root, sched_in, &sched->cmp_pid); + if (!in_events) { + pr_err("in-event: Internal tree error"); + return -1; + } + /* + * Take came in we have not heard about yet, + * add in an initial atom in runnable state: + */ + if (add_sched_out_event(in_events, 'R', timestamp)) + return -1; + } + add_sched_in_event(in_events, timestamp); + + return 0; +} + +static int latency_runtime_event(struct perf_sched *sched, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine) +{ + const u32 pid = perf_evsel__intval(evsel, sample, "pid"); + const u64 runtime = perf_evsel__intval(evsel, sample, "runtime"); + struct thread *thread = machine__findnew_thread(machine, pid); + struct work_atoms *atoms = thread_atoms_search(&sched->atom_root, thread, &sched->cmp_pid); + u64 timestamp = sample->time; + int cpu = sample->cpu; + + BUG_ON(cpu >= MAX_CPUS || cpu < 0); + if (!atoms) { + if (thread_atoms_insert(sched, thread)) + return -1; + atoms = thread_atoms_search(&sched->atom_root, thread, &sched->cmp_pid); + if (!atoms) { + pr_err("in-event: Internal tree error"); + return -1; + } + if (add_sched_out_event(atoms, 'R', timestamp)) + return -1; + } + + add_runtime_event(atoms, runtime, timestamp); + return 0; +} + +static int latency_wakeup_event(struct perf_sched *sched, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine) +{ + const u32 pid = perf_evsel__intval(evsel, sample, "pid"), + success = perf_evsel__intval(evsel, sample, "success"); + struct work_atoms *atoms; + struct work_atom *atom; + struct thread *wakee; + u64 timestamp = sample->time; + + /* Note for later, it may be interesting to observe the failing cases */ + if (!success) + return 0; + + wakee = machine__findnew_thread(machine, pid); + atoms = thread_atoms_search(&sched->atom_root, wakee, &sched->cmp_pid); + if (!atoms) { + if (thread_atoms_insert(sched, wakee)) + return -1; + atoms = thread_atoms_search(&sched->atom_root, wakee, &sched->cmp_pid); + if (!atoms) { + pr_err("wakeup-event: Internal tree error"); + return -1; + } + if (add_sched_out_event(atoms, 'S', timestamp)) + return -1; + } + + BUG_ON(list_empty(&atoms->work_list)); + + atom = list_entry(atoms->work_list.prev, struct work_atom, list); + + /* + * You WILL be missing events if you've recorded only + * one CPU, or are only looking at only one, so don't + * make useless noise. + */ + if (sched->profile_cpu == -1 && atom->state != THREAD_SLEEPING) + sched->nr_state_machine_bugs++; + + sched->nr_timestamps++; + if (atom->sched_out_time > timestamp) { + sched->nr_unordered_timestamps++; + return 0; + } + + atom->state = THREAD_WAIT_CPU; + atom->wake_up_time = timestamp; + return 0; +} + +static int latency_migrate_task_event(struct perf_sched *sched, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine) +{ + const u32 pid = perf_evsel__intval(evsel, sample, "pid"); + u64 timestamp = sample->time; + struct work_atoms *atoms; + struct work_atom *atom; + struct thread *migrant; + + /* + * Only need to worry about migration when profiling one CPU. + */ + if (sched->profile_cpu == -1) + return 0; + + migrant = machine__findnew_thread(machine, pid); + atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid); + if (!atoms) { + if (thread_atoms_insert(sched, migrant)) + return -1; + register_pid(sched, migrant->pid, migrant->comm); + atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid); + if (!atoms) { + pr_err("migration-event: Internal tree error"); + return -1; + } + if (add_sched_out_event(atoms, 'R', timestamp)) + return -1; + } + + BUG_ON(list_empty(&atoms->work_list)); + + atom = list_entry(atoms->work_list.prev, struct work_atom, list); + atom->sched_in_time = atom->sched_out_time = atom->wake_up_time = timestamp; + + sched->nr_timestamps++; + + if (atom->sched_out_time > timestamp) + sched->nr_unordered_timestamps++; + + return 0; +} + +static void output_lat_thread(struct perf_sched *sched, struct work_atoms *work_list) +{ + int i; + int ret; + u64 avg; + + if (!work_list->nb_atoms) + return; + /* + * Ignore idle threads: + */ + if (!strcmp(work_list->thread->comm, "swapper")) + return; + + sched->all_runtime += work_list->total_runtime; + sched->all_count += work_list->nb_atoms; + + ret = printf(" %s:%d ", work_list->thread->comm, work_list->thread->pid); + + for (i = 0; i < 24 - ret; i++) + printf(" "); + + avg = work_list->total_lat / work_list->nb_atoms; + + printf("|%11.3f ms |%9" PRIu64 " | avg:%9.3f ms | max:%9.3f ms | max at: %9.6f s\n", + (double)work_list->total_runtime / 1e6, + work_list->nb_atoms, (double)avg / 1e6, + (double)work_list->max_lat / 1e6, + (double)work_list->max_lat_at / 1e9); +} + +static int pid_cmp(struct work_atoms *l, struct work_atoms *r) +{ + if (l->thread->pid < r->thread->pid) + return -1; + if (l->thread->pid > r->thread->pid) + return 1; + + return 0; +} + +static int avg_cmp(struct work_atoms *l, struct work_atoms *r) +{ + u64 avgl, avgr; + + if (!l->nb_atoms) + return -1; + + if (!r->nb_atoms) + return 1; + + avgl = l->total_lat / l->nb_atoms; + avgr = r->total_lat / r->nb_atoms; + + if (avgl < avgr) + return -1; + if (avgl > avgr) + return 1; + + return 0; +} + +static int max_cmp(struct work_atoms *l, struct work_atoms *r) +{ + if (l->max_lat < r->max_lat) + return -1; + if (l->max_lat > r->max_lat) + return 1; + + return 0; +} + +static int switch_cmp(struct work_atoms *l, struct work_atoms *r) +{ + if (l->nb_atoms < r->nb_atoms) + return -1; + if (l->nb_atoms > r->nb_atoms) + return 1; + + return 0; +} + +static int runtime_cmp(struct work_atoms *l, struct work_atoms *r) +{ + if (l->total_runtime < r->total_runtime) + return -1; + if (l->total_runtime > r->total_runtime) + return 1; + + return 0; +} + +static int sort_dimension__add(const char *tok, struct list_head *list) +{ + size_t i; + static struct sort_dimension avg_sort_dimension = { + .name = "avg", + .cmp = avg_cmp, + }; + static struct sort_dimension max_sort_dimension = { + .name = "max", + .cmp = max_cmp, + }; + static struct sort_dimension pid_sort_dimension = { + .name = "pid", + .cmp = pid_cmp, + }; + static struct sort_dimension runtime_sort_dimension = { + .name = "runtime", + .cmp = runtime_cmp, + }; + static struct sort_dimension switch_sort_dimension = { + .name = "switch", + .cmp = switch_cmp, + }; + struct sort_dimension *available_sorts[] = { + &pid_sort_dimension, + &avg_sort_dimension, + &max_sort_dimension, + &switch_sort_dimension, + &runtime_sort_dimension, + }; + + for (i = 0; i < ARRAY_SIZE(available_sorts); i++) { + if (!strcmp(available_sorts[i]->name, tok)) { + list_add_tail(&available_sorts[i]->list, list); + + return 0; + } + } + + return -1; +} + +static void perf_sched__sort_lat(struct perf_sched *sched) +{ + struct rb_node *node; + + for (;;) { + struct work_atoms *data; + node = rb_first(&sched->atom_root); + if (!node) + break; + + rb_erase(node, &sched->atom_root); + data = rb_entry(node, struct work_atoms, node); + __thread_latency_insert(&sched->sorted_atom_root, data, &sched->sort_list); + } +} + +static int process_sched_wakeup_event(struct perf_tool *tool, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine) +{ + struct perf_sched *sched = container_of(tool, struct perf_sched, tool); + + if (sched->tp_handler->wakeup_event) + return sched->tp_handler->wakeup_event(sched, evsel, sample, machine); + + return 0; +} + +static int map_switch_event(struct perf_sched *sched, struct perf_evsel *evsel, + struct perf_sample *sample, struct machine *machine) +{ + const u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"), + next_pid = perf_evsel__intval(evsel, sample, "next_pid"); + struct thread *sched_out __maybe_unused, *sched_in; + int new_shortname; + u64 timestamp0, timestamp = sample->time; + s64 delta; + int cpu, this_cpu = sample->cpu; + + BUG_ON(this_cpu >= MAX_CPUS || this_cpu < 0); + + if (this_cpu > sched->max_cpu) + sched->max_cpu = this_cpu; + + timestamp0 = sched->cpu_last_switched[this_cpu]; + sched->cpu_last_switched[this_cpu] = timestamp; + if (timestamp0) + delta = timestamp - timestamp0; + else + delta = 0; + + if (delta < 0) { + pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta); + return -1; + } + + sched_out = machine__findnew_thread(machine, prev_pid); + sched_in = machine__findnew_thread(machine, next_pid); + + sched->curr_thread[this_cpu] = sched_in; + + printf(" "); + + new_shortname = 0; + if (!sched_in->shortname[0]) { + sched_in->shortname[0] = sched->next_shortname1; + sched_in->shortname[1] = sched->next_shortname2; + + if (sched->next_shortname1 < 'Z') { + sched->next_shortname1++; + } else { + sched->next_shortname1='A'; + if (sched->next_shortname2 < '9') { + sched->next_shortname2++; + } else { + sched->next_shortname2='0'; + } + } + new_shortname = 1; + } + + for (cpu = 0; cpu <= sched->max_cpu; cpu++) { + if (cpu != this_cpu) + printf(" "); + else + printf("*"); + + if (sched->curr_thread[cpu]) { + if (sched->curr_thread[cpu]->pid) + printf("%2s ", sched->curr_thread[cpu]->shortname); + else + printf(". "); + } else + printf(" "); + } + + printf(" %12.6f secs ", (double)timestamp/1e9); + if (new_shortname) { + printf("%s => %s:%d\n", + sched_in->shortname, sched_in->comm, sched_in->pid); + } else { + printf("\n"); + } + + return 0; +} + +static int process_sched_switch_event(struct perf_tool *tool, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine) +{ + struct perf_sched *sched = container_of(tool, struct perf_sched, tool); + int this_cpu = sample->cpu, err = 0; + u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"), + next_pid = perf_evsel__intval(evsel, sample, "next_pid"); + + if (sched->curr_pid[this_cpu] != (u32)-1) { + /* + * Are we trying to switch away a PID that is + * not current? + */ + if (sched->curr_pid[this_cpu] != prev_pid) + sched->nr_context_switch_bugs++; + } + + if (sched->tp_handler->switch_event) + err = sched->tp_handler->switch_event(sched, evsel, sample, machine); + + sched->curr_pid[this_cpu] = next_pid; + return err; +} + +static int process_sched_runtime_event(struct perf_tool *tool, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine) +{ + struct perf_sched *sched = container_of(tool, struct perf_sched, tool); + + if (sched->tp_handler->runtime_event) + return sched->tp_handler->runtime_event(sched, evsel, sample, machine); + + return 0; +} + +static int process_sched_fork_event(struct perf_tool *tool, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine __maybe_unused) +{ + struct perf_sched *sched = container_of(tool, struct perf_sched, tool); + + if (sched->tp_handler->fork_event) + return sched->tp_handler->fork_event(sched, evsel, sample); + + return 0; +} + +static int process_sched_exit_event(struct perf_tool *tool __maybe_unused, + struct perf_evsel *evsel, + struct perf_sample *sample __maybe_unused, + struct machine *machine __maybe_unused) +{ + pr_debug("sched_exit event %p\n", evsel); + return 0; +} + +static int process_sched_migrate_task_event(struct perf_tool *tool, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine) +{ + struct perf_sched *sched = container_of(tool, struct perf_sched, tool); + + if (sched->tp_handler->migrate_task_event) + return sched->tp_handler->migrate_task_event(sched, evsel, sample, machine); + + return 0; +} + +typedef int (*tracepoint_handler)(struct perf_tool *tool, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine); + +static int perf_sched__process_tracepoint_sample(struct perf_tool *tool __maybe_unused, + union perf_event *event __maybe_unused, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct thread *thread = machine__findnew_thread(machine, sample->tid); + int err = 0; + + if (thread == NULL) { + pr_debug("problem processing %s event, skipping it.\n", + perf_evsel__name(evsel)); + return -1; + } + + evsel->hists.stats.total_period += sample->period; + hists__inc_nr_events(&evsel->hists, PERF_RECORD_SAMPLE); + + if (evsel->handler.func != NULL) { + tracepoint_handler f = evsel->handler.func; + err = f(tool, evsel, sample, machine); + } + + return err; +} + +static int perf_sched__read_events(struct perf_sched *sched, bool destroy, + struct perf_session **psession) +{ + const struct perf_evsel_str_handler handlers[] = { + { "sched:sched_switch", process_sched_switch_event, }, + { "sched:sched_stat_runtime", process_sched_runtime_event, }, + { "sched:sched_wakeup", process_sched_wakeup_event, }, + { "sched:sched_wakeup_new", process_sched_wakeup_event, }, + { "sched:sched_process_fork", process_sched_fork_event, }, + { "sched:sched_process_exit", process_sched_exit_event, }, + { "sched:sched_migrate_task", process_sched_migrate_task_event, }, + }; + struct perf_session *session; + + session = perf_session__new(input_name, O_RDONLY, 0, false, &sched->tool); + if (session == NULL) { + pr_debug("No Memory for session\n"); + return -1; + } + + if (perf_session__set_tracepoints_handlers(session, handlers)) + goto out_delete; + + if (perf_session__has_traces(session, "record -R")) { + int err = perf_session__process_events(session, &sched->tool); + if (err) { + pr_err("Failed to process events, error %d", err); + goto out_delete; + } + + sched->nr_events = session->stats.nr_events[0]; + sched->nr_lost_events = session->stats.total_lost; + sched->nr_lost_chunks = session->stats.nr_events[PERF_RECORD_LOST]; + } + + if (destroy) + perf_session__delete(session); + + if (psession) + *psession = session; + + return 0; + +out_delete: + perf_session__delete(session); + return -1; +} + +static void print_bad_events(struct perf_sched *sched) +{ + if (sched->nr_unordered_timestamps && sched->nr_timestamps) { + printf(" INFO: %.3f%% unordered timestamps (%ld out of %ld)\n", + (double)sched->nr_unordered_timestamps/(double)sched->nr_timestamps*100.0, + sched->nr_unordered_timestamps, sched->nr_timestamps); + } + if (sched->nr_lost_events && sched->nr_events) { + printf(" INFO: %.3f%% lost events (%ld out of %ld, in %ld chunks)\n", + (double)sched->nr_lost_events/(double)sched->nr_events * 100.0, + sched->nr_lost_events, sched->nr_events, sched->nr_lost_chunks); + } + if (sched->nr_state_machine_bugs && sched->nr_timestamps) { + printf(" INFO: %.3f%% state machine bugs (%ld out of %ld)", + (double)sched->nr_state_machine_bugs/(double)sched->nr_timestamps*100.0, + sched->nr_state_machine_bugs, sched->nr_timestamps); + if (sched->nr_lost_events) + printf(" (due to lost events?)"); + printf("\n"); + } + if (sched->nr_context_switch_bugs && sched->nr_timestamps) { + printf(" INFO: %.3f%% context switch bugs (%ld out of %ld)", + (double)sched->nr_context_switch_bugs/(double)sched->nr_timestamps*100.0, + sched->nr_context_switch_bugs, sched->nr_timestamps); + if (sched->nr_lost_events) + printf(" (due to lost events?)"); + printf("\n"); + } +} + +static int perf_sched__lat(struct perf_sched *sched) +{ + struct rb_node *next; + struct perf_session *session; + + setup_pager(); + if (perf_sched__read_events(sched, false, &session)) + return -1; + perf_sched__sort_lat(sched); + + printf("\n ---------------------------------------------------------------------------------------------------------------\n"); + printf(" Task | Runtime ms | Switches | Average delay ms | Maximum delay ms | Maximum delay at |\n"); + printf(" ---------------------------------------------------------------------------------------------------------------\n"); + + next = rb_first(&sched->sorted_atom_root); + + while (next) { + struct work_atoms *work_list; + + work_list = rb_entry(next, struct work_atoms, node); + output_lat_thread(sched, work_list); + next = rb_next(next); + } + + printf(" -----------------------------------------------------------------------------------------\n"); + printf(" TOTAL: |%11.3f ms |%9" PRIu64 " |\n", + (double)sched->all_runtime / 1e6, sched->all_count); + + printf(" ---------------------------------------------------\n"); + + print_bad_events(sched); + printf("\n"); + + perf_session__delete(session); + return 0; +} + +static int perf_sched__map(struct perf_sched *sched) +{ + sched->max_cpu = sysconf(_SC_NPROCESSORS_CONF); + + setup_pager(); + if (perf_sched__read_events(sched, true, NULL)) + return -1; + print_bad_events(sched); + return 0; +} + +static int perf_sched__replay(struct perf_sched *sched) +{ + unsigned long i; + + calibrate_run_measurement_overhead(sched); + calibrate_sleep_measurement_overhead(sched); + + test_calibrations(sched); + + if (perf_sched__read_events(sched, true, NULL)) + return -1; + + printf("nr_run_events: %ld\n", sched->nr_run_events); + printf("nr_sleep_events: %ld\n", sched->nr_sleep_events); + printf("nr_wakeup_events: %ld\n", sched->nr_wakeup_events); + + if (sched->targetless_wakeups) + printf("target-less wakeups: %ld\n", sched->targetless_wakeups); + if (sched->multitarget_wakeups) + printf("multi-target wakeups: %ld\n", sched->multitarget_wakeups); + if (sched->nr_run_events_optimized) + printf("run atoms optimized: %ld\n", + sched->nr_run_events_optimized); + + print_task_traces(sched); + add_cross_task_wakeups(sched); + + create_tasks(sched); + printf("------------------------------------------------------------\n"); + for (i = 0; i < sched->replay_repeat; i++) + run_one_test(sched); + + return 0; +} + +static void setup_sorting(struct perf_sched *sched, const struct option *options, + const char * const usage_msg[]) +{ + char *tmp, *tok, *str = strdup(sched->sort_order); + + for (tok = strtok_r(str, ", ", &tmp); + tok; tok = strtok_r(NULL, ", ", &tmp)) { + if (sort_dimension__add(tok, &sched->sort_list) < 0) { + error("Unknown --sort key: `%s'", tok); + usage_with_options(usage_msg, options); + } + } + + free(str); + + sort_dimension__add("pid", &sched->cmp_pid); +} + +static int __cmd_record(int argc, const char **argv) +{ + unsigned int rec_argc, i, j; + const char **rec_argv; + const char * const record_args[] = { + "record", + "-a", + "-R", + "-f", + "-m", "1024", + "-c", "1", + "-e", "sched:sched_switch", + "-e", "sched:sched_stat_wait", + "-e", "sched:sched_stat_sleep", + "-e", "sched:sched_stat_iowait", + "-e", "sched:sched_stat_runtime", + "-e", "sched:sched_process_exit", + "-e", "sched:sched_process_fork", + "-e", "sched:sched_wakeup", + "-e", "sched:sched_migrate_task", + }; + + rec_argc = ARRAY_SIZE(record_args) + argc - 1; + rec_argv = calloc(rec_argc + 1, sizeof(char *)); + + if (rec_argv == NULL) + return -ENOMEM; + + for (i = 0; i < ARRAY_SIZE(record_args); i++) + rec_argv[i] = strdup(record_args[i]); + + for (j = 1; j < (unsigned int)argc; j++, i++) + rec_argv[i] = argv[j]; + + BUG_ON(i != rec_argc); + + return cmd_record(i, rec_argv, NULL); +} + +int cmd_sched(int argc, const char **argv, const char *prefix __maybe_unused) +{ + const char default_sort_order[] = "avg, max, switch, runtime"; + struct perf_sched sched = { + .tool = { + .sample = perf_sched__process_tracepoint_sample, + .comm = perf_event__process_comm, + .lost = perf_event__process_lost, + .exit = perf_event__process_exit, + .fork = perf_event__process_fork, + .ordered_samples = true, + }, + .cmp_pid = LIST_HEAD_INIT(sched.cmp_pid), + .sort_list = LIST_HEAD_INIT(sched.sort_list), + .start_work_mutex = PTHREAD_MUTEX_INITIALIZER, + .work_done_wait_mutex = PTHREAD_MUTEX_INITIALIZER, + .curr_pid = { [0 ... MAX_CPUS - 1] = -1 }, + .sort_order = default_sort_order, + .replay_repeat = 10, + .profile_cpu = -1, + .next_shortname1 = 'A', + .next_shortname2 = '0', + }; + const struct option latency_options[] = { + OPT_STRING('s', "sort", &sched.sort_order, "key[,key2...]", + "sort by key(s): runtime, switch, avg, max"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show symbol address, etc)"), + OPT_INTEGER('C', "CPU", &sched.profile_cpu, + "CPU to profile on"), + OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, + "dump raw trace in ASCII"), + OPT_END() + }; + const struct option replay_options[] = { + OPT_UINTEGER('r', "repeat", &sched.replay_repeat, + "repeat the workload replay N times (-1: infinite)"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show symbol address, etc)"), + OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, + "dump raw trace in ASCII"), + OPT_END() + }; + const struct option sched_options[] = { + OPT_STRING('i', "input", &input_name, "file", + "input file name"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show symbol address, etc)"), + OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, + "dump raw trace in ASCII"), + OPT_END() + }; + const char * const latency_usage[] = { + "perf sched latency [<options>]", + NULL + }; + const char * const replay_usage[] = { + "perf sched replay [<options>]", + NULL + }; + const char * const sched_usage[] = { + "perf sched [<options>] {record|latency|map|replay|script}", + NULL + }; + struct trace_sched_handler lat_ops = { + .wakeup_event = latency_wakeup_event, + .switch_event = latency_switch_event, + .runtime_event = latency_runtime_event, + .fork_event = latency_fork_event, + .migrate_task_event = latency_migrate_task_event, + }; + struct trace_sched_handler map_ops = { + .switch_event = map_switch_event, + }; + struct trace_sched_handler replay_ops = { + .wakeup_event = replay_wakeup_event, + .switch_event = replay_switch_event, + .fork_event = replay_fork_event, + }; + + argc = parse_options(argc, argv, sched_options, sched_usage, + PARSE_OPT_STOP_AT_NON_OPTION); + if (!argc) + usage_with_options(sched_usage, sched_options); + + /* + * Aliased to 'perf script' for now: + */ + if (!strcmp(argv[0], "script")) + return cmd_script(argc, argv, prefix); + + symbol__init(); + if (!strncmp(argv[0], "rec", 3)) { + return __cmd_record(argc, argv); + } else if (!strncmp(argv[0], "lat", 3)) { + sched.tp_handler = &lat_ops; + if (argc > 1) { + argc = parse_options(argc, argv, latency_options, latency_usage, 0); + if (argc) + usage_with_options(latency_usage, latency_options); + } + setup_sorting(&sched, latency_options, latency_usage); + return perf_sched__lat(&sched); + } else if (!strcmp(argv[0], "map")) { + sched.tp_handler = &map_ops; + setup_sorting(&sched, latency_options, latency_usage); + return perf_sched__map(&sched); + } else if (!strncmp(argv[0], "rep", 3)) { + sched.tp_handler = &replay_ops; + if (argc) { + argc = parse_options(argc, argv, replay_options, replay_usage, 0); + if (argc) + usage_with_options(replay_usage, replay_options); + } + return perf_sched__replay(&sched); + } else { + usage_with_options(sched_usage, sched_options); + } + + return 0; +} diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c new file mode 100644 index 00000000..92d4658f --- /dev/null +++ b/tools/perf/builtin-script.c @@ -0,0 +1,1542 @@ +#include "builtin.h" + +#include "perf.h" +#include "util/cache.h" +#include "util/debug.h" +#include "util/exec_cmd.h" +#include "util/header.h" +#include "util/parse-options.h" +#include "util/session.h" +#include "util/tool.h" +#include "util/symbol.h" +#include "util/thread.h" +#include "util/trace-event.h" +#include "util/util.h" +#include "util/evlist.h" +#include "util/evsel.h" +#include "util/sort.h" +#include <linux/bitmap.h> + +static char const *script_name; +static char const *generate_script_lang; +static bool debug_mode; +static u64 last_timestamp; +static u64 nr_unordered; +extern const struct option record_options[]; +static bool no_callchain; +static bool system_wide; +static const char *cpu_list; +static DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS); + +enum perf_output_field { + PERF_OUTPUT_COMM = 1U << 0, + PERF_OUTPUT_TID = 1U << 1, + PERF_OUTPUT_PID = 1U << 2, + PERF_OUTPUT_TIME = 1U << 3, + PERF_OUTPUT_CPU = 1U << 4, + PERF_OUTPUT_EVNAME = 1U << 5, + PERF_OUTPUT_TRACE = 1U << 6, + PERF_OUTPUT_IP = 1U << 7, + PERF_OUTPUT_SYM = 1U << 8, + PERF_OUTPUT_DSO = 1U << 9, + PERF_OUTPUT_ADDR = 1U << 10, + PERF_OUTPUT_SYMOFFSET = 1U << 11, +}; + +struct output_option { + const char *str; + enum perf_output_field field; +} all_output_options[] = { + {.str = "comm", .field = PERF_OUTPUT_COMM}, + {.str = "tid", .field = PERF_OUTPUT_TID}, + {.str = "pid", .field = PERF_OUTPUT_PID}, + {.str = "time", .field = PERF_OUTPUT_TIME}, + {.str = "cpu", .field = PERF_OUTPUT_CPU}, + {.str = "event", .field = PERF_OUTPUT_EVNAME}, + {.str = "trace", .field = PERF_OUTPUT_TRACE}, + {.str = "ip", .field = PERF_OUTPUT_IP}, + {.str = "sym", .field = PERF_OUTPUT_SYM}, + {.str = "dso", .field = PERF_OUTPUT_DSO}, + {.str = "addr", .field = PERF_OUTPUT_ADDR}, + {.str = "symoff", .field = PERF_OUTPUT_SYMOFFSET}, +}; + +/* default set to maintain compatibility with current format */ +static struct { + bool user_set; + bool wildcard_set; + u64 fields; + u64 invalid_fields; +} output[PERF_TYPE_MAX] = { + + [PERF_TYPE_HARDWARE] = { + .user_set = false, + + .fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID | + PERF_OUTPUT_CPU | PERF_OUTPUT_TIME | + PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP | + PERF_OUTPUT_SYM | PERF_OUTPUT_DSO, + + .invalid_fields = PERF_OUTPUT_TRACE, + }, + + [PERF_TYPE_SOFTWARE] = { + .user_set = false, + + .fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID | + PERF_OUTPUT_CPU | PERF_OUTPUT_TIME | + PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP | + PERF_OUTPUT_SYM | PERF_OUTPUT_DSO, + + .invalid_fields = PERF_OUTPUT_TRACE, + }, + + [PERF_TYPE_TRACEPOINT] = { + .user_set = false, + + .fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID | + PERF_OUTPUT_CPU | PERF_OUTPUT_TIME | + PERF_OUTPUT_EVNAME | PERF_OUTPUT_TRACE, + }, + + [PERF_TYPE_RAW] = { + .user_set = false, + + .fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID | + PERF_OUTPUT_CPU | PERF_OUTPUT_TIME | + PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP | + PERF_OUTPUT_SYM | PERF_OUTPUT_DSO, + + .invalid_fields = PERF_OUTPUT_TRACE, + }, +}; + +static bool output_set_by_user(void) +{ + int j; + for (j = 0; j < PERF_TYPE_MAX; ++j) { + if (output[j].user_set) + return true; + } + return false; +} + +static const char *output_field2str(enum perf_output_field field) +{ + int i, imax = ARRAY_SIZE(all_output_options); + const char *str = ""; + + for (i = 0; i < imax; ++i) { + if (all_output_options[i].field == field) { + str = all_output_options[i].str; + break; + } + } + return str; +} + +#define PRINT_FIELD(x) (output[attr->type].fields & PERF_OUTPUT_##x) + +static int perf_evsel__check_stype(struct perf_evsel *evsel, + u64 sample_type, const char *sample_msg, + enum perf_output_field field) +{ + struct perf_event_attr *attr = &evsel->attr; + int type = attr->type; + const char *evname; + + if (attr->sample_type & sample_type) + return 0; + + if (output[type].user_set) { + evname = perf_evsel__name(evsel); + pr_err("Samples for '%s' event do not have %s attribute set. " + "Cannot print '%s' field.\n", + evname, sample_msg, output_field2str(field)); + return -1; + } + + /* user did not ask for it explicitly so remove from the default list */ + output[type].fields &= ~field; + evname = perf_evsel__name(evsel); + pr_debug("Samples for '%s' event do not have %s attribute set. " + "Skipping '%s' field.\n", + evname, sample_msg, output_field2str(field)); + + return 0; +} + +static int perf_evsel__check_attr(struct perf_evsel *evsel, + struct perf_session *session) +{ + struct perf_event_attr *attr = &evsel->attr; + + if (PRINT_FIELD(TRACE) && + !perf_session__has_traces(session, "record -R")) + return -EINVAL; + + if (PRINT_FIELD(IP)) { + if (perf_evsel__check_stype(evsel, PERF_SAMPLE_IP, "IP", + PERF_OUTPUT_IP)) + return -EINVAL; + + if (!no_callchain && + !(attr->sample_type & PERF_SAMPLE_CALLCHAIN)) + symbol_conf.use_callchain = false; + } + + if (PRINT_FIELD(ADDR) && + perf_evsel__check_stype(evsel, PERF_SAMPLE_ADDR, "ADDR", + PERF_OUTPUT_ADDR)) + return -EINVAL; + + if (PRINT_FIELD(SYM) && !PRINT_FIELD(IP) && !PRINT_FIELD(ADDR)) { + pr_err("Display of symbols requested but neither sample IP nor " + "sample address\nis selected. Hence, no addresses to convert " + "to symbols.\n"); + return -EINVAL; + } + if (PRINT_FIELD(SYMOFFSET) && !PRINT_FIELD(SYM)) { + pr_err("Display of offsets requested but symbol is not" + "selected.\n"); + return -EINVAL; + } + if (PRINT_FIELD(DSO) && !PRINT_FIELD(IP) && !PRINT_FIELD(ADDR)) { + pr_err("Display of DSO requested but neither sample IP nor " + "sample address\nis selected. Hence, no addresses to convert " + "to DSO.\n"); + return -EINVAL; + } + + if ((PRINT_FIELD(PID) || PRINT_FIELD(TID)) && + perf_evsel__check_stype(evsel, PERF_SAMPLE_TID, "TID", + PERF_OUTPUT_TID|PERF_OUTPUT_PID)) + return -EINVAL; + + if (PRINT_FIELD(TIME) && + perf_evsel__check_stype(evsel, PERF_SAMPLE_TIME, "TIME", + PERF_OUTPUT_TIME)) + return -EINVAL; + + if (PRINT_FIELD(CPU) && + perf_evsel__check_stype(evsel, PERF_SAMPLE_CPU, "CPU", + PERF_OUTPUT_CPU)) + return -EINVAL; + + return 0; +} + +/* + * verify all user requested events exist and the samples + * have the expected data + */ +static int perf_session__check_output_opt(struct perf_session *session) +{ + int j; + struct perf_evsel *evsel; + + for (j = 0; j < PERF_TYPE_MAX; ++j) { + evsel = perf_session__find_first_evtype(session, j); + + /* + * even if fields is set to 0 (ie., show nothing) event must + * exist if user explicitly includes it on the command line + */ + if (!evsel && output[j].user_set && !output[j].wildcard_set) { + pr_err("%s events do not exist. " + "Remove corresponding -f option to proceed.\n", + event_type(j)); + return -1; + } + + if (evsel && output[j].fields && + perf_evsel__check_attr(evsel, session)) + return -1; + } + + return 0; +} + +static void print_sample_start(struct perf_sample *sample, + struct thread *thread, + struct perf_evsel *evsel) +{ + struct perf_event_attr *attr = &evsel->attr; + const char *evname = NULL; + unsigned long secs; + unsigned long usecs; + unsigned long long nsecs; + + if (PRINT_FIELD(COMM)) { + if (latency_format) + printf("%8.8s ", thread->comm); + else if (PRINT_FIELD(IP) && symbol_conf.use_callchain) + printf("%s ", thread->comm); + else + printf("%16s ", thread->comm); + } + + if (PRINT_FIELD(PID) && PRINT_FIELD(TID)) + printf("%5d/%-5d ", sample->pid, sample->tid); + else if (PRINT_FIELD(PID)) + printf("%5d ", sample->pid); + else if (PRINT_FIELD(TID)) + printf("%5d ", sample->tid); + + if (PRINT_FIELD(CPU)) { + if (latency_format) + printf("%3d ", sample->cpu); + else + printf("[%03d] ", sample->cpu); + } + + if (PRINT_FIELD(TIME)) { + nsecs = sample->time; + secs = nsecs / NSECS_PER_SEC; + nsecs -= secs * NSECS_PER_SEC; + usecs = nsecs / NSECS_PER_USEC; + printf("%5lu.%06lu: ", secs, usecs); + } + + if (PRINT_FIELD(EVNAME)) { + evname = perf_evsel__name(evsel); + printf("%s: ", evname ? evname : "[unknown]"); + } +} + +static bool is_bts_event(struct perf_event_attr *attr) +{ + return ((attr->type == PERF_TYPE_HARDWARE) && + (attr->config & PERF_COUNT_HW_BRANCH_INSTRUCTIONS) && + (attr->sample_period == 1)); +} + +static bool sample_addr_correlates_sym(struct perf_event_attr *attr) +{ + if ((attr->type == PERF_TYPE_SOFTWARE) && + ((attr->config == PERF_COUNT_SW_PAGE_FAULTS) || + (attr->config == PERF_COUNT_SW_PAGE_FAULTS_MIN) || + (attr->config == PERF_COUNT_SW_PAGE_FAULTS_MAJ))) + return true; + + if (is_bts_event(attr)) + return true; + + return false; +} + +static void print_sample_addr(union perf_event *event, + struct perf_sample *sample, + struct machine *machine, + struct thread *thread, + struct perf_event_attr *attr) +{ + struct addr_location al; + u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; + + printf("%16" PRIx64, sample->addr); + + if (!sample_addr_correlates_sym(attr)) + return; + + thread__find_addr_map(thread, machine, cpumode, MAP__FUNCTION, + sample->addr, &al); + if (!al.map) + thread__find_addr_map(thread, machine, cpumode, MAP__VARIABLE, + sample->addr, &al); + + al.cpu = sample->cpu; + al.sym = NULL; + + if (al.map) + al.sym = map__find_symbol(al.map, al.addr, NULL); + + if (PRINT_FIELD(SYM)) { + printf(" "); + if (PRINT_FIELD(SYMOFFSET)) + symbol__fprintf_symname_offs(al.sym, &al, stdout); + else + symbol__fprintf_symname(al.sym, stdout); + } + + if (PRINT_FIELD(DSO)) { + printf(" ("); + map__fprintf_dsoname(al.map, stdout); + printf(")"); + } +} + +static void print_sample_bts(union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine, + struct thread *thread) +{ + struct perf_event_attr *attr = &evsel->attr; + + /* print branch_from information */ + if (PRINT_FIELD(IP)) { + if (!symbol_conf.use_callchain) + printf(" "); + else + printf("\n"); + perf_evsel__print_ip(evsel, event, sample, machine, + PRINT_FIELD(SYM), PRINT_FIELD(DSO), + PRINT_FIELD(SYMOFFSET)); + } + + printf(" => "); + + /* print branch_to information */ + if (PRINT_FIELD(ADDR)) + print_sample_addr(event, sample, machine, thread, attr); + + printf("\n"); +} + +static void process_event(union perf_event *event, struct perf_sample *sample, + struct perf_evsel *evsel, struct machine *machine, + struct addr_location *al) +{ + struct perf_event_attr *attr = &evsel->attr; + struct thread *thread = al->thread; + + if (output[attr->type].fields == 0) + return; + + print_sample_start(sample, thread, evsel); + + if (is_bts_event(attr)) { + print_sample_bts(event, sample, evsel, machine, thread); + return; + } + + if (PRINT_FIELD(TRACE)) + event_format__print(evsel->tp_format, sample->cpu, + sample->raw_data, sample->raw_size); + if (PRINT_FIELD(ADDR)) + print_sample_addr(event, sample, machine, thread, attr); + + if (PRINT_FIELD(IP)) { + if (!symbol_conf.use_callchain) + printf(" "); + else + printf("\n"); + perf_evsel__print_ip(evsel, event, sample, machine, + PRINT_FIELD(SYM), PRINT_FIELD(DSO), + PRINT_FIELD(SYMOFFSET)); + } + + printf("\n"); +} + +static int default_start_script(const char *script __maybe_unused, + int argc __maybe_unused, + const char **argv __maybe_unused) +{ + return 0; +} + +static int default_stop_script(void) +{ + return 0; +} + +static int default_generate_script(struct pevent *pevent __maybe_unused, + const char *outfile __maybe_unused) +{ + return 0; +} + +static struct scripting_ops default_scripting_ops = { + .start_script = default_start_script, + .stop_script = default_stop_script, + .process_event = process_event, + .generate_script = default_generate_script, +}; + +static struct scripting_ops *scripting_ops; + +static void setup_scripting(void) +{ + setup_perl_scripting(); + setup_python_scripting(); + + scripting_ops = &default_scripting_ops; +} + +static int cleanup_scripting(void) +{ + pr_debug("\nperf script stopped\n"); + + return scripting_ops->stop_script(); +} + +static int process_sample_event(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine) +{ + struct addr_location al; + struct thread *thread = machine__findnew_thread(machine, event->ip.tid); + + if (thread == NULL) { + pr_debug("problem processing %d event, skipping it.\n", + event->header.type); + return -1; + } + + if (debug_mode) { + if (sample->time < last_timestamp) { + pr_err("Samples misordered, previous: %" PRIu64 + " this: %" PRIu64 "\n", last_timestamp, + sample->time); + nr_unordered++; + } + last_timestamp = sample->time; + return 0; + } + + if (perf_event__preprocess_sample(event, machine, &al, sample, 0) < 0) { + pr_err("problem processing %d event, skipping it.\n", + event->header.type); + return -1; + } + + if (al.filtered) + return 0; + + if (cpu_list && !test_bit(sample->cpu, cpu_bitmap)) + return 0; + + scripting_ops->process_event(event, sample, evsel, machine, &al); + + evsel->hists.stats.total_period += sample->period; + return 0; +} + +static struct perf_tool perf_script = { + .sample = process_sample_event, + .mmap = perf_event__process_mmap, + .comm = perf_event__process_comm, + .exit = perf_event__process_exit, + .fork = perf_event__process_fork, + .attr = perf_event__process_attr, + .event_type = perf_event__process_event_type, + .tracing_data = perf_event__process_tracing_data, + .build_id = perf_event__process_build_id, + .ordered_samples = true, + .ordering_requires_timestamps = true, +}; + +extern volatile int session_done; + +static void sig_handler(int sig __maybe_unused) +{ + session_done = 1; +} + +static int __cmd_script(struct perf_session *session) +{ + int ret; + + signal(SIGINT, sig_handler); + + ret = perf_session__process_events(session, &perf_script); + + if (debug_mode) + pr_err("Misordered timestamps: %" PRIu64 "\n", nr_unordered); + + return ret; +} + +struct script_spec { + struct list_head node; + struct scripting_ops *ops; + char spec[0]; +}; + +static LIST_HEAD(script_specs); + +static struct script_spec *script_spec__new(const char *spec, + struct scripting_ops *ops) +{ + struct script_spec *s = malloc(sizeof(*s) + strlen(spec) + 1); + + if (s != NULL) { + strcpy(s->spec, spec); + s->ops = ops; + } + + return s; +} + +static void script_spec__add(struct script_spec *s) +{ + list_add_tail(&s->node, &script_specs); +} + +static struct script_spec *script_spec__find(const char *spec) +{ + struct script_spec *s; + + list_for_each_entry(s, &script_specs, node) + if (strcasecmp(s->spec, spec) == 0) + return s; + return NULL; +} + +static struct script_spec *script_spec__findnew(const char *spec, + struct scripting_ops *ops) +{ + struct script_spec *s = script_spec__find(spec); + + if (s) + return s; + + s = script_spec__new(spec, ops); + if (!s) + return NULL; + + script_spec__add(s); + + return s; +} + +int script_spec_register(const char *spec, struct scripting_ops *ops) +{ + struct script_spec *s; + + s = script_spec__find(spec); + if (s) + return -1; + + s = script_spec__findnew(spec, ops); + if (!s) + return -1; + + return 0; +} + +static struct scripting_ops *script_spec__lookup(const char *spec) +{ + struct script_spec *s = script_spec__find(spec); + if (!s) + return NULL; + + return s->ops; +} + +static void list_available_languages(void) +{ + struct script_spec *s; + + fprintf(stderr, "\n"); + fprintf(stderr, "Scripting language extensions (used in " + "perf script -s [spec:]script.[spec]):\n\n"); + + list_for_each_entry(s, &script_specs, node) + fprintf(stderr, " %-42s [%s]\n", s->spec, s->ops->name); + + fprintf(stderr, "\n"); +} + +static int parse_scriptname(const struct option *opt __maybe_unused, + const char *str, int unset __maybe_unused) +{ + char spec[PATH_MAX]; + const char *script, *ext; + int len; + + if (strcmp(str, "lang") == 0) { + list_available_languages(); + exit(0); + } + + script = strchr(str, ':'); + if (script) { + len = script - str; + if (len >= PATH_MAX) { + fprintf(stderr, "invalid language specifier"); + return -1; + } + strncpy(spec, str, len); + spec[len] = '\0'; + scripting_ops = script_spec__lookup(spec); + if (!scripting_ops) { + fprintf(stderr, "invalid language specifier"); + return -1; + } + script++; + } else { + script = str; + ext = strrchr(script, '.'); + if (!ext) { + fprintf(stderr, "invalid script extension"); + return -1; + } + scripting_ops = script_spec__lookup(++ext); + if (!scripting_ops) { + fprintf(stderr, "invalid script extension"); + return -1; + } + } + + script_name = strdup(script); + + return 0; +} + +static int parse_output_fields(const struct option *opt __maybe_unused, + const char *arg, int unset __maybe_unused) +{ + char *tok; + int i, imax = ARRAY_SIZE(all_output_options); + int j; + int rc = 0; + char *str = strdup(arg); + int type = -1; + + if (!str) + return -ENOMEM; + + /* first word can state for which event type the user is specifying + * the fields. If no type exists, the specified fields apply to all + * event types found in the file minus the invalid fields for a type. + */ + tok = strchr(str, ':'); + if (tok) { + *tok = '\0'; + tok++; + if (!strcmp(str, "hw")) + type = PERF_TYPE_HARDWARE; + else if (!strcmp(str, "sw")) + type = PERF_TYPE_SOFTWARE; + else if (!strcmp(str, "trace")) + type = PERF_TYPE_TRACEPOINT; + else if (!strcmp(str, "raw")) + type = PERF_TYPE_RAW; + else { + fprintf(stderr, "Invalid event type in field string.\n"); + rc = -EINVAL; + goto out; + } + + if (output[type].user_set) + pr_warning("Overriding previous field request for %s events.\n", + event_type(type)); + + output[type].fields = 0; + output[type].user_set = true; + output[type].wildcard_set = false; + + } else { + tok = str; + if (strlen(str) == 0) { + fprintf(stderr, + "Cannot set fields to 'none' for all event types.\n"); + rc = -EINVAL; + goto out; + } + + if (output_set_by_user()) + pr_warning("Overriding previous field request for all events.\n"); + + for (j = 0; j < PERF_TYPE_MAX; ++j) { + output[j].fields = 0; + output[j].user_set = true; + output[j].wildcard_set = true; + } + } + + tok = strtok(tok, ","); + while (tok) { + for (i = 0; i < imax; ++i) { + if (strcmp(tok, all_output_options[i].str) == 0) + break; + } + if (i == imax) { + fprintf(stderr, "Invalid field requested.\n"); + rc = -EINVAL; + goto out; + } + + if (type == -1) { + /* add user option to all events types for + * which it is valid + */ + for (j = 0; j < PERF_TYPE_MAX; ++j) { + if (output[j].invalid_fields & all_output_options[i].field) { + pr_warning("\'%s\' not valid for %s events. Ignoring.\n", + all_output_options[i].str, event_type(j)); + } else + output[j].fields |= all_output_options[i].field; + } + } else { + if (output[type].invalid_fields & all_output_options[i].field) { + fprintf(stderr, "\'%s\' not valid for %s events.\n", + all_output_options[i].str, event_type(type)); + + rc = -EINVAL; + goto out; + } + output[type].fields |= all_output_options[i].field; + } + + tok = strtok(NULL, ","); + } + + if (type >= 0) { + if (output[type].fields == 0) { + pr_debug("No fields requested for %s type. " + "Events will not be displayed.\n", event_type(type)); + } + } + +out: + free(str); + return rc; +} + +/* Helper function for filesystems that return a dent->d_type DT_UNKNOWN */ +static int is_directory(const char *base_path, const struct dirent *dent) +{ + char path[PATH_MAX]; + struct stat st; + + sprintf(path, "%s/%s", base_path, dent->d_name); + if (stat(path, &st)) + return 0; + + return S_ISDIR(st.st_mode); +} + +#define for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next)\ + while (!readdir_r(scripts_dir, &lang_dirent, &lang_next) && \ + lang_next) \ + if ((lang_dirent.d_type == DT_DIR || \ + (lang_dirent.d_type == DT_UNKNOWN && \ + is_directory(scripts_path, &lang_dirent))) && \ + (strcmp(lang_dirent.d_name, ".")) && \ + (strcmp(lang_dirent.d_name, ".."))) + +#define for_each_script(lang_path, lang_dir, script_dirent, script_next)\ + while (!readdir_r(lang_dir, &script_dirent, &script_next) && \ + script_next) \ + if (script_dirent.d_type != DT_DIR && \ + (script_dirent.d_type != DT_UNKNOWN || \ + !is_directory(lang_path, &script_dirent))) + + +#define RECORD_SUFFIX "-record" +#define REPORT_SUFFIX "-report" + +struct script_desc { + struct list_head node; + char *name; + char *half_liner; + char *args; +}; + +static LIST_HEAD(script_descs); + +static struct script_desc *script_desc__new(const char *name) +{ + struct script_desc *s = zalloc(sizeof(*s)); + + if (s != NULL && name) + s->name = strdup(name); + + return s; +} + +static void script_desc__delete(struct script_desc *s) +{ + free(s->name); + free(s->half_liner); + free(s->args); + free(s); +} + +static void script_desc__add(struct script_desc *s) +{ + list_add_tail(&s->node, &script_descs); +} + +static struct script_desc *script_desc__find(const char *name) +{ + struct script_desc *s; + + list_for_each_entry(s, &script_descs, node) + if (strcasecmp(s->name, name) == 0) + return s; + return NULL; +} + +static struct script_desc *script_desc__findnew(const char *name) +{ + struct script_desc *s = script_desc__find(name); + + if (s) + return s; + + s = script_desc__new(name); + if (!s) + goto out_delete_desc; + + script_desc__add(s); + + return s; + +out_delete_desc: + script_desc__delete(s); + + return NULL; +} + +static const char *ends_with(const char *str, const char *suffix) +{ + size_t suffix_len = strlen(suffix); + const char *p = str; + + if (strlen(str) > suffix_len) { + p = str + strlen(str) - suffix_len; + if (!strncmp(p, suffix, suffix_len)) + return p; + } + + return NULL; +} + +static int read_script_info(struct script_desc *desc, const char *filename) +{ + char line[BUFSIZ], *p; + FILE *fp; + + fp = fopen(filename, "r"); + if (!fp) + return -1; + + while (fgets(line, sizeof(line), fp)) { + p = ltrim(line); + if (strlen(p) == 0) + continue; + if (*p != '#') + continue; + p++; + if (strlen(p) && *p == '!') + continue; + + p = ltrim(p); + if (strlen(p) && p[strlen(p) - 1] == '\n') + p[strlen(p) - 1] = '\0'; + + if (!strncmp(p, "description:", strlen("description:"))) { + p += strlen("description:"); + desc->half_liner = strdup(ltrim(p)); + continue; + } + + if (!strncmp(p, "args:", strlen("args:"))) { + p += strlen("args:"); + desc->args = strdup(ltrim(p)); + continue; + } + } + + fclose(fp); + + return 0; +} + +static char *get_script_root(struct dirent *script_dirent, const char *suffix) +{ + char *script_root, *str; + + script_root = strdup(script_dirent->d_name); + if (!script_root) + return NULL; + + str = (char *)ends_with(script_root, suffix); + if (!str) { + free(script_root); + return NULL; + } + + *str = '\0'; + return script_root; +} + +static int list_available_scripts(const struct option *opt __maybe_unused, + const char *s __maybe_unused, + int unset __maybe_unused) +{ + struct dirent *script_next, *lang_next, script_dirent, lang_dirent; + char scripts_path[MAXPATHLEN]; + DIR *scripts_dir, *lang_dir; + char script_path[MAXPATHLEN]; + char lang_path[MAXPATHLEN]; + struct script_desc *desc; + char first_half[BUFSIZ]; + char *script_root; + + snprintf(scripts_path, MAXPATHLEN, "%s/scripts", perf_exec_path()); + + scripts_dir = opendir(scripts_path); + if (!scripts_dir) + return -1; + + for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next) { + snprintf(lang_path, MAXPATHLEN, "%s/%s/bin", scripts_path, + lang_dirent.d_name); + lang_dir = opendir(lang_path); + if (!lang_dir) + continue; + + for_each_script(lang_path, lang_dir, script_dirent, script_next) { + script_root = get_script_root(&script_dirent, REPORT_SUFFIX); + if (script_root) { + desc = script_desc__findnew(script_root); + snprintf(script_path, MAXPATHLEN, "%s/%s", + lang_path, script_dirent.d_name); + read_script_info(desc, script_path); + free(script_root); + } + } + } + + fprintf(stdout, "List of available trace scripts:\n"); + list_for_each_entry(desc, &script_descs, node) { + sprintf(first_half, "%s %s", desc->name, + desc->args ? desc->args : ""); + fprintf(stdout, " %-36s %s\n", first_half, + desc->half_liner ? desc->half_liner : ""); + } + + exit(0); +} + +/* + * Some scripts specify the required events in their "xxx-record" file, + * this function will check if the events in perf.data match those + * mentioned in the "xxx-record". + * + * Fixme: All existing "xxx-record" are all in good formats "-e event ", + * which is covered well now. And new parsing code should be added to + * cover the future complexing formats like event groups etc. + */ +static int check_ev_match(char *dir_name, char *scriptname, + struct perf_session *session) +{ + char filename[MAXPATHLEN], evname[128]; + char line[BUFSIZ], *p; + struct perf_evsel *pos; + int match, len; + FILE *fp; + + sprintf(filename, "%s/bin/%s-record", dir_name, scriptname); + + fp = fopen(filename, "r"); + if (!fp) + return -1; + + while (fgets(line, sizeof(line), fp)) { + p = ltrim(line); + if (*p == '#') + continue; + + while (strlen(p)) { + p = strstr(p, "-e"); + if (!p) + break; + + p += 2; + p = ltrim(p); + len = strcspn(p, " \t"); + if (!len) + break; + + snprintf(evname, len + 1, "%s", p); + + match = 0; + list_for_each_entry(pos, + &session->evlist->entries, node) { + if (!strcmp(perf_evsel__name(pos), evname)) { + match = 1; + break; + } + } + + if (!match) { + fclose(fp); + return -1; + } + } + } + + fclose(fp); + return 0; +} + +/* + * Return -1 if none is found, otherwise the actual scripts number. + * + * Currently the only user of this function is the script browser, which + * will list all statically runnable scripts, select one, execute it and + * show the output in a perf browser. + */ +int find_scripts(char **scripts_array, char **scripts_path_array) +{ + struct dirent *script_next, *lang_next, script_dirent, lang_dirent; + char scripts_path[MAXPATHLEN], lang_path[MAXPATHLEN]; + DIR *scripts_dir, *lang_dir; + struct perf_session *session; + char *temp; + int i = 0; + + session = perf_session__new(input_name, O_RDONLY, 0, false, NULL); + if (!session) + return -1; + + snprintf(scripts_path, MAXPATHLEN, "%s/scripts", perf_exec_path()); + + scripts_dir = opendir(scripts_path); + if (!scripts_dir) { + perf_session__delete(session); + return -1; + } + + for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next) { + snprintf(lang_path, MAXPATHLEN, "%s/%s", scripts_path, + lang_dirent.d_name); +#ifdef NO_LIBPERL + if (strstr(lang_path, "perl")) + continue; +#endif +#ifdef NO_LIBPYTHON + if (strstr(lang_path, "python")) + continue; +#endif + + lang_dir = opendir(lang_path); + if (!lang_dir) + continue; + + for_each_script(lang_path, lang_dir, script_dirent, script_next) { + /* Skip those real time scripts: xxxtop.p[yl] */ + if (strstr(script_dirent.d_name, "top.")) + continue; + sprintf(scripts_path_array[i], "%s/%s", lang_path, + script_dirent.d_name); + temp = strchr(script_dirent.d_name, '.'); + snprintf(scripts_array[i], + (temp - script_dirent.d_name) + 1, + "%s", script_dirent.d_name); + + if (check_ev_match(lang_path, + scripts_array[i], session)) + continue; + + i++; + } + closedir(lang_dir); + } + + closedir(scripts_dir); + perf_session__delete(session); + return i; +} + +static char *get_script_path(const char *script_root, const char *suffix) +{ + struct dirent *script_next, *lang_next, script_dirent, lang_dirent; + char scripts_path[MAXPATHLEN]; + char script_path[MAXPATHLEN]; + DIR *scripts_dir, *lang_dir; + char lang_path[MAXPATHLEN]; + char *__script_root; + + snprintf(scripts_path, MAXPATHLEN, "%s/scripts", perf_exec_path()); + + scripts_dir = opendir(scripts_path); + if (!scripts_dir) + return NULL; + + for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next) { + snprintf(lang_path, MAXPATHLEN, "%s/%s/bin", scripts_path, + lang_dirent.d_name); + lang_dir = opendir(lang_path); + if (!lang_dir) + continue; + + for_each_script(lang_path, lang_dir, script_dirent, script_next) { + __script_root = get_script_root(&script_dirent, suffix); + if (__script_root && !strcmp(script_root, __script_root)) { + free(__script_root); + closedir(lang_dir); + closedir(scripts_dir); + snprintf(script_path, MAXPATHLEN, "%s/%s", + lang_path, script_dirent.d_name); + return strdup(script_path); + } + free(__script_root); + } + closedir(lang_dir); + } + closedir(scripts_dir); + + return NULL; +} + +static bool is_top_script(const char *script_path) +{ + return ends_with(script_path, "top") == NULL ? false : true; +} + +static int has_required_arg(char *script_path) +{ + struct script_desc *desc; + int n_args = 0; + char *p; + + desc = script_desc__new(NULL); + + if (read_script_info(desc, script_path)) + goto out; + + if (!desc->args) + goto out; + + for (p = desc->args; *p; p++) + if (*p == '<') + n_args++; +out: + script_desc__delete(desc); + + return n_args; +} + +static int have_cmd(int argc, const char **argv) +{ + char **__argv = malloc(sizeof(const char *) * argc); + + if (!__argv) { + pr_err("malloc failed\n"); + return -1; + } + + memcpy(__argv, argv, sizeof(const char *) * argc); + argc = parse_options(argc, (const char **)__argv, record_options, + NULL, PARSE_OPT_STOP_AT_NON_OPTION); + free(__argv); + + system_wide = (argc == 0); + + return 0; +} + +int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused) +{ + bool show_full_info = false; + char *rec_script_path = NULL; + char *rep_script_path = NULL; + struct perf_session *session; + char *script_path = NULL; + const char **__argv; + int i, j, err; + const struct option options[] = { + OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, + "dump raw trace in ASCII"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show symbol address, etc)"), + OPT_BOOLEAN('L', "Latency", &latency_format, + "show latency attributes (irqs/preemption disabled, etc)"), + OPT_CALLBACK_NOOPT('l', "list", NULL, NULL, "list available scripts", + list_available_scripts), + OPT_CALLBACK('s', "script", NULL, "name", + "script file name (lang:script name, script name, or *)", + parse_scriptname), + OPT_STRING('g', "gen-script", &generate_script_lang, "lang", + "generate perf-script.xx script in specified language"), + OPT_STRING('i', "input", &input_name, "file", "input file name"), + OPT_BOOLEAN('d', "debug-mode", &debug_mode, + "do various checks like samples ordering and lost events"), + OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name, + "file", "vmlinux pathname"), + OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name, + "file", "kallsyms pathname"), + OPT_BOOLEAN('G', "hide-call-graph", &no_callchain, + "When printing symbols do not display call chain"), + OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory", + "Look for files with symbols relative to this directory"), + OPT_CALLBACK('f', "fields", NULL, "str", + "comma separated output fields prepend with 'type:'. " + "Valid types: hw,sw,trace,raw. " + "Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso," + "addr,symoff", parse_output_fields), + OPT_BOOLEAN('a', "all-cpus", &system_wide, + "system-wide collection from all CPUs"), + OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]", + "only consider these symbols"), + OPT_STRING('C', "cpu", &cpu_list, "cpu", "list of cpus to profile"), + OPT_STRING('c', "comms", &symbol_conf.comm_list_str, "comm[,comm...]", + "only display events for these comms"), + OPT_BOOLEAN('I', "show-info", &show_full_info, + "display extended information from perf.data file"), + OPT_BOOLEAN('\0', "show-kernel-path", &symbol_conf.show_kernel_path, + "Show the path of [kernel.kallsyms]"), + OPT_END() + }; + const char * const script_usage[] = { + "perf script [<options>]", + "perf script [<options>] record <script> [<record-options>] <command>", + "perf script [<options>] report <script> [script-args]", + "perf script [<options>] <script> [<record-options>] <command>", + "perf script [<options>] <top-script> [script-args]", + NULL + }; + + setup_scripting(); + + argc = parse_options(argc, argv, options, script_usage, + PARSE_OPT_STOP_AT_NON_OPTION); + + if (argc > 1 && !strncmp(argv[0], "rec", strlen("rec"))) { + rec_script_path = get_script_path(argv[1], RECORD_SUFFIX); + if (!rec_script_path) + return cmd_record(argc, argv, NULL); + } + + if (argc > 1 && !strncmp(argv[0], "rep", strlen("rep"))) { + rep_script_path = get_script_path(argv[1], REPORT_SUFFIX); + if (!rep_script_path) { + fprintf(stderr, + "Please specify a valid report script" + "(see 'perf script -l' for listing)\n"); + return -1; + } + } + + /* make sure PERF_EXEC_PATH is set for scripts */ + perf_set_argv_exec_path(perf_exec_path()); + + if (argc && !script_name && !rec_script_path && !rep_script_path) { + int live_pipe[2]; + int rep_args; + pid_t pid; + + rec_script_path = get_script_path(argv[0], RECORD_SUFFIX); + rep_script_path = get_script_path(argv[0], REPORT_SUFFIX); + + if (!rec_script_path && !rep_script_path) { + fprintf(stderr, " Couldn't find script %s\n\n See perf" + " script -l for available scripts.\n", argv[0]); + usage_with_options(script_usage, options); + } + + if (is_top_script(argv[0])) { + rep_args = argc - 1; + } else { + int rec_args; + + rep_args = has_required_arg(rep_script_path); + rec_args = (argc - 1) - rep_args; + if (rec_args < 0) { + fprintf(stderr, " %s script requires options." + "\n\n See perf script -l for available " + "scripts and options.\n", argv[0]); + usage_with_options(script_usage, options); + } + } + + if (pipe(live_pipe) < 0) { + perror("failed to create pipe"); + return -1; + } + + pid = fork(); + if (pid < 0) { + perror("failed to fork"); + return -1; + } + + if (!pid) { + j = 0; + + dup2(live_pipe[1], 1); + close(live_pipe[0]); + + if (is_top_script(argv[0])) { + system_wide = true; + } else if (!system_wide) { + if (have_cmd(argc - rep_args, &argv[rep_args]) != 0) { + err = -1; + goto out; + } + } + + __argv = malloc((argc + 6) * sizeof(const char *)); + if (!__argv) { + pr_err("malloc failed\n"); + err = -ENOMEM; + goto out; + } + + __argv[j++] = "/bin/sh"; + __argv[j++] = rec_script_path; + if (system_wide) + __argv[j++] = "-a"; + __argv[j++] = "-q"; + __argv[j++] = "-o"; + __argv[j++] = "-"; + for (i = rep_args + 1; i < argc; i++) + __argv[j++] = argv[i]; + __argv[j++] = NULL; + + execvp("/bin/sh", (char **)__argv); + free(__argv); + exit(-1); + } + + dup2(live_pipe[0], 0); + close(live_pipe[1]); + + __argv = malloc((argc + 4) * sizeof(const char *)); + if (!__argv) { + pr_err("malloc failed\n"); + err = -ENOMEM; + goto out; + } + + j = 0; + __argv[j++] = "/bin/sh"; + __argv[j++] = rep_script_path; + for (i = 1; i < rep_args + 1; i++) + __argv[j++] = argv[i]; + __argv[j++] = "-i"; + __argv[j++] = "-"; + __argv[j++] = NULL; + + execvp("/bin/sh", (char **)__argv); + free(__argv); + exit(-1); + } + + if (rec_script_path) + script_path = rec_script_path; + if (rep_script_path) + script_path = rep_script_path; + + if (script_path) { + j = 0; + + if (!rec_script_path) + system_wide = false; + else if (!system_wide) { + if (have_cmd(argc - 1, &argv[1]) != 0) { + err = -1; + goto out; + } + } + + __argv = malloc((argc + 2) * sizeof(const char *)); + if (!__argv) { + pr_err("malloc failed\n"); + err = -ENOMEM; + goto out; + } + + __argv[j++] = "/bin/sh"; + __argv[j++] = script_path; + if (system_wide) + __argv[j++] = "-a"; + for (i = 2; i < argc; i++) + __argv[j++] = argv[i]; + __argv[j++] = NULL; + + execvp("/bin/sh", (char **)__argv); + free(__argv); + exit(-1); + } + + if (symbol__init() < 0) + return -1; + if (!script_name) + setup_pager(); + + session = perf_session__new(input_name, O_RDONLY, 0, false, + &perf_script); + if (session == NULL) + return -ENOMEM; + + if (cpu_list) { + if (perf_session__cpu_bitmap(session, cpu_list, cpu_bitmap)) + return -1; + } + + if (!script_name && !generate_script_lang) + perf_session__fprintf_info(session, stdout, show_full_info); + + if (!no_callchain) + symbol_conf.use_callchain = true; + else + symbol_conf.use_callchain = false; + + if (generate_script_lang) { + struct stat perf_stat; + int input; + + if (output_set_by_user()) { + fprintf(stderr, + "custom fields not supported for generated scripts"); + return -1; + } + + input = open(session->filename, O_RDONLY); /* input_name */ + if (input < 0) { + perror("failed to open file"); + return -1; + } + + err = fstat(input, &perf_stat); + if (err < 0) { + perror("failed to stat file"); + return -1; + } + + if (!perf_stat.st_size) { + fprintf(stderr, "zero-sized file, nothing to do!\n"); + return 0; + } + + scripting_ops = script_spec__lookup(generate_script_lang); + if (!scripting_ops) { + fprintf(stderr, "invalid language specifier"); + return -1; + } + + err = scripting_ops->generate_script(session->pevent, + "perf-script"); + goto out; + } + + if (script_name) { + err = scripting_ops->start_script(script_name, argc, argv); + if (err) + goto out; + pr_debug("perf script started with script %s\n\n", script_name); + } + + + err = perf_session__check_output_opt(session); + if (err < 0) + goto out; + + err = __cmd_script(session); + + perf_session__delete(session); + cleanup_scripting(); +out: + return err; +} diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c new file mode 100644 index 00000000..99848761 --- /dev/null +++ b/tools/perf/builtin-stat.c @@ -0,0 +1,1486 @@ +/* + * 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: 0.118 + + Performance counter stats for './hackbench 10': + + 1708.761321 task-clock # 11.037 CPUs utilized + 41,190 context-switches # 0.024 M/sec + 6,735 CPU-migrations # 0.004 M/sec + 17,318 page-faults # 0.010 M/sec + 5,205,202,243 cycles # 3.046 GHz + 3,856,436,920 stalled-cycles-frontend # 74.09% frontend cycles idle + 1,600,790,871 stalled-cycles-backend # 30.75% backend cycles idle + 2,603,501,247 instructions # 0.50 insns per cycle + # 1.48 stalled cycles per insn + 484,357,498 branches # 283.455 M/sec + 6,388,934 branch-misses # 1.32% of all branches + + 0.154822978 seconds time elapsed + + * + * Copyright (C) 2008-2011, 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> + * Jaswinder Singh Rajput <jaswinder@kernel.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 "util/event.h" +#include "util/evlist.h" +#include "util/evsel.h" +#include "util/debug.h" +#include "util/color.h" +#include "util/stat.h" +#include "util/header.h" +#include "util/cpumap.h" +#include "util/thread.h" +#include "util/thread_map.h" + +#include <stdlib.h> +#include <sys/prctl.h> +#include <locale.h> + +#define DEFAULT_SEPARATOR " " +#define CNTR_NOT_SUPPORTED "<not supported>" +#define CNTR_NOT_COUNTED "<not counted>" + +static void print_stat(int argc, const char **argv); +static void print_counter_aggr(struct perf_evsel *counter, char *prefix); +static void print_counter(struct perf_evsel *counter, char *prefix); +static void print_aggr_socket(char *prefix); + +static struct perf_evlist *evsel_list; + +static struct perf_target target = { + .uid = UINT_MAX, +}; + +static int run_count = 1; +static bool no_inherit = false; +static bool scale = true; +static bool no_aggr = false; +static bool aggr_socket = false; +static pid_t child_pid = -1; +static bool null_run = false; +static int detailed_run = 0; +static bool big_num = true; +static int big_num_opt = -1; +static const char *csv_sep = NULL; +static bool csv_output = false; +static bool group = false; +static FILE *output = NULL; +static const char *pre_cmd = NULL; +static const char *post_cmd = NULL; +static bool sync_run = false; +static unsigned int interval = 0; +static struct timespec ref_time; +static struct cpu_map *sock_map; + +static volatile int done = 0; + +struct perf_stat { + struct stats res_stats[3]; +}; + +static inline void diff_timespec(struct timespec *r, struct timespec *a, + struct timespec *b) +{ + r->tv_sec = a->tv_sec - b->tv_sec; + if (a->tv_nsec < b->tv_nsec) { + r->tv_nsec = a->tv_nsec + 1000000000L - b->tv_nsec; + r->tv_sec--; + } else { + r->tv_nsec = a->tv_nsec - b->tv_nsec ; + } +} + +static inline struct cpu_map *perf_evsel__cpus(struct perf_evsel *evsel) +{ + return (evsel->cpus && !target.cpu_list) ? evsel->cpus : evsel_list->cpus; +} + +static inline int perf_evsel__nr_cpus(struct perf_evsel *evsel) +{ + return perf_evsel__cpus(evsel)->nr; +} + +static int perf_evsel__alloc_stat_priv(struct perf_evsel *evsel) +{ + evsel->priv = zalloc(sizeof(struct perf_stat)); + return evsel->priv == NULL ? -ENOMEM : 0; +} + +static void perf_evsel__free_stat_priv(struct perf_evsel *evsel) +{ + free(evsel->priv); + evsel->priv = NULL; +} + +static int perf_evsel__alloc_prev_raw_counts(struct perf_evsel *evsel) +{ + void *addr; + size_t sz; + + sz = sizeof(*evsel->counts) + + (perf_evsel__nr_cpus(evsel) * sizeof(struct perf_counts_values)); + + addr = zalloc(sz); + if (!addr) + return -ENOMEM; + + evsel->prev_raw_counts = addr; + + return 0; +} + +static void perf_evsel__free_prev_raw_counts(struct perf_evsel *evsel) +{ + free(evsel->prev_raw_counts); + evsel->prev_raw_counts = NULL; +} + +static struct stats runtime_nsecs_stats[MAX_NR_CPUS]; +static struct stats runtime_cycles_stats[MAX_NR_CPUS]; +static struct stats runtime_stalled_cycles_front_stats[MAX_NR_CPUS]; +static struct stats runtime_stalled_cycles_back_stats[MAX_NR_CPUS]; +static struct stats runtime_branches_stats[MAX_NR_CPUS]; +static struct stats runtime_cacherefs_stats[MAX_NR_CPUS]; +static struct stats runtime_l1_dcache_stats[MAX_NR_CPUS]; +static struct stats runtime_l1_icache_stats[MAX_NR_CPUS]; +static struct stats runtime_ll_cache_stats[MAX_NR_CPUS]; +static struct stats runtime_itlb_cache_stats[MAX_NR_CPUS]; +static struct stats runtime_dtlb_cache_stats[MAX_NR_CPUS]; +static struct stats walltime_nsecs_stats; + +static int create_perf_stat_counter(struct perf_evsel *evsel) +{ + struct perf_event_attr *attr = &evsel->attr; + + if (scale) + attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED | + PERF_FORMAT_TOTAL_TIME_RUNNING; + + attr->inherit = !no_inherit; + + if (perf_target__has_cpu(&target)) + return perf_evsel__open_per_cpu(evsel, perf_evsel__cpus(evsel)); + + if (!perf_target__has_task(&target) && + perf_evsel__is_group_leader(evsel)) { + attr->disabled = 1; + attr->enable_on_exec = 1; + } + + return perf_evsel__open_per_thread(evsel, evsel_list->threads); +} + +/* + * Does the counter have nsecs as a unit? + */ +static inline int nsec_counter(struct perf_evsel *evsel) +{ + if (perf_evsel__match(evsel, SOFTWARE, SW_CPU_CLOCK) || + perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK)) + return 1; + + return 0; +} + +/* + * Update various tracking values we maintain to print + * more semantic information such as miss/hit ratios, + * instruction rates, etc: + */ +static void update_shadow_stats(struct perf_evsel *counter, u64 *count) +{ + if (perf_evsel__match(counter, SOFTWARE, SW_TASK_CLOCK)) + update_stats(&runtime_nsecs_stats[0], count[0]); + else if (perf_evsel__match(counter, HARDWARE, HW_CPU_CYCLES)) + update_stats(&runtime_cycles_stats[0], count[0]); + else if (perf_evsel__match(counter, HARDWARE, HW_STALLED_CYCLES_FRONTEND)) + update_stats(&runtime_stalled_cycles_front_stats[0], count[0]); + else if (perf_evsel__match(counter, HARDWARE, HW_STALLED_CYCLES_BACKEND)) + update_stats(&runtime_stalled_cycles_back_stats[0], count[0]); + else if (perf_evsel__match(counter, HARDWARE, HW_BRANCH_INSTRUCTIONS)) + update_stats(&runtime_branches_stats[0], count[0]); + else if (perf_evsel__match(counter, HARDWARE, HW_CACHE_REFERENCES)) + update_stats(&runtime_cacherefs_stats[0], count[0]); + else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_L1D)) + update_stats(&runtime_l1_dcache_stats[0], count[0]); + else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_L1I)) + update_stats(&runtime_l1_icache_stats[0], count[0]); + else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_LL)) + update_stats(&runtime_ll_cache_stats[0], count[0]); + else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_DTLB)) + update_stats(&runtime_dtlb_cache_stats[0], count[0]); + else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_ITLB)) + update_stats(&runtime_itlb_cache_stats[0], count[0]); +} + +/* + * Read out the results of a single counter: + * aggregate counts across CPUs in system-wide mode + */ +static int read_counter_aggr(struct perf_evsel *counter) +{ + struct perf_stat *ps = counter->priv; + u64 *count = counter->counts->aggr.values; + int i; + + if (__perf_evsel__read(counter, perf_evsel__nr_cpus(counter), + evsel_list->threads->nr, scale) < 0) + return -1; + + for (i = 0; i < 3; i++) + update_stats(&ps->res_stats[i], count[i]); + + if (verbose) { + fprintf(output, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n", + perf_evsel__name(counter), count[0], count[1], count[2]); + } + + /* + * Save the full runtime - to allow normalization during printout: + */ + update_shadow_stats(counter, count); + + return 0; +} + +/* + * Read out the results of a single counter: + * do not aggregate counts across CPUs in system-wide mode + */ +static int read_counter(struct perf_evsel *counter) +{ + u64 *count; + int cpu; + + for (cpu = 0; cpu < perf_evsel__nr_cpus(counter); cpu++) { + if (__perf_evsel__read_on_cpu(counter, cpu, 0, scale) < 0) + return -1; + + count = counter->counts->cpu[cpu].values; + + update_shadow_stats(counter, count); + } + + return 0; +} + +static void print_interval(void) +{ + static int num_print_interval; + struct perf_evsel *counter; + struct perf_stat *ps; + struct timespec ts, rs; + char prefix[64]; + + if (no_aggr) { + list_for_each_entry(counter, &evsel_list->entries, node) { + ps = counter->priv; + memset(ps->res_stats, 0, sizeof(ps->res_stats)); + read_counter(counter); + } + } else { + list_for_each_entry(counter, &evsel_list->entries, node) { + ps = counter->priv; + memset(ps->res_stats, 0, sizeof(ps->res_stats)); + read_counter_aggr(counter); + } + } + clock_gettime(CLOCK_MONOTONIC, &ts); + diff_timespec(&rs, &ts, &ref_time); + sprintf(prefix, "%6lu.%09lu%s", rs.tv_sec, rs.tv_nsec, csv_sep); + + if (num_print_interval == 0 && !csv_output) { + if (aggr_socket) + fprintf(output, "# time socket cpus counts events\n"); + else if (no_aggr) + fprintf(output, "# time CPU counts events\n"); + else + fprintf(output, "# time counts events\n"); + } + + if (++num_print_interval == 25) + num_print_interval = 0; + + if (aggr_socket) + print_aggr_socket(prefix); + else if (no_aggr) { + list_for_each_entry(counter, &evsel_list->entries, node) + print_counter(counter, prefix); + } else { + list_for_each_entry(counter, &evsel_list->entries, node) + print_counter_aggr(counter, prefix); + } +} + +static int __run_perf_stat(int argc __maybe_unused, const char **argv) +{ + char msg[512]; + unsigned long long t0, t1; + struct perf_evsel *counter; + struct timespec ts; + int status = 0; + int child_ready_pipe[2], go_pipe[2]; + const bool forks = (argc > 0); + char buf; + + if (interval) { + ts.tv_sec = interval / 1000; + ts.tv_nsec = (interval % 1000) * 1000000; + } else { + ts.tv_sec = 1; + ts.tv_nsec = 0; + } + + if (aggr_socket + && cpu_map__build_socket_map(evsel_list->cpus, &sock_map)) { + perror("cannot build socket map"); + return -1; + } + + if (forks && (pipe(child_ready_pipe) < 0 || pipe(go_pipe) < 0)) { + perror("failed to create pipes"); + return -1; + } + + if (forks) { + if ((child_pid = fork()) < 0) + perror("failed to fork"); + + if (!child_pid) { + close(child_ready_pipe[0]); + close(go_pipe[1]); + fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC); + + /* + * Do a dummy execvp to get the PLT entry resolved, + * so we avoid the resolver overhead on the real + * execvp call. + */ + execvp("", (char **)argv); + + /* + * Tell the parent we're ready to go + */ + close(child_ready_pipe[1]); + + /* + * Wait until the parent tells us to go. + */ + if (read(go_pipe[0], &buf, 1) == -1) + perror("unable to read pipe"); + + execvp(argv[0], (char **)argv); + + perror(argv[0]); + exit(-1); + } + + if (perf_target__none(&target)) + evsel_list->threads->map[0] = child_pid; + + /* + * Wait for the child to be ready to exec. + */ + close(child_ready_pipe[1]); + close(go_pipe[0]); + if (read(child_ready_pipe[0], &buf, 1) == -1) + perror("unable to read pipe"); + close(child_ready_pipe[0]); + } + + if (group) + perf_evlist__set_leader(evsel_list); + + list_for_each_entry(counter, &evsel_list->entries, node) { + if (create_perf_stat_counter(counter) < 0) { + /* + * PPC returns ENXIO for HW counters until 2.6.37 + * (behavior changed with commit b0a873e). + */ + if (errno == EINVAL || errno == ENOSYS || + errno == ENOENT || errno == EOPNOTSUPP || + errno == ENXIO) { + if (verbose) + ui__warning("%s event is not supported by the kernel.\n", + perf_evsel__name(counter)); + counter->supported = false; + continue; + } + + perf_evsel__open_strerror(counter, &target, + errno, msg, sizeof(msg)); + ui__error("%s\n", msg); + + if (child_pid != -1) + kill(child_pid, SIGTERM); + + return -1; + } + counter->supported = true; + } + + if (perf_evlist__apply_filters(evsel_list)) { + error("failed to set filter with %d (%s)\n", errno, + strerror(errno)); + return -1; + } + + /* + * Enable counters and exec the command: + */ + t0 = rdclock(); + clock_gettime(CLOCK_MONOTONIC, &ref_time); + + if (forks) { + close(go_pipe[1]); + if (interval) { + while (!waitpid(child_pid, &status, WNOHANG)) { + nanosleep(&ts, NULL); + print_interval(); + } + } + wait(&status); + if (WIFSIGNALED(status)) + psignal(WTERMSIG(status), argv[0]); + } else { + while (!done) { + nanosleep(&ts, NULL); + if (interval) + print_interval(); + } + } + + t1 = rdclock(); + + update_stats(&walltime_nsecs_stats, t1 - t0); + + if (no_aggr) { + list_for_each_entry(counter, &evsel_list->entries, node) { + read_counter(counter); + perf_evsel__close_fd(counter, perf_evsel__nr_cpus(counter), 1); + } + } else { + list_for_each_entry(counter, &evsel_list->entries, node) { + read_counter_aggr(counter); + perf_evsel__close_fd(counter, perf_evsel__nr_cpus(counter), + evsel_list->threads->nr); + } + } + + return WEXITSTATUS(status); +} + +static int run_perf_stat(int argc __maybe_unused, const char **argv) +{ + int ret; + + if (pre_cmd) { + ret = system(pre_cmd); + if (ret) + return ret; + } + + if (sync_run) + sync(); + + ret = __run_perf_stat(argc, argv); + if (ret) + return ret; + + if (post_cmd) { + ret = system(post_cmd); + if (ret) + return ret; + } + + return ret; +} + +static void print_noise_pct(double total, double avg) +{ + double pct = rel_stddev_stats(total, avg); + + if (csv_output) + fprintf(output, "%s%.2f%%", csv_sep, pct); + else if (pct) + fprintf(output, " ( +-%6.2f%% )", pct); +} + +static void print_noise(struct perf_evsel *evsel, double avg) +{ + struct perf_stat *ps; + + if (run_count == 1) + return; + + ps = evsel->priv; + print_noise_pct(stddev_stats(&ps->res_stats[0]), avg); +} + +static void nsec_printout(int cpu, int nr, struct perf_evsel *evsel, double avg) +{ + double msecs = avg / 1e6; + char cpustr[16] = { '\0', }; + const char *fmt = csv_output ? "%s%.6f%s%s" : "%s%18.6f%s%-25s"; + + if (aggr_socket) + sprintf(cpustr, "S%*d%s%*d%s", + csv_output ? 0 : -5, + cpu, + csv_sep, + csv_output ? 0 : 4, + nr, + csv_sep); + else if (no_aggr) + sprintf(cpustr, "CPU%*d%s", + csv_output ? 0 : -4, + perf_evsel__cpus(evsel)->map[cpu], csv_sep); + + fprintf(output, fmt, cpustr, msecs, csv_sep, perf_evsel__name(evsel)); + + if (evsel->cgrp) + fprintf(output, "%s%s", csv_sep, evsel->cgrp->name); + + if (csv_output || interval) + return; + + if (perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK)) + fprintf(output, " # %8.3f CPUs utilized ", + avg / avg_stats(&walltime_nsecs_stats)); + else + fprintf(output, " "); +} + +/* used for get_ratio_color() */ +enum grc_type { + GRC_STALLED_CYCLES_FE, + GRC_STALLED_CYCLES_BE, + GRC_CACHE_MISSES, + GRC_MAX_NR +}; + +static const char *get_ratio_color(enum grc_type type, double ratio) +{ + static const double grc_table[GRC_MAX_NR][3] = { + [GRC_STALLED_CYCLES_FE] = { 50.0, 30.0, 10.0 }, + [GRC_STALLED_CYCLES_BE] = { 75.0, 50.0, 20.0 }, + [GRC_CACHE_MISSES] = { 20.0, 10.0, 5.0 }, + }; + const char *color = PERF_COLOR_NORMAL; + + if (ratio > grc_table[type][0]) + color = PERF_COLOR_RED; + else if (ratio > grc_table[type][1]) + color = PERF_COLOR_MAGENTA; + else if (ratio > grc_table[type][2]) + color = PERF_COLOR_YELLOW; + + return color; +} + +static void print_stalled_cycles_frontend(int cpu, + struct perf_evsel *evsel + __maybe_unused, double avg) +{ + double total, ratio = 0.0; + const char *color; + + total = avg_stats(&runtime_cycles_stats[cpu]); + + if (total) + ratio = avg / total * 100.0; + + color = get_ratio_color(GRC_STALLED_CYCLES_FE, ratio); + + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " frontend cycles idle "); +} + +static void print_stalled_cycles_backend(int cpu, + struct perf_evsel *evsel + __maybe_unused, double avg) +{ + double total, ratio = 0.0; + const char *color; + + total = avg_stats(&runtime_cycles_stats[cpu]); + + if (total) + ratio = avg / total * 100.0; + + color = get_ratio_color(GRC_STALLED_CYCLES_BE, ratio); + + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " backend cycles idle "); +} + +static void print_branch_misses(int cpu, + struct perf_evsel *evsel __maybe_unused, + double avg) +{ + double total, ratio = 0.0; + const char *color; + + total = avg_stats(&runtime_branches_stats[cpu]); + + if (total) + ratio = avg / total * 100.0; + + color = get_ratio_color(GRC_CACHE_MISSES, ratio); + + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all branches "); +} + +static void print_l1_dcache_misses(int cpu, + struct perf_evsel *evsel __maybe_unused, + double avg) +{ + double total, ratio = 0.0; + const char *color; + + total = avg_stats(&runtime_l1_dcache_stats[cpu]); + + if (total) + ratio = avg / total * 100.0; + + color = get_ratio_color(GRC_CACHE_MISSES, ratio); + + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all L1-dcache hits "); +} + +static void print_l1_icache_misses(int cpu, + struct perf_evsel *evsel __maybe_unused, + double avg) +{ + double total, ratio = 0.0; + const char *color; + + total = avg_stats(&runtime_l1_icache_stats[cpu]); + + if (total) + ratio = avg / total * 100.0; + + color = get_ratio_color(GRC_CACHE_MISSES, ratio); + + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all L1-icache hits "); +} + +static void print_dtlb_cache_misses(int cpu, + struct perf_evsel *evsel __maybe_unused, + double avg) +{ + double total, ratio = 0.0; + const char *color; + + total = avg_stats(&runtime_dtlb_cache_stats[cpu]); + + if (total) + ratio = avg / total * 100.0; + + color = get_ratio_color(GRC_CACHE_MISSES, ratio); + + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all dTLB cache hits "); +} + +static void print_itlb_cache_misses(int cpu, + struct perf_evsel *evsel __maybe_unused, + double avg) +{ + double total, ratio = 0.0; + const char *color; + + total = avg_stats(&runtime_itlb_cache_stats[cpu]); + + if (total) + ratio = avg / total * 100.0; + + color = get_ratio_color(GRC_CACHE_MISSES, ratio); + + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all iTLB cache hits "); +} + +static void print_ll_cache_misses(int cpu, + struct perf_evsel *evsel __maybe_unused, + double avg) +{ + double total, ratio = 0.0; + const char *color; + + total = avg_stats(&runtime_ll_cache_stats[cpu]); + + if (total) + ratio = avg / total * 100.0; + + color = get_ratio_color(GRC_CACHE_MISSES, ratio); + + fprintf(output, " # "); + color_fprintf(output, color, "%6.2f%%", ratio); + fprintf(output, " of all LL-cache hits "); +} + +static void abs_printout(int cpu, int nr, struct perf_evsel *evsel, double avg) +{ + double total, ratio = 0.0; + char cpustr[16] = { '\0', }; + const char *fmt; + + if (csv_output) + fmt = "%s%.0f%s%s"; + else if (big_num) + fmt = "%s%'18.0f%s%-25s"; + else + fmt = "%s%18.0f%s%-25s"; + + if (aggr_socket) + sprintf(cpustr, "S%*d%s%*d%s", + csv_output ? 0 : -5, + cpu, + csv_sep, + csv_output ? 0 : 4, + nr, + csv_sep); + else if (no_aggr) + sprintf(cpustr, "CPU%*d%s", + csv_output ? 0 : -4, + perf_evsel__cpus(evsel)->map[cpu], csv_sep); + else + cpu = 0; + + fprintf(output, fmt, cpustr, avg, csv_sep, perf_evsel__name(evsel)); + + if (evsel->cgrp) + fprintf(output, "%s%s", csv_sep, evsel->cgrp->name); + + if (csv_output || interval) + return; + + if (perf_evsel__match(evsel, HARDWARE, HW_INSTRUCTIONS)) { + total = avg_stats(&runtime_cycles_stats[cpu]); + if (total) + ratio = avg / total; + + fprintf(output, " # %5.2f insns per cycle ", ratio); + + total = avg_stats(&runtime_stalled_cycles_front_stats[cpu]); + total = max(total, avg_stats(&runtime_stalled_cycles_back_stats[cpu])); + + if (total && avg) { + ratio = total / avg; + fprintf(output, "\n # %5.2f stalled cycles per insn", ratio); + } + + } else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) && + runtime_branches_stats[cpu].n != 0) { + print_branch_misses(cpu, evsel, avg); + } else if ( + evsel->attr.type == PERF_TYPE_HW_CACHE && + evsel->attr.config == ( PERF_COUNT_HW_CACHE_L1D | + ((PERF_COUNT_HW_CACHE_OP_READ) << 8) | + ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) && + runtime_l1_dcache_stats[cpu].n != 0) { + print_l1_dcache_misses(cpu, evsel, avg); + } else if ( + evsel->attr.type == PERF_TYPE_HW_CACHE && + evsel->attr.config == ( PERF_COUNT_HW_CACHE_L1I | + ((PERF_COUNT_HW_CACHE_OP_READ) << 8) | + ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) && + runtime_l1_icache_stats[cpu].n != 0) { + print_l1_icache_misses(cpu, evsel, avg); + } else if ( + evsel->attr.type == PERF_TYPE_HW_CACHE && + evsel->attr.config == ( PERF_COUNT_HW_CACHE_DTLB | + ((PERF_COUNT_HW_CACHE_OP_READ) << 8) | + ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) && + runtime_dtlb_cache_stats[cpu].n != 0) { + print_dtlb_cache_misses(cpu, evsel, avg); + } else if ( + evsel->attr.type == PERF_TYPE_HW_CACHE && + evsel->attr.config == ( PERF_COUNT_HW_CACHE_ITLB | + ((PERF_COUNT_HW_CACHE_OP_READ) << 8) | + ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) && + runtime_itlb_cache_stats[cpu].n != 0) { + print_itlb_cache_misses(cpu, evsel, avg); + } else if ( + evsel->attr.type == PERF_TYPE_HW_CACHE && + evsel->attr.config == ( PERF_COUNT_HW_CACHE_LL | + ((PERF_COUNT_HW_CACHE_OP_READ) << 8) | + ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) && + runtime_ll_cache_stats[cpu].n != 0) { + print_ll_cache_misses(cpu, evsel, avg); + } else if (perf_evsel__match(evsel, HARDWARE, HW_CACHE_MISSES) && + runtime_cacherefs_stats[cpu].n != 0) { + total = avg_stats(&runtime_cacherefs_stats[cpu]); + + if (total) + ratio = avg * 100 / total; + + fprintf(output, " # %8.3f %% of all cache refs ", ratio); + + } else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_FRONTEND)) { + print_stalled_cycles_frontend(cpu, evsel, avg); + } else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_BACKEND)) { + print_stalled_cycles_backend(cpu, evsel, avg); + } else if (perf_evsel__match(evsel, HARDWARE, HW_CPU_CYCLES)) { + total = avg_stats(&runtime_nsecs_stats[cpu]); + + if (total) + ratio = 1.0 * avg / total; + + fprintf(output, " # %8.3f GHz ", ratio); + } else if (runtime_nsecs_stats[cpu].n != 0) { + char unit = 'M'; + + total = avg_stats(&runtime_nsecs_stats[cpu]); + + if (total) + ratio = 1000.0 * avg / total; + if (ratio < 0.001) { + ratio *= 1000; + unit = 'K'; + } + + fprintf(output, " # %8.3f %c/sec ", ratio, unit); + } else { + fprintf(output, " "); + } +} + +static void print_aggr_socket(char *prefix) +{ + struct perf_evsel *counter; + u64 ena, run, val; + int cpu, s, s2, sock, nr; + + if (!sock_map) + return; + + for (s = 0; s < sock_map->nr; s++) { + sock = cpu_map__socket(sock_map, s); + list_for_each_entry(counter, &evsel_list->entries, node) { + val = ena = run = 0; + nr = 0; + for (cpu = 0; cpu < perf_evsel__nr_cpus(counter); cpu++) { + s2 = cpu_map__get_socket(evsel_list->cpus, cpu); + if (s2 != sock) + continue; + val += counter->counts->cpu[cpu].val; + ena += counter->counts->cpu[cpu].ena; + run += counter->counts->cpu[cpu].run; + nr++; + } + if (prefix) + fprintf(output, "%s", prefix); + + if (run == 0 || ena == 0) { + fprintf(output, "S%*d%s%*d%s%*s%s%*s", + csv_output ? 0 : -5, + s, + csv_sep, + csv_output ? 0 : 4, + nr, + csv_sep, + csv_output ? 0 : 18, + counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED, + csv_sep, + csv_output ? 0 : -24, + perf_evsel__name(counter)); + if (counter->cgrp) + fprintf(output, "%s%s", + csv_sep, counter->cgrp->name); + + fputc('\n', output); + continue; + } + + if (nsec_counter(counter)) + nsec_printout(sock, nr, counter, val); + else + abs_printout(sock, nr, counter, val); + + if (!csv_output) { + print_noise(counter, 1.0); + + if (run != ena) + fprintf(output, " (%.2f%%)", + 100.0 * run / ena); + } + fputc('\n', output); + } + } +} + +/* + * Print out the results of a single counter: + * aggregated counts in system-wide mode + */ +static void print_counter_aggr(struct perf_evsel *counter, char *prefix) +{ + struct perf_stat *ps = counter->priv; + double avg = avg_stats(&ps->res_stats[0]); + int scaled = counter->counts->scaled; + + if (prefix) + fprintf(output, "%s", prefix); + + if (scaled == -1) { + fprintf(output, "%*s%s%*s", + csv_output ? 0 : 18, + counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED, + csv_sep, + csv_output ? 0 : -24, + perf_evsel__name(counter)); + + if (counter->cgrp) + fprintf(output, "%s%s", csv_sep, counter->cgrp->name); + + fputc('\n', output); + return; + } + + if (nsec_counter(counter)) + nsec_printout(-1, 0, counter, avg); + else + abs_printout(-1, 0, counter, avg); + + print_noise(counter, avg); + + if (csv_output) { + fputc('\n', output); + return; + } + + if (scaled) { + double avg_enabled, avg_running; + + avg_enabled = avg_stats(&ps->res_stats[1]); + avg_running = avg_stats(&ps->res_stats[2]); + + fprintf(output, " [%5.2f%%]", 100 * avg_running / avg_enabled); + } + fprintf(output, "\n"); +} + +/* + * Print out the results of a single counter: + * does not use aggregated count in system-wide + */ +static void print_counter(struct perf_evsel *counter, char *prefix) +{ + u64 ena, run, val; + int cpu; + + for (cpu = 0; cpu < perf_evsel__nr_cpus(counter); cpu++) { + val = counter->counts->cpu[cpu].val; + ena = counter->counts->cpu[cpu].ena; + run = counter->counts->cpu[cpu].run; + + if (prefix) + fprintf(output, "%s", prefix); + + if (run == 0 || ena == 0) { + fprintf(output, "CPU%*d%s%*s%s%*s", + csv_output ? 0 : -4, + perf_evsel__cpus(counter)->map[cpu], csv_sep, + csv_output ? 0 : 18, + counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED, + csv_sep, + csv_output ? 0 : -24, + perf_evsel__name(counter)); + + if (counter->cgrp) + fprintf(output, "%s%s", + csv_sep, counter->cgrp->name); + + fputc('\n', output); + continue; + } + + if (nsec_counter(counter)) + nsec_printout(cpu, 0, counter, val); + else + abs_printout(cpu, 0, counter, val); + + if (!csv_output) { + print_noise(counter, 1.0); + + if (run != ena) + fprintf(output, " (%.2f%%)", + 100.0 * run / ena); + } + fputc('\n', output); + } +} + +static void print_stat(int argc, const char **argv) +{ + struct perf_evsel *counter; + int i; + + fflush(stdout); + + if (!csv_output) { + fprintf(output, "\n"); + fprintf(output, " Performance counter stats for "); + if (!perf_target__has_task(&target)) { + fprintf(output, "\'%s", argv[0]); + for (i = 1; i < argc; i++) + fprintf(output, " %s", argv[i]); + } else if (target.pid) + fprintf(output, "process id \'%s", target.pid); + else + fprintf(output, "thread id \'%s", target.tid); + + fprintf(output, "\'"); + if (run_count > 1) + fprintf(output, " (%d runs)", run_count); + fprintf(output, ":\n\n"); + } + + if (aggr_socket) + print_aggr_socket(NULL); + else if (no_aggr) { + list_for_each_entry(counter, &evsel_list->entries, node) + print_counter(counter, NULL); + } else { + list_for_each_entry(counter, &evsel_list->entries, node) + print_counter_aggr(counter, NULL); + } + + if (!csv_output) { + if (!null_run) + fprintf(output, "\n"); + fprintf(output, " %17.9f seconds time elapsed", + avg_stats(&walltime_nsecs_stats)/1e9); + if (run_count > 1) { + fprintf(output, " "); + print_noise_pct(stddev_stats(&walltime_nsecs_stats), + avg_stats(&walltime_nsecs_stats)); + } + fprintf(output, "\n\n"); + } +} + +static volatile int signr = -1; + +static void skip_signal(int signo) +{ + if ((child_pid == -1) || interval) + done = 1; + + signr = signo; +} + +static void sig_atexit(void) +{ + if (child_pid != -1) + kill(child_pid, SIGTERM); + + if (signr == -1) + return; + + signal(signr, SIG_DFL); + kill(getpid(), signr); +} + +static int stat__set_big_num(const struct option *opt __maybe_unused, + const char *s __maybe_unused, int unset) +{ + big_num_opt = unset ? 0 : 1; + return 0; +} + +/* + * Add default attributes, if there were no attributes specified or + * if -d/--detailed, -d -d or -d -d -d is used: + */ +static int add_default_attributes(void) +{ + struct perf_event_attr default_attrs[] = { + + { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK }, + { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES }, + { .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_STALLED_CYCLES_FRONTEND }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_STALLED_CYCLES_BACKEND }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES }, + +}; + +/* + * Detailed stats (-d), covering the L1 and last level data caches: + */ + struct perf_event_attr detailed_attrs[] = { + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1D << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1D << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_LL << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_LL << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, +}; + +/* + * Very detailed stats (-d -d), covering the instruction cache and the TLB caches: + */ + struct perf_event_attr very_detailed_attrs[] = { + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1I << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1I << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_DTLB << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_DTLB << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_ITLB << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_ITLB << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, + +}; + +/* + * Very, very detailed stats (-d -d -d), adding prefetch events: + */ + struct perf_event_attr very_very_detailed_attrs[] = { + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1D << 0 | + (PERF_COUNT_HW_CACHE_OP_PREFETCH << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1D << 0 | + (PERF_COUNT_HW_CACHE_OP_PREFETCH << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, +}; + + /* Set attrs if no event is selected and !null_run: */ + if (null_run) + return 0; + + if (!evsel_list->nr_entries) { + if (perf_evlist__add_default_attrs(evsel_list, default_attrs) < 0) + return -1; + } + + /* Detailed events get appended to the event list: */ + + if (detailed_run < 1) + return 0; + + /* Append detailed run extra attributes: */ + if (perf_evlist__add_default_attrs(evsel_list, detailed_attrs) < 0) + return -1; + + if (detailed_run < 2) + return 0; + + /* Append very detailed run extra attributes: */ + if (perf_evlist__add_default_attrs(evsel_list, very_detailed_attrs) < 0) + return -1; + + if (detailed_run < 3) + return 0; + + /* Append very, very detailed run extra attributes: */ + return perf_evlist__add_default_attrs(evsel_list, very_very_detailed_attrs); +} + +int cmd_stat(int argc, const char **argv, const char *prefix __maybe_unused) +{ + bool append_file = false; + int output_fd = 0; + const char *output_name = NULL; + const struct option options[] = { + OPT_CALLBACK('e', "event", &evsel_list, "event", + "event selector. use 'perf list' to list available events", + parse_events_option), + OPT_CALLBACK(0, "filter", &evsel_list, "filter", + "event filter", parse_filter), + OPT_BOOLEAN('i', "no-inherit", &no_inherit, + "child tasks do not inherit counters"), + OPT_STRING('p', "pid", &target.pid, "pid", + "stat events on existing process id"), + OPT_STRING('t', "tid", &target.tid, "tid", + "stat events on existing thread id"), + OPT_BOOLEAN('a', "all-cpus", &target.system_wide, + "system-wide collection from all CPUs"), + OPT_BOOLEAN('g', "group", &group, + "put the counters into a counter group"), + OPT_BOOLEAN('c', "scale", &scale, "scale/normalize counters"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show counter open errors, etc)"), + OPT_INTEGER('r', "repeat", &run_count, + "repeat command and print average + stddev (max: 100)"), + OPT_BOOLEAN('n', "null", &null_run, + "null run - dont start any counters"), + OPT_INCR('d', "detailed", &detailed_run, + "detailed run - start a lot of events"), + OPT_BOOLEAN('S', "sync", &sync_run, + "call sync() before starting a run"), + OPT_CALLBACK_NOOPT('B', "big-num", NULL, NULL, + "print large numbers with thousands\' separators", + stat__set_big_num), + OPT_STRING('C', "cpu", &target.cpu_list, "cpu", + "list of cpus to monitor in system-wide"), + OPT_BOOLEAN('A', "no-aggr", &no_aggr, "disable CPU count aggregation"), + OPT_STRING('x', "field-separator", &csv_sep, "separator", + "print counts with custom separator"), + OPT_CALLBACK('G', "cgroup", &evsel_list, "name", + "monitor event in cgroup name only", parse_cgroups), + OPT_STRING('o', "output", &output_name, "file", "output file name"), + OPT_BOOLEAN(0, "append", &append_file, "append to the output file"), + OPT_INTEGER(0, "log-fd", &output_fd, + "log output to fd, instead of stderr"), + OPT_STRING(0, "pre", &pre_cmd, "command", + "command to run prior to the measured command"), + OPT_STRING(0, "post", &post_cmd, "command", + "command to run after to the measured command"), + OPT_UINTEGER('I', "interval-print", &interval, + "print counts at regular interval in ms (>= 100)"), + OPT_BOOLEAN(0, "aggr-socket", &aggr_socket, "aggregate counts per processor socket"), + OPT_END() + }; + const char * const stat_usage[] = { + "perf stat [<options>] [<command>]", + NULL + }; + struct perf_evsel *pos; + int status = -ENOMEM, run_idx; + const char *mode; + + setlocale(LC_ALL, ""); + + evsel_list = perf_evlist__new(NULL, NULL); + if (evsel_list == NULL) + return -ENOMEM; + + argc = parse_options(argc, argv, options, stat_usage, + PARSE_OPT_STOP_AT_NON_OPTION); + + output = stderr; + if (output_name && strcmp(output_name, "-")) + output = NULL; + + if (output_name && output_fd) { + fprintf(stderr, "cannot use both --output and --log-fd\n"); + usage_with_options(stat_usage, options); + } + + if (output_fd < 0) { + fprintf(stderr, "argument to --log-fd must be a > 0\n"); + usage_with_options(stat_usage, options); + } + + if (!output) { + struct timespec tm; + mode = append_file ? "a" : "w"; + + output = fopen(output_name, mode); + if (!output) { + perror("failed to create output file"); + return -1; + } + clock_gettime(CLOCK_REALTIME, &tm); + fprintf(output, "# started on %s\n", ctime(&tm.tv_sec)); + } else if (output_fd > 0) { + mode = append_file ? "a" : "w"; + output = fdopen(output_fd, mode); + if (!output) { + perror("Failed opening logfd"); + return -errno; + } + } + + if (csv_sep) { + csv_output = true; + if (!strcmp(csv_sep, "\\t")) + csv_sep = "\t"; + } else + csv_sep = DEFAULT_SEPARATOR; + + /* + * let the spreadsheet do the pretty-printing + */ + if (csv_output) { + /* User explicitly passed -B? */ + if (big_num_opt == 1) { + fprintf(stderr, "-B option not supported with -x\n"); + usage_with_options(stat_usage, options); + } else /* Nope, so disable big number formatting */ + big_num = false; + } else if (big_num_opt == 0) /* User passed --no-big-num */ + big_num = false; + + if (!argc && !perf_target__has_task(&target)) + usage_with_options(stat_usage, options); + if (run_count <= 0) + usage_with_options(stat_usage, options); + + /* no_aggr, cgroup are for system-wide only */ + if ((no_aggr || nr_cgroups) && !perf_target__has_cpu(&target)) { + fprintf(stderr, "both cgroup and no-aggregation " + "modes only available in system-wide mode\n"); + + usage_with_options(stat_usage, options); + } + + if (aggr_socket) { + if (!perf_target__has_cpu(&target)) { + fprintf(stderr, "--aggr-socket only available in system-wide mode (-a)\n"); + usage_with_options(stat_usage, options); + } + no_aggr = true; + } + + if (add_default_attributes()) + goto out; + + perf_target__validate(&target); + + if (perf_evlist__create_maps(evsel_list, &target) < 0) { + if (perf_target__has_task(&target)) + pr_err("Problems finding threads of monitor\n"); + if (perf_target__has_cpu(&target)) + perror("failed to parse CPUs map"); + + usage_with_options(stat_usage, options); + return -1; + } + if (interval && interval < 100) { + pr_err("print interval must be >= 100ms\n"); + usage_with_options(stat_usage, options); + return -1; + } + + list_for_each_entry(pos, &evsel_list->entries, node) { + if (perf_evsel__alloc_stat_priv(pos) < 0 || + perf_evsel__alloc_counts(pos, perf_evsel__nr_cpus(pos)) < 0) + goto out_free_fd; + } + if (interval) { + list_for_each_entry(pos, &evsel_list->entries, node) { + if (perf_evsel__alloc_prev_raw_counts(pos) < 0) + goto out_free_fd; + } + } + + /* + * 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(SIGCHLD, skip_signal); + signal(SIGALRM, skip_signal); + signal(SIGABRT, skip_signal); + + status = 0; + for (run_idx = 0; run_idx < run_count; run_idx++) { + if (run_count != 1 && verbose) + fprintf(output, "[ perf stat: executing run #%d ... ]\n", + run_idx + 1); + + status = run_perf_stat(argc, argv); + } + + if (status != -1 && !interval) + print_stat(argc, argv); +out_free_fd: + list_for_each_entry(pos, &evsel_list->entries, node) { + perf_evsel__free_stat_priv(pos); + perf_evsel__free_counts(pos); + perf_evsel__free_prev_raw_counts(pos); + } + perf_evlist__delete_maps(evsel_list); +out: + perf_evlist__delete(evsel_list); + return status; +} diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c new file mode 100644 index 00000000..ab4cf232 --- /dev/null +++ b/tools/perf/builtin-timechart.c @@ -0,0 +1,1095 @@ +/* + * builtin-timechart.c - make an svg timechart of system activity + * + * (C) Copyright 2009 Intel Corporation + * + * Authors: + * Arjan van de Ven <arjan@linux.intel.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 + * of the License. + */ + +#include "builtin.h" + +#include "util/util.h" + +#include "util/color.h" +#include <linux/list.h> +#include "util/cache.h" +#include "util/evsel.h" +#include <linux/rbtree.h> +#include "util/symbol.h" +#include "util/callchain.h" +#include "util/strlist.h" + +#include "perf.h" +#include "util/header.h" +#include "util/parse-options.h" +#include "util/parse-events.h" +#include "util/event.h" +#include "util/session.h" +#include "util/svghelper.h" +#include "util/tool.h" + +#define SUPPORT_OLD_POWER_EVENTS 1 +#define PWR_EVENT_EXIT -1 + + +static unsigned int numcpus; +static u64 min_freq; /* Lowest CPU frequency seen */ +static u64 max_freq; /* Highest CPU frequency seen */ +static u64 turbo_frequency; + +static u64 first_time, last_time; + +static bool power_only; + + +struct per_pid; +struct per_pidcomm; + +struct cpu_sample; +struct power_event; +struct wake_event; + +struct sample_wrapper; + +/* + * Datastructure layout: + * We keep an list of "pid"s, matching the kernels notion of a task struct. + * Each "pid" entry, has a list of "comm"s. + * this is because we want to track different programs different, while + * exec will reuse the original pid (by design). + * Each comm has a list of samples that will be used to draw + * final graph. + */ + +struct per_pid { + struct per_pid *next; + + int pid; + int ppid; + + u64 start_time; + u64 end_time; + u64 total_time; + int display; + + struct per_pidcomm *all; + struct per_pidcomm *current; +}; + + +struct per_pidcomm { + struct per_pidcomm *next; + + u64 start_time; + u64 end_time; + u64 total_time; + + int Y; + int display; + + long state; + u64 state_since; + + char *comm; + + struct cpu_sample *samples; +}; + +struct sample_wrapper { + struct sample_wrapper *next; + + u64 timestamp; + unsigned char data[0]; +}; + +#define TYPE_NONE 0 +#define TYPE_RUNNING 1 +#define TYPE_WAITING 2 +#define TYPE_BLOCKED 3 + +struct cpu_sample { + struct cpu_sample *next; + + u64 start_time; + u64 end_time; + int type; + int cpu; +}; + +static struct per_pid *all_data; + +#define CSTATE 1 +#define PSTATE 2 + +struct power_event { + struct power_event *next; + int type; + int state; + u64 start_time; + u64 end_time; + int cpu; +}; + +struct wake_event { + struct wake_event *next; + int waker; + int wakee; + u64 time; +}; + +static struct power_event *power_events; +static struct wake_event *wake_events; + +struct process_filter; +struct process_filter { + char *name; + int pid; + struct process_filter *next; +}; + +static struct process_filter *process_filter; + + +static struct per_pid *find_create_pid(int pid) +{ + struct per_pid *cursor = all_data; + + while (cursor) { + if (cursor->pid == pid) + return cursor; + cursor = cursor->next; + } + cursor = zalloc(sizeof(*cursor)); + assert(cursor != NULL); + cursor->pid = pid; + cursor->next = all_data; + all_data = cursor; + return cursor; +} + +static void pid_set_comm(int pid, char *comm) +{ + struct per_pid *p; + struct per_pidcomm *c; + p = find_create_pid(pid); + c = p->all; + while (c) { + if (c->comm && strcmp(c->comm, comm) == 0) { + p->current = c; + return; + } + if (!c->comm) { + c->comm = strdup(comm); + p->current = c; + return; + } + c = c->next; + } + c = zalloc(sizeof(*c)); + assert(c != NULL); + c->comm = strdup(comm); + p->current = c; + c->next = p->all; + p->all = c; +} + +static void pid_fork(int pid, int ppid, u64 timestamp) +{ + struct per_pid *p, *pp; + p = find_create_pid(pid); + pp = find_create_pid(ppid); + p->ppid = ppid; + if (pp->current && pp->current->comm && !p->current) + pid_set_comm(pid, pp->current->comm); + + p->start_time = timestamp; + if (p->current) { + p->current->start_time = timestamp; + p->current->state_since = timestamp; + } +} + +static void pid_exit(int pid, u64 timestamp) +{ + struct per_pid *p; + p = find_create_pid(pid); + p->end_time = timestamp; + if (p->current) + p->current->end_time = timestamp; +} + +static void +pid_put_sample(int pid, int type, unsigned int cpu, u64 start, u64 end) +{ + struct per_pid *p; + struct per_pidcomm *c; + struct cpu_sample *sample; + + p = find_create_pid(pid); + c = p->current; + if (!c) { + c = zalloc(sizeof(*c)); + assert(c != NULL); + p->current = c; + c->next = p->all; + p->all = c; + } + + sample = zalloc(sizeof(*sample)); + assert(sample != NULL); + sample->start_time = start; + sample->end_time = end; + sample->type = type; + sample->next = c->samples; + sample->cpu = cpu; + c->samples = sample; + + if (sample->type == TYPE_RUNNING && end > start && start > 0) { + c->total_time += (end-start); + p->total_time += (end-start); + } + + if (c->start_time == 0 || c->start_time > start) + c->start_time = start; + if (p->start_time == 0 || p->start_time > start) + p->start_time = start; +} + +#define MAX_CPUS 4096 + +static u64 cpus_cstate_start_times[MAX_CPUS]; +static int cpus_cstate_state[MAX_CPUS]; +static u64 cpus_pstate_start_times[MAX_CPUS]; +static u64 cpus_pstate_state[MAX_CPUS]; + +static int process_comm_event(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine __maybe_unused) +{ + pid_set_comm(event->comm.tid, event->comm.comm); + return 0; +} + +static int process_fork_event(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine __maybe_unused) +{ + pid_fork(event->fork.pid, event->fork.ppid, event->fork.time); + return 0; +} + +static int process_exit_event(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine __maybe_unused) +{ + pid_exit(event->fork.pid, event->fork.time); + return 0; +} + +struct trace_entry { + unsigned short type; + unsigned char flags; + unsigned char preempt_count; + int pid; + int lock_depth; +}; + +#ifdef SUPPORT_OLD_POWER_EVENTS +static int use_old_power_events; +struct power_entry_old { + struct trace_entry te; + u64 type; + u64 value; + u64 cpu_id; +}; +#endif + +struct power_processor_entry { + struct trace_entry te; + u32 state; + u32 cpu_id; +}; + +#define TASK_COMM_LEN 16 +struct wakeup_entry { + struct trace_entry te; + char comm[TASK_COMM_LEN]; + int pid; + int prio; + int success; +}; + +/* + * trace_flag_type is an enumeration that holds different + * states when a trace occurs. These are: + * IRQS_OFF - interrupts were disabled + * IRQS_NOSUPPORT - arch does not support irqs_disabled_flags + * NEED_RESCED - reschedule is requested + * HARDIRQ - inside an interrupt handler + * SOFTIRQ - inside a softirq handler + */ +enum trace_flag_type { + TRACE_FLAG_IRQS_OFF = 0x01, + TRACE_FLAG_IRQS_NOSUPPORT = 0x02, + TRACE_FLAG_NEED_RESCHED = 0x04, + TRACE_FLAG_HARDIRQ = 0x08, + TRACE_FLAG_SOFTIRQ = 0x10, +}; + + + +struct sched_switch { + struct trace_entry te; + char prev_comm[TASK_COMM_LEN]; + int prev_pid; + int prev_prio; + long prev_state; /* Arjan weeps. */ + char next_comm[TASK_COMM_LEN]; + int next_pid; + int next_prio; +}; + +static void c_state_start(int cpu, u64 timestamp, int state) +{ + cpus_cstate_start_times[cpu] = timestamp; + cpus_cstate_state[cpu] = state; +} + +static void c_state_end(int cpu, u64 timestamp) +{ + struct power_event *pwr = zalloc(sizeof(*pwr)); + + if (!pwr) + return; + + pwr->state = cpus_cstate_state[cpu]; + pwr->start_time = cpus_cstate_start_times[cpu]; + pwr->end_time = timestamp; + pwr->cpu = cpu; + pwr->type = CSTATE; + pwr->next = power_events; + + power_events = pwr; +} + +static void p_state_change(int cpu, u64 timestamp, u64 new_freq) +{ + struct power_event *pwr; + + if (new_freq > 8000000) /* detect invalid data */ + return; + + pwr = zalloc(sizeof(*pwr)); + if (!pwr) + return; + + pwr->state = cpus_pstate_state[cpu]; + pwr->start_time = cpus_pstate_start_times[cpu]; + pwr->end_time = timestamp; + pwr->cpu = cpu; + pwr->type = PSTATE; + pwr->next = power_events; + + if (!pwr->start_time) + pwr->start_time = first_time; + + power_events = pwr; + + cpus_pstate_state[cpu] = new_freq; + cpus_pstate_start_times[cpu] = timestamp; + + if ((u64)new_freq > max_freq) + max_freq = new_freq; + + if (new_freq < min_freq || min_freq == 0) + min_freq = new_freq; + + if (new_freq == max_freq - 1000) + turbo_frequency = max_freq; +} + +static void +sched_wakeup(int cpu, u64 timestamp, int pid, struct trace_entry *te) +{ + struct per_pid *p; + struct wakeup_entry *wake = (void *)te; + struct wake_event *we = zalloc(sizeof(*we)); + + if (!we) + return; + + we->time = timestamp; + we->waker = pid; + + if ((te->flags & TRACE_FLAG_HARDIRQ) || (te->flags & TRACE_FLAG_SOFTIRQ)) + we->waker = -1; + + we->wakee = wake->pid; + we->next = wake_events; + wake_events = we; + p = find_create_pid(we->wakee); + + if (p && p->current && p->current->state == TYPE_NONE) { + p->current->state_since = timestamp; + p->current->state = TYPE_WAITING; + } + if (p && p->current && p->current->state == TYPE_BLOCKED) { + pid_put_sample(p->pid, p->current->state, cpu, p->current->state_since, timestamp); + p->current->state_since = timestamp; + p->current->state = TYPE_WAITING; + } +} + +static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te) +{ + struct per_pid *p = NULL, *prev_p; + struct sched_switch *sw = (void *)te; + + + prev_p = find_create_pid(sw->prev_pid); + + p = find_create_pid(sw->next_pid); + + if (prev_p->current && prev_p->current->state != TYPE_NONE) + pid_put_sample(sw->prev_pid, TYPE_RUNNING, cpu, prev_p->current->state_since, timestamp); + if (p && p->current) { + if (p->current->state != TYPE_NONE) + pid_put_sample(sw->next_pid, p->current->state, cpu, p->current->state_since, timestamp); + + p->current->state_since = timestamp; + p->current->state = TYPE_RUNNING; + } + + if (prev_p->current) { + prev_p->current->state = TYPE_NONE; + prev_p->current->state_since = timestamp; + if (sw->prev_state & 2) + prev_p->current->state = TYPE_BLOCKED; + if (sw->prev_state == 0) + prev_p->current->state = TYPE_WAITING; + } +} + + +static int process_sample_event(struct perf_tool *tool __maybe_unused, + union perf_event *event __maybe_unused, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine __maybe_unused) +{ + struct trace_entry *te; + + if (evsel->attr.sample_type & PERF_SAMPLE_TIME) { + if (!first_time || first_time > sample->time) + first_time = sample->time; + if (last_time < sample->time) + last_time = sample->time; + } + + te = (void *)sample->raw_data; + if ((evsel->attr.sample_type & PERF_SAMPLE_RAW) && sample->raw_size > 0) { + char *event_str; +#ifdef SUPPORT_OLD_POWER_EVENTS + struct power_entry_old *peo; + peo = (void *)te; +#endif + /* + * FIXME: use evsel, its already mapped from id to perf_evsel, + * remove perf_header__find_event infrastructure bits. + * Mapping all these "power:cpu_idle" strings to the tracepoint + * ID and then just comparing against evsel->attr.config. + * + * e.g.: + * + * if (evsel->attr.config == power_cpu_idle_id) + */ + event_str = perf_header__find_event(te->type); + + if (!event_str) + return 0; + + if (sample->cpu > numcpus) + numcpus = sample->cpu; + + if (strcmp(event_str, "power:cpu_idle") == 0) { + struct power_processor_entry *ppe = (void *)te; + if (ppe->state == (u32)PWR_EVENT_EXIT) + c_state_end(ppe->cpu_id, sample->time); + else + c_state_start(ppe->cpu_id, sample->time, + ppe->state); + } + else if (strcmp(event_str, "power:cpu_frequency") == 0) { + struct power_processor_entry *ppe = (void *)te; + p_state_change(ppe->cpu_id, sample->time, ppe->state); + } + + else if (strcmp(event_str, "sched:sched_wakeup") == 0) + sched_wakeup(sample->cpu, sample->time, sample->pid, te); + + else if (strcmp(event_str, "sched:sched_switch") == 0) + sched_switch(sample->cpu, sample->time, te); + +#ifdef SUPPORT_OLD_POWER_EVENTS + if (use_old_power_events) { + if (strcmp(event_str, "power:power_start") == 0) + c_state_start(peo->cpu_id, sample->time, + peo->value); + + else if (strcmp(event_str, "power:power_end") == 0) + c_state_end(sample->cpu, sample->time); + + else if (strcmp(event_str, + "power:power_frequency") == 0) + p_state_change(peo->cpu_id, sample->time, + peo->value); + } +#endif + } + return 0; +} + +/* + * After the last sample we need to wrap up the current C/P state + * and close out each CPU for these. + */ +static void end_sample_processing(void) +{ + u64 cpu; + struct power_event *pwr; + + for (cpu = 0; cpu <= numcpus; cpu++) { + /* C state */ +#if 0 + pwr = zalloc(sizeof(*pwr)); + if (!pwr) + return; + + pwr->state = cpus_cstate_state[cpu]; + pwr->start_time = cpus_cstate_start_times[cpu]; + pwr->end_time = last_time; + pwr->cpu = cpu; + pwr->type = CSTATE; + pwr->next = power_events; + + power_events = pwr; +#endif + /* P state */ + + pwr = zalloc(sizeof(*pwr)); + if (!pwr) + return; + + pwr->state = cpus_pstate_state[cpu]; + pwr->start_time = cpus_pstate_start_times[cpu]; + pwr->end_time = last_time; + pwr->cpu = cpu; + pwr->type = PSTATE; + pwr->next = power_events; + + if (!pwr->start_time) + pwr->start_time = first_time; + if (!pwr->state) + pwr->state = min_freq; + power_events = pwr; + } +} + +/* + * Sort the pid datastructure + */ +static void sort_pids(void) +{ + struct per_pid *new_list, *p, *cursor, *prev; + /* sort by ppid first, then by pid, lowest to highest */ + + new_list = NULL; + + while (all_data) { + p = all_data; + all_data = p->next; + p->next = NULL; + + if (new_list == NULL) { + new_list = p; + p->next = NULL; + continue; + } + prev = NULL; + cursor = new_list; + while (cursor) { + if (cursor->ppid > p->ppid || + (cursor->ppid == p->ppid && cursor->pid > p->pid)) { + /* must insert before */ + if (prev) { + p->next = prev->next; + prev->next = p; + cursor = NULL; + continue; + } else { + p->next = new_list; + new_list = p; + cursor = NULL; + continue; + } + } + + prev = cursor; + cursor = cursor->next; + if (!cursor) + prev->next = p; + } + } + all_data = new_list; +} + + +static void draw_c_p_states(void) +{ + struct power_event *pwr; + pwr = power_events; + + /* + * two pass drawing so that the P state bars are on top of the C state blocks + */ + while (pwr) { + if (pwr->type == CSTATE) + svg_cstate(pwr->cpu, pwr->start_time, pwr->end_time, pwr->state); + pwr = pwr->next; + } + + pwr = power_events; + while (pwr) { + if (pwr->type == PSTATE) { + if (!pwr->state) + pwr->state = min_freq; + svg_pstate(pwr->cpu, pwr->start_time, pwr->end_time, pwr->state); + } + pwr = pwr->next; + } +} + +static void draw_wakeups(void) +{ + struct wake_event *we; + struct per_pid *p; + struct per_pidcomm *c; + + we = wake_events; + while (we) { + int from = 0, to = 0; + char *task_from = NULL, *task_to = NULL; + + /* locate the column of the waker and wakee */ + p = all_data; + while (p) { + if (p->pid == we->waker || p->pid == we->wakee) { + c = p->all; + while (c) { + if (c->Y && c->start_time <= we->time && c->end_time >= we->time) { + if (p->pid == we->waker && !from) { + from = c->Y; + task_from = strdup(c->comm); + } + if (p->pid == we->wakee && !to) { + to = c->Y; + task_to = strdup(c->comm); + } + } + c = c->next; + } + c = p->all; + while (c) { + if (p->pid == we->waker && !from) { + from = c->Y; + task_from = strdup(c->comm); + } + if (p->pid == we->wakee && !to) { + to = c->Y; + task_to = strdup(c->comm); + } + c = c->next; + } + } + p = p->next; + } + + if (!task_from) { + task_from = malloc(40); + sprintf(task_from, "[%i]", we->waker); + } + if (!task_to) { + task_to = malloc(40); + sprintf(task_to, "[%i]", we->wakee); + } + + if (we->waker == -1) + svg_interrupt(we->time, to); + else if (from && to && abs(from - to) == 1) + svg_wakeline(we->time, from, to); + else + svg_partial_wakeline(we->time, from, task_from, to, task_to); + we = we->next; + + free(task_from); + free(task_to); + } +} + +static void draw_cpu_usage(void) +{ + struct per_pid *p; + struct per_pidcomm *c; + struct cpu_sample *sample; + p = all_data; + while (p) { + c = p->all; + while (c) { + sample = c->samples; + while (sample) { + if (sample->type == TYPE_RUNNING) + svg_process(sample->cpu, sample->start_time, sample->end_time, "sample", c->comm); + + sample = sample->next; + } + c = c->next; + } + p = p->next; + } +} + +static void draw_process_bars(void) +{ + struct per_pid *p; + struct per_pidcomm *c; + struct cpu_sample *sample; + int Y = 0; + + Y = 2 * numcpus + 2; + + p = all_data; + while (p) { + c = p->all; + while (c) { + if (!c->display) { + c->Y = 0; + c = c->next; + continue; + } + + svg_box(Y, c->start_time, c->end_time, "process"); + sample = c->samples; + while (sample) { + if (sample->type == TYPE_RUNNING) + svg_sample(Y, sample->cpu, sample->start_time, sample->end_time); + if (sample->type == TYPE_BLOCKED) + svg_box(Y, sample->start_time, sample->end_time, "blocked"); + if (sample->type == TYPE_WAITING) + svg_waiting(Y, sample->start_time, sample->end_time); + sample = sample->next; + } + + if (c->comm) { + char comm[256]; + if (c->total_time > 5000000000) /* 5 seconds */ + sprintf(comm, "%s:%i (%2.2fs)", c->comm, p->pid, c->total_time / 1000000000.0); + else + sprintf(comm, "%s:%i (%3.1fms)", c->comm, p->pid, c->total_time / 1000000.0); + + svg_text(Y, c->start_time, comm); + } + c->Y = Y; + Y++; + c = c->next; + } + p = p->next; + } +} + +static void add_process_filter(const char *string) +{ + int pid = strtoull(string, NULL, 10); + struct process_filter *filt = malloc(sizeof(*filt)); + + if (!filt) + return; + + filt->name = strdup(string); + filt->pid = pid; + filt->next = process_filter; + + process_filter = filt; +} + +static int passes_filter(struct per_pid *p, struct per_pidcomm *c) +{ + struct process_filter *filt; + if (!process_filter) + return 1; + + filt = process_filter; + while (filt) { + if (filt->pid && p->pid == filt->pid) + return 1; + if (strcmp(filt->name, c->comm) == 0) + return 1; + filt = filt->next; + } + return 0; +} + +static int determine_display_tasks_filtered(void) +{ + struct per_pid *p; + struct per_pidcomm *c; + int count = 0; + + p = all_data; + while (p) { + p->display = 0; + if (p->start_time == 1) + p->start_time = first_time; + + /* no exit marker, task kept running to the end */ + if (p->end_time == 0) + p->end_time = last_time; + + c = p->all; + + while (c) { + c->display = 0; + + if (c->start_time == 1) + c->start_time = first_time; + + if (passes_filter(p, c)) { + c->display = 1; + p->display = 1; + count++; + } + + if (c->end_time == 0) + c->end_time = last_time; + + c = c->next; + } + p = p->next; + } + return count; +} + +static int determine_display_tasks(u64 threshold) +{ + struct per_pid *p; + struct per_pidcomm *c; + int count = 0; + + if (process_filter) + return determine_display_tasks_filtered(); + + p = all_data; + while (p) { + p->display = 0; + if (p->start_time == 1) + p->start_time = first_time; + + /* no exit marker, task kept running to the end */ + if (p->end_time == 0) + p->end_time = last_time; + if (p->total_time >= threshold && !power_only) + p->display = 1; + + c = p->all; + + while (c) { + c->display = 0; + + if (c->start_time == 1) + c->start_time = first_time; + + if (c->total_time >= threshold && !power_only) { + c->display = 1; + count++; + } + + if (c->end_time == 0) + c->end_time = last_time; + + c = c->next; + } + p = p->next; + } + return count; +} + + + +#define TIME_THRESH 10000000 + +static void write_svg_file(const char *filename) +{ + u64 i; + int count; + + numcpus++; + + + count = determine_display_tasks(TIME_THRESH); + + /* We'd like to show at least 15 tasks; be less picky if we have fewer */ + if (count < 15) + count = determine_display_tasks(TIME_THRESH / 10); + + open_svg(filename, numcpus, count, first_time, last_time); + + svg_time_grid(); + svg_legenda(); + + for (i = 0; i < numcpus; i++) + svg_cpu_box(i, max_freq, turbo_frequency); + + draw_cpu_usage(); + draw_process_bars(); + draw_c_p_states(); + draw_wakeups(); + + svg_close(); +} + +static int __cmd_timechart(const char *output_name) +{ + struct perf_tool perf_timechart = { + .comm = process_comm_event, + .fork = process_fork_event, + .exit = process_exit_event, + .sample = process_sample_event, + .ordered_samples = true, + }; + struct perf_session *session = perf_session__new(input_name, O_RDONLY, + 0, false, &perf_timechart); + int ret = -EINVAL; + + if (session == NULL) + return -ENOMEM; + + if (!perf_session__has_traces(session, "timechart record")) + goto out_delete; + + ret = perf_session__process_events(session, &perf_timechart); + if (ret) + goto out_delete; + + end_sample_processing(); + + sort_pids(); + + write_svg_file(output_name); + + pr_info("Written %2.1f seconds of trace to %s.\n", + (last_time - first_time) / 1000000000.0, output_name); +out_delete: + perf_session__delete(session); + return ret; +} + +static int __cmd_record(int argc, const char **argv) +{ +#ifdef SUPPORT_OLD_POWER_EVENTS + const char * const record_old_args[] = { + "record", "-a", "-R", "-f", "-c", "1", + "-e", "power:power_start", + "-e", "power:power_end", + "-e", "power:power_frequency", + "-e", "sched:sched_wakeup", + "-e", "sched:sched_switch", + }; +#endif + const char * const record_new_args[] = { + "record", "-a", "-R", "-f", "-c", "1", + "-e", "power:cpu_frequency", + "-e", "power:cpu_idle", + "-e", "sched:sched_wakeup", + "-e", "sched:sched_switch", + }; + unsigned int rec_argc, i, j; + const char **rec_argv; + const char * const *record_args = record_new_args; + unsigned int record_elems = ARRAY_SIZE(record_new_args); + +#ifdef SUPPORT_OLD_POWER_EVENTS + if (!is_valid_tracepoint("power:cpu_idle") && + is_valid_tracepoint("power:power_start")) { + use_old_power_events = 1; + record_args = record_old_args; + record_elems = ARRAY_SIZE(record_old_args); + } +#endif + + rec_argc = record_elems + argc - 1; + rec_argv = calloc(rec_argc + 1, sizeof(char *)); + + if (rec_argv == NULL) + return -ENOMEM; + + for (i = 0; i < record_elems; i++) + rec_argv[i] = strdup(record_args[i]); + + for (j = 1; j < (unsigned int)argc; j++, i++) + rec_argv[i] = argv[j]; + + return cmd_record(i, rec_argv, NULL); +} + +static int +parse_process(const struct option *opt __maybe_unused, const char *arg, + int __maybe_unused unset) +{ + if (arg) + add_process_filter(arg); + return 0; +} + +int cmd_timechart(int argc, const char **argv, + const char *prefix __maybe_unused) +{ + const char *output_name = "output.svg"; + const struct option options[] = { + OPT_STRING('i', "input", &input_name, "file", "input file name"), + OPT_STRING('o', "output", &output_name, "file", "output file name"), + OPT_INTEGER('w', "width", &svg_page_width, "page width"), + OPT_BOOLEAN('P', "power-only", &power_only, "output power data only"), + OPT_CALLBACK('p', "process", NULL, "process", + "process selector. Pass a pid or process name.", + parse_process), + OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory", + "Look for files with symbols relative to this directory"), + OPT_END() + }; + const char * const timechart_usage[] = { + "perf timechart [<options>] {record}", + NULL + }; + + argc = parse_options(argc, argv, options, timechart_usage, + PARSE_OPT_STOP_AT_NON_OPTION); + + symbol__init(); + + if (argc && !strncmp(argv[0], "rec", 3)) + return __cmd_record(argc, argv); + else if (argc) + usage_with_options(timechart_usage, options); + + setup_pager(); + + return __cmd_timechart(output_name); +} diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c new file mode 100644 index 00000000..72f6eb7b --- /dev/null +++ b/tools/perf/builtin-top.c @@ -0,0 +1,1230 @@ +/* + * 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> + * 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@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/annotate.h" +#include "util/cache.h" +#include "util/color.h" +#include "util/evlist.h" +#include "util/evsel.h" +#include "util/machine.h" +#include "util/session.h" +#include "util/symbol.h" +#include "util/thread.h" +#include "util/thread_map.h" +#include "util/top.h" +#include "util/util.h" +#include <linux/rbtree.h> +#include "util/parse-options.h" +#include "util/parse-events.h" +#include "util/cpumap.h" +#include "util/xyarray.h" +#include "util/sort.h" +#include "util/intlist.h" + +#include "util/debug.h" + +#include <assert.h> +#include <elf.h> +#include <fcntl.h> + +#include <stdio.h> +#include <termios.h> +#include <unistd.h> +#include <inttypes.h> + +#include <errno.h> +#include <time.h> +#include <sched.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/utsname.h> +#include <sys/mman.h> + +#include <linux/unistd.h> +#include <linux/types.h> + +static volatile int done; + +static void perf_top__update_print_entries(struct perf_top *top) +{ + if (top->print_entries > 9) + top->print_entries -= 9; +} + +static void perf_top__sig_winch(int sig __maybe_unused, + siginfo_t *info __maybe_unused, void *arg) +{ + struct perf_top *top = arg; + + get_term_dimensions(&top->winsize); + if (!top->print_entries + || (top->print_entries+4) > top->winsize.ws_row) { + top->print_entries = top->winsize.ws_row; + } else { + top->print_entries += 4; + top->winsize.ws_row = top->print_entries; + } + perf_top__update_print_entries(top); +} + +static int perf_top__parse_source(struct perf_top *top, struct hist_entry *he) +{ + struct symbol *sym; + struct annotation *notes; + struct map *map; + int err = -1; + + if (!he || !he->ms.sym) + return -1; + + sym = he->ms.sym; + map = he->ms.map; + + /* + * We can't annotate with just /proc/kallsyms + */ + if (map->dso->symtab_type == DSO_BINARY_TYPE__KALLSYMS) { + pr_err("Can't annotate %s: No vmlinux file was found in the " + "path\n", sym->name); + sleep(1); + return -1; + } + + notes = symbol__annotation(sym); + if (notes->src != NULL) { + pthread_mutex_lock(¬es->lock); + goto out_assign; + } + + pthread_mutex_lock(¬es->lock); + + if (symbol__alloc_hist(sym) < 0) { + pthread_mutex_unlock(¬es->lock); + pr_err("Not enough memory for annotating '%s' symbol!\n", + sym->name); + sleep(1); + return err; + } + + err = symbol__annotate(sym, map, 0); + if (err == 0) { +out_assign: + top->sym_filter_entry = he; + } + + pthread_mutex_unlock(¬es->lock); + return err; +} + +static void __zero_source_counters(struct hist_entry *he) +{ + struct symbol *sym = he->ms.sym; + symbol__annotate_zero_histograms(sym); +} + +static void ui__warn_map_erange(struct map *map, struct symbol *sym, u64 ip) +{ + struct utsname uts; + int err = uname(&uts); + + ui__warning("Out of bounds address found:\n\n" + "Addr: %" PRIx64 "\n" + "DSO: %s %c\n" + "Map: %" PRIx64 "-%" PRIx64 "\n" + "Symbol: %" PRIx64 "-%" PRIx64 " %c %s\n" + "Arch: %s\n" + "Kernel: %s\n" + "Tools: %s\n\n" + "Not all samples will be on the annotation output.\n\n" + "Please report to linux-kernel@vger.kernel.org\n", + ip, map->dso->long_name, dso__symtab_origin(map->dso), + map->start, map->end, sym->start, sym->end, + sym->binding == STB_GLOBAL ? 'g' : + sym->binding == STB_LOCAL ? 'l' : 'w', sym->name, + err ? "[unknown]" : uts.machine, + err ? "[unknown]" : uts.release, perf_version_string); + if (use_browser <= 0) + sleep(5); + + map->erange_warned = true; +} + +static void perf_top__record_precise_ip(struct perf_top *top, + struct hist_entry *he, + int counter, u64 ip) +{ + struct annotation *notes; + struct symbol *sym; + int err; + + if (he == NULL || he->ms.sym == NULL || + ((top->sym_filter_entry == NULL || + top->sym_filter_entry->ms.sym != he->ms.sym) && use_browser != 1)) + return; + + sym = he->ms.sym; + notes = symbol__annotation(sym); + + if (pthread_mutex_trylock(¬es->lock)) + return; + + if (notes->src == NULL && symbol__alloc_hist(sym) < 0) { + pthread_mutex_unlock(¬es->lock); + pr_err("Not enough memory for annotating '%s' symbol!\n", + sym->name); + sleep(1); + return; + } + + ip = he->ms.map->map_ip(he->ms.map, ip); + err = symbol__inc_addr_samples(sym, he->ms.map, counter, ip); + + pthread_mutex_unlock(¬es->lock); + + if (err == -ERANGE && !he->ms.map->erange_warned) + ui__warn_map_erange(he->ms.map, sym, ip); +} + +static void perf_top__show_details(struct perf_top *top) +{ + struct hist_entry *he = top->sym_filter_entry; + struct annotation *notes; + struct symbol *symbol; + int more; + + if (!he) + return; + + symbol = he->ms.sym; + notes = symbol__annotation(symbol); + + pthread_mutex_lock(¬es->lock); + + if (notes->src == NULL) + goto out_unlock; + + printf("Showing %s for %s\n", perf_evsel__name(top->sym_evsel), symbol->name); + printf(" Events Pcnt (>=%d%%)\n", top->sym_pcnt_filter); + + more = symbol__annotate_printf(symbol, he->ms.map, top->sym_evsel->idx, + 0, top->sym_pcnt_filter, top->print_entries, 4); + if (top->zero) + symbol__annotate_zero_histogram(symbol, top->sym_evsel->idx); + else + symbol__annotate_decay_histogram(symbol, top->sym_evsel->idx); + if (more != 0) + printf("%d lines not displayed, maybe increase display entries [e]\n", more); +out_unlock: + pthread_mutex_unlock(¬es->lock); +} + +static const char CONSOLE_CLEAR[] = "[H[2J"; + +static struct hist_entry *perf_evsel__add_hist_entry(struct perf_evsel *evsel, + struct addr_location *al, + struct perf_sample *sample) +{ + struct hist_entry *he; + + he = __hists__add_entry(&evsel->hists, al, NULL, sample->period); + if (he == NULL) + return NULL; + + hists__inc_nr_events(&evsel->hists, PERF_RECORD_SAMPLE); + return he; +} + +static void perf_top__print_sym_table(struct perf_top *top) +{ + char bf[160]; + int printed = 0; + const int win_width = top->winsize.ws_col - 1; + + puts(CONSOLE_CLEAR); + + perf_top__header_snprintf(top, bf, sizeof(bf)); + printf("%s\n", bf); + + perf_top__reset_sample_counters(top); + + printf("%-*.*s\n", win_width, win_width, graph_dotted_line); + + if (top->sym_evsel->hists.stats.nr_lost_warned != + top->sym_evsel->hists.stats.nr_events[PERF_RECORD_LOST]) { + top->sym_evsel->hists.stats.nr_lost_warned = + top->sym_evsel->hists.stats.nr_events[PERF_RECORD_LOST]; + color_fprintf(stdout, PERF_COLOR_RED, + "WARNING: LOST %d chunks, Check IO/CPU overload", + top->sym_evsel->hists.stats.nr_lost_warned); + ++printed; + } + + if (top->sym_filter_entry) { + perf_top__show_details(top); + return; + } + + hists__collapse_resort_threaded(&top->sym_evsel->hists); + hists__output_resort_threaded(&top->sym_evsel->hists); + hists__decay_entries_threaded(&top->sym_evsel->hists, + top->hide_user_symbols, + top->hide_kernel_symbols); + hists__output_recalc_col_len(&top->sym_evsel->hists, + top->winsize.ws_row - 3); + putchar('\n'); + hists__fprintf(&top->sym_evsel->hists, false, + top->winsize.ws_row - 4 - printed, win_width, stdout); +} + +static void prompt_integer(int *target, const char *msg) +{ + char *buf = malloc(0), *p; + size_t dummy = 0; + int tmp; + + fprintf(stdout, "\n%s: ", msg); + if (getline(&buf, &dummy, stdin) < 0) + return; + + p = strchr(buf, '\n'); + if (p) + *p = 0; + + p = buf; + while(*p) { + if (!isdigit(*p)) + goto out_free; + p++; + } + tmp = strtoul(buf, NULL, 10); + *target = tmp; +out_free: + free(buf); +} + +static void prompt_percent(int *target, const char *msg) +{ + int tmp = 0; + + prompt_integer(&tmp, msg); + if (tmp >= 0 && tmp <= 100) + *target = tmp; +} + +static void perf_top__prompt_symbol(struct perf_top *top, const char *msg) +{ + char *buf = malloc(0), *p; + struct hist_entry *syme = top->sym_filter_entry, *n, *found = NULL; + struct rb_node *next; + size_t dummy = 0; + + /* zero counters of active symbol */ + if (syme) { + __zero_source_counters(syme); + top->sym_filter_entry = NULL; + } + + fprintf(stdout, "\n%s: ", msg); + if (getline(&buf, &dummy, stdin) < 0) + goto out_free; + + p = strchr(buf, '\n'); + if (p) + *p = 0; + + next = rb_first(&top->sym_evsel->hists.entries); + while (next) { + n = rb_entry(next, struct hist_entry, rb_node); + if (n->ms.sym && !strcmp(buf, n->ms.sym->name)) { + found = n; + break; + } + next = rb_next(&n->rb_node); + } + + if (!found) { + fprintf(stderr, "Sorry, %s is not active.\n", buf); + sleep(1); + } else + perf_top__parse_source(top, found); + +out_free: + free(buf); +} + +static void perf_top__print_mapped_keys(struct perf_top *top) +{ + char *name = NULL; + + if (top->sym_filter_entry) { + struct symbol *sym = top->sym_filter_entry->ms.sym; + name = sym->name; + } + + fprintf(stdout, "\nMapped keys:\n"); + fprintf(stdout, "\t[d] display refresh delay. \t(%d)\n", top->delay_secs); + fprintf(stdout, "\t[e] display entries (lines). \t(%d)\n", top->print_entries); + + if (top->evlist->nr_entries > 1) + fprintf(stdout, "\t[E] active event counter. \t(%s)\n", perf_evsel__name(top->sym_evsel)); + + fprintf(stdout, "\t[f] profile display filter (count). \t(%d)\n", top->count_filter); + + fprintf(stdout, "\t[F] annotate display filter (percent). \t(%d%%)\n", top->sym_pcnt_filter); + fprintf(stdout, "\t[s] annotate symbol. \t(%s)\n", name?: "NULL"); + fprintf(stdout, "\t[S] stop annotation.\n"); + + fprintf(stdout, + "\t[K] hide kernel_symbols symbols. \t(%s)\n", + top->hide_kernel_symbols ? "yes" : "no"); + fprintf(stdout, + "\t[U] hide user symbols. \t(%s)\n", + top->hide_user_symbols ? "yes" : "no"); + fprintf(stdout, "\t[z] toggle sample zeroing. \t(%d)\n", top->zero ? 1 : 0); + fprintf(stdout, "\t[qQ] quit.\n"); +} + +static int perf_top__key_mapped(struct perf_top *top, int c) +{ + switch (c) { + case 'd': + case 'e': + case 'f': + case 'z': + case 'q': + case 'Q': + case 'K': + case 'U': + case 'F': + case 's': + case 'S': + return 1; + case 'E': + return top->evlist->nr_entries > 1 ? 1 : 0; + default: + break; + } + + return 0; +} + +static bool perf_top__handle_keypress(struct perf_top *top, int c) +{ + bool ret = true; + + if (!perf_top__key_mapped(top, c)) { + struct pollfd stdin_poll = { .fd = 0, .events = POLLIN }; + struct termios tc, save; + + perf_top__print_mapped_keys(top); + fprintf(stdout, "\nEnter selection, or unmapped key to continue: "); + fflush(stdout); + + tcgetattr(0, &save); + tc = save; + tc.c_lflag &= ~(ICANON | ECHO); + tc.c_cc[VMIN] = 0; + tc.c_cc[VTIME] = 0; + tcsetattr(0, TCSANOW, &tc); + + poll(&stdin_poll, 1, -1); + c = getc(stdin); + + tcsetattr(0, TCSAFLUSH, &save); + if (!perf_top__key_mapped(top, c)) + return ret; + } + + switch (c) { + case 'd': + prompt_integer(&top->delay_secs, "Enter display delay"); + if (top->delay_secs < 1) + top->delay_secs = 1; + break; + case 'e': + prompt_integer(&top->print_entries, "Enter display entries (lines)"); + if (top->print_entries == 0) { + struct sigaction act = { + .sa_sigaction = perf_top__sig_winch, + .sa_flags = SA_SIGINFO, + }; + perf_top__sig_winch(SIGWINCH, NULL, top); + sigaction(SIGWINCH, &act, NULL); + } else { + perf_top__sig_winch(SIGWINCH, NULL, top); + signal(SIGWINCH, SIG_DFL); + } + break; + case 'E': + if (top->evlist->nr_entries > 1) { + /* Select 0 as the default event: */ + int counter = 0; + + fprintf(stderr, "\nAvailable events:"); + + list_for_each_entry(top->sym_evsel, &top->evlist->entries, node) + fprintf(stderr, "\n\t%d %s", top->sym_evsel->idx, perf_evsel__name(top->sym_evsel)); + + prompt_integer(&counter, "Enter details event counter"); + + if (counter >= top->evlist->nr_entries) { + top->sym_evsel = perf_evlist__first(top->evlist); + fprintf(stderr, "Sorry, no such event, using %s.\n", perf_evsel__name(top->sym_evsel)); + sleep(1); + break; + } + list_for_each_entry(top->sym_evsel, &top->evlist->entries, node) + if (top->sym_evsel->idx == counter) + break; + } else + top->sym_evsel = perf_evlist__first(top->evlist); + break; + case 'f': + prompt_integer(&top->count_filter, "Enter display event count filter"); + break; + case 'F': + prompt_percent(&top->sym_pcnt_filter, + "Enter details display event filter (percent)"); + break; + case 'K': + top->hide_kernel_symbols = !top->hide_kernel_symbols; + break; + case 'q': + case 'Q': + printf("exiting.\n"); + if (top->dump_symtab) + perf_session__fprintf_dsos(top->session, stderr); + ret = false; + break; + case 's': + perf_top__prompt_symbol(top, "Enter details symbol"); + break; + case 'S': + if (!top->sym_filter_entry) + break; + else { + struct hist_entry *syme = top->sym_filter_entry; + + top->sym_filter_entry = NULL; + __zero_source_counters(syme); + } + break; + case 'U': + top->hide_user_symbols = !top->hide_user_symbols; + break; + case 'z': + top->zero = !top->zero; + break; + default: + break; + } + + return ret; +} + +static void perf_top__sort_new_samples(void *arg) +{ + struct perf_top *t = arg; + perf_top__reset_sample_counters(t); + + if (t->evlist->selected != NULL) + t->sym_evsel = t->evlist->selected; + + hists__collapse_resort_threaded(&t->sym_evsel->hists); + hists__output_resort_threaded(&t->sym_evsel->hists); + hists__decay_entries_threaded(&t->sym_evsel->hists, + t->hide_user_symbols, + t->hide_kernel_symbols); +} + +static void *display_thread_tui(void *arg) +{ + struct perf_evsel *pos; + struct perf_top *top = arg; + const char *help = "For a higher level overview, try: perf top --sort comm,dso"; + struct hist_browser_timer hbt = { + .timer = perf_top__sort_new_samples, + .arg = top, + .refresh = top->delay_secs, + }; + + perf_top__sort_new_samples(top); + + /* + * Initialize the uid_filter_str, in the future the TUI will allow + * Zooming in/out UIDs. For now juse use whatever the user passed + * via --uid. + */ + list_for_each_entry(pos, &top->evlist->entries, node) + pos->hists.uid_filter_str = top->record_opts.target.uid_str; + + perf_evlist__tui_browse_hists(top->evlist, help, &hbt, + &top->session->header.env); + + done = 1; + return NULL; +} + +static void *display_thread(void *arg) +{ + struct pollfd stdin_poll = { .fd = 0, .events = POLLIN }; + struct termios tc, save; + struct perf_top *top = arg; + int delay_msecs, c; + + tcgetattr(0, &save); + tc = save; + tc.c_lflag &= ~(ICANON | ECHO); + tc.c_cc[VMIN] = 0; + tc.c_cc[VTIME] = 0; + + pthread__unblock_sigwinch(); +repeat: + delay_msecs = top->delay_secs * 1000; + tcsetattr(0, TCSANOW, &tc); + /* trash return*/ + getc(stdin); + + while (!done) { + perf_top__print_sym_table(top); + /* + * Either timeout expired or we got an EINTR due to SIGWINCH, + * refresh screen in both cases. + */ + switch (poll(&stdin_poll, 1, delay_msecs)) { + case 0: + continue; + case -1: + if (errno == EINTR) + continue; + /* Fall trhu */ + default: + c = getc(stdin); + tcsetattr(0, TCSAFLUSH, &save); + + if (perf_top__handle_keypress(top, c)) + goto repeat; + done = 1; + } + } + + return NULL; +} + +/* Tag samples to be skipped. */ +static const char *skip_symbols[] = { + "intel_idle", + "default_idle", + "native_safe_halt", + "cpu_idle", + "enter_idle", + "exit_idle", + "mwait_idle", + "mwait_idle_with_hints", + "poll_idle", + "ppc64_runlatch_off", + "pseries_dedicated_idle_sleep", + NULL +}; + +static int symbol_filter(struct map *map __maybe_unused, struct symbol *sym) +{ + const char *name = sym->name; + int i; + + /* + * ppc64 uses function descriptors and appends a '.' to the + * start of every instruction address. Remove it. + */ + if (name[0] == '.') + name++; + + if (!strcmp(name, "_text") || + !strcmp(name, "_etext") || + !strcmp(name, "_sinittext") || + !strncmp("init_module", name, 11) || + !strncmp("cleanup_module", name, 14) || + strstr(name, "_text_start") || + strstr(name, "_text_end")) + return 1; + + for (i = 0; skip_symbols[i]; i++) { + if (!strcmp(skip_symbols[i], name)) { + sym->ignore = true; + break; + } + } + + return 0; +} + +static void perf_event__process_sample(struct perf_tool *tool, + const union perf_event *event, + struct perf_evsel *evsel, + struct perf_sample *sample, + struct machine *machine) +{ + struct perf_top *top = container_of(tool, struct perf_top, tool); + struct symbol *parent = NULL; + u64 ip = event->ip.ip; + struct addr_location al; + int err; + + if (!machine && perf_guest) { + static struct intlist *seen; + + if (!seen) + seen = intlist__new(NULL); + + if (!intlist__has_entry(seen, event->ip.pid)) { + pr_err("Can't find guest [%d]'s kernel information\n", + event->ip.pid); + intlist__add(seen, event->ip.pid); + } + return; + } + + if (!machine) { + pr_err("%u unprocessable samples recorded.\r", + top->session->stats.nr_unprocessable_samples++); + return; + } + + if (event->header.misc & PERF_RECORD_MISC_EXACT_IP) + top->exact_samples++; + + if (perf_event__preprocess_sample(event, machine, &al, sample, + symbol_filter) < 0 || + al.filtered) + return; + + if (!top->kptr_restrict_warned && + symbol_conf.kptr_restrict && + al.cpumode == PERF_RECORD_MISC_KERNEL) { + ui__warning( +"Kernel address maps (/proc/{kallsyms,modules}) are restricted.\n\n" +"Check /proc/sys/kernel/kptr_restrict.\n\n" +"Kernel%s samples will not be resolved.\n", + !RB_EMPTY_ROOT(&al.map->dso->symbols[MAP__FUNCTION]) ? + " modules" : ""); + if (use_browser <= 0) + sleep(5); + top->kptr_restrict_warned = true; + } + + if (al.sym == NULL) { + const char *msg = "Kernel samples will not be resolved.\n"; + /* + * As we do lazy loading of symtabs we only will know if the + * specified vmlinux file is invalid when we actually have a + * hit in kernel space and then try to load it. So if we get + * here and there are _no_ symbols in the DSO backing the + * kernel map, bail out. + * + * We may never get here, for instance, if we use -K/ + * --hide-kernel-symbols, even if the user specifies an + * invalid --vmlinux ;-) + */ + if (!top->kptr_restrict_warned && !top->vmlinux_warned && + al.map == machine->vmlinux_maps[MAP__FUNCTION] && + RB_EMPTY_ROOT(&al.map->dso->symbols[MAP__FUNCTION])) { + if (symbol_conf.vmlinux_name) { + ui__warning("The %s file can't be used.\n%s", + symbol_conf.vmlinux_name, msg); + } else { + ui__warning("A vmlinux file was not found.\n%s", + msg); + } + + if (use_browser <= 0) + sleep(5); + top->vmlinux_warned = true; + } + } + + if (al.sym == NULL || !al.sym->ignore) { + struct hist_entry *he; + + if ((sort__has_parent || symbol_conf.use_callchain) && + sample->callchain) { + err = machine__resolve_callchain(machine, evsel, + al.thread, sample, + &parent); + + if (err) + return; + } + + he = perf_evsel__add_hist_entry(evsel, &al, sample); + if (he == NULL) { + pr_err("Problem incrementing symbol period, skipping event\n"); + return; + } + + if (symbol_conf.use_callchain) { + err = callchain_append(he->callchain, &callchain_cursor, + sample->period); + if (err) + return; + } + + if (top->sort_has_symbols) + perf_top__record_precise_ip(top, he, evsel->idx, ip); + } + + return; +} + +static void perf_top__mmap_read_idx(struct perf_top *top, int idx) +{ + struct perf_sample sample; + struct perf_evsel *evsel; + struct perf_session *session = top->session; + union perf_event *event; + struct machine *machine; + u8 origin; + int ret; + + while ((event = perf_evlist__mmap_read(top->evlist, idx)) != NULL) { + ret = perf_evlist__parse_sample(top->evlist, event, &sample); + if (ret) { + pr_err("Can't parse sample, err = %d\n", ret); + continue; + } + + evsel = perf_evlist__id2evsel(session->evlist, sample.id); + assert(evsel != NULL); + + origin = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; + + if (event->header.type == PERF_RECORD_SAMPLE) + ++top->samples; + + switch (origin) { + case PERF_RECORD_MISC_USER: + ++top->us_samples; + if (top->hide_user_symbols) + continue; + machine = &session->machines.host; + break; + case PERF_RECORD_MISC_KERNEL: + ++top->kernel_samples; + if (top->hide_kernel_symbols) + continue; + machine = &session->machines.host; + break; + case PERF_RECORD_MISC_GUEST_KERNEL: + ++top->guest_kernel_samples; + machine = perf_session__find_machine(session, event->ip.pid); + break; + case PERF_RECORD_MISC_GUEST_USER: + ++top->guest_us_samples; + /* + * TODO: we don't process guest user from host side + * except simple counting. + */ + /* Fall thru */ + default: + continue; + } + + + if (event->header.type == PERF_RECORD_SAMPLE) { + perf_event__process_sample(&top->tool, event, evsel, + &sample, machine); + } else if (event->header.type < PERF_RECORD_MAX) { + hists__inc_nr_events(&evsel->hists, event->header.type); + machine__process_event(machine, event); + } else + ++session->stats.nr_unknown_events; + } +} + +static void perf_top__mmap_read(struct perf_top *top) +{ + int i; + + for (i = 0; i < top->evlist->nr_mmaps; i++) + perf_top__mmap_read_idx(top, i); +} + +static int perf_top__start_counters(struct perf_top *top) +{ + char msg[512]; + struct perf_evsel *counter; + struct perf_evlist *evlist = top->evlist; + struct perf_record_opts *opts = &top->record_opts; + + perf_evlist__config(evlist, opts); + + list_for_each_entry(counter, &evlist->entries, node) { +try_again: + if (perf_evsel__open(counter, top->evlist->cpus, + top->evlist->threads) < 0) { + if (perf_evsel__fallback(counter, errno, msg, sizeof(msg))) { + if (verbose) + ui__warning("%s\n", msg); + goto try_again; + } + + perf_evsel__open_strerror(counter, &opts->target, + errno, msg, sizeof(msg)); + ui__error("%s\n", msg); + goto out_err; + } + } + + if (perf_evlist__mmap(evlist, opts->mmap_pages, false) < 0) { + ui__error("Failed to mmap with %d (%s)\n", + errno, strerror(errno)); + goto out_err; + } + + return 0; + +out_err: + return -1; +} + +static int perf_top__setup_sample_type(struct perf_top *top) +{ + if (!top->sort_has_symbols) { + if (symbol_conf.use_callchain) { + ui__error("Selected -g but \"sym\" not present in --sort/-s."); + return -EINVAL; + } + } else if (callchain_param.mode != CHAIN_NONE) { + if (callchain_register_param(&callchain_param) < 0) { + ui__error("Can't register callchain params.\n"); + return -EINVAL; + } + } + + return 0; +} + +static int __cmd_top(struct perf_top *top) +{ + struct perf_record_opts *opts = &top->record_opts; + pthread_t thread; + int ret; + /* + * FIXME: perf_session__new should allow passing a O_MMAP, so that all this + * mmap reading, etc is encapsulated in it. Use O_WRONLY for now. + */ + top->session = perf_session__new(NULL, O_WRONLY, false, false, NULL); + if (top->session == NULL) + return -ENOMEM; + + ret = perf_top__setup_sample_type(top); + if (ret) + goto out_delete; + + if (perf_target__has_task(&opts->target)) + perf_event__synthesize_thread_map(&top->tool, top->evlist->threads, + perf_event__process, + &top->session->machines.host); + else + perf_event__synthesize_threads(&top->tool, perf_event__process, + &top->session->machines.host); + + ret = perf_top__start_counters(top); + if (ret) + goto out_delete; + + top->session->evlist = top->evlist; + perf_session__set_id_hdr_size(top->session); + + /* + * When perf is starting the traced process, all the events (apart from + * group members) have enable_on_exec=1 set, so don't spoil it by + * prematurely enabling them. + * + * XXX 'top' still doesn't start workloads like record, trace, but should, + * so leave the check here. + */ + if (!perf_target__none(&opts->target)) + perf_evlist__enable(top->evlist); + + /* Wait for a minimal set of events before starting the snapshot */ + poll(top->evlist->pollfd, top->evlist->nr_fds, 100); + + perf_top__mmap_read(top); + + ret = -1; + if (pthread_create(&thread, NULL, (use_browser > 0 ? display_thread_tui : + display_thread), top)) { + ui__error("Could not create display thread.\n"); + goto out_delete; + } + + if (top->realtime_prio) { + struct sched_param param; + + param.sched_priority = top->realtime_prio; + if (sched_setscheduler(0, SCHED_FIFO, ¶m)) { + ui__error("Could not set realtime priority.\n"); + goto out_delete; + } + } + + while (!done) { + u64 hits = top->samples; + + perf_top__mmap_read(top); + + if (hits == top->samples) + ret = poll(top->evlist->pollfd, top->evlist->nr_fds, 100); + } + + ret = 0; +out_delete: + perf_session__delete(top->session); + top->session = NULL; + + return ret; +} + +static int +parse_callchain_opt(const struct option *opt, const char *arg, int unset) +{ + /* + * --no-call-graph + */ + if (unset) + return 0; + + symbol_conf.use_callchain = true; + + return record_parse_callchain_opt(opt, arg, unset); +} + +int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused) +{ + int status; + char errbuf[BUFSIZ]; + struct perf_top top = { + .count_filter = 5, + .delay_secs = 2, + .record_opts = { + .mmap_pages = UINT_MAX, + .user_freq = UINT_MAX, + .user_interval = ULLONG_MAX, + .freq = 4000, /* 4 KHz */ + .target = { + .uses_mmap = true, + }, + }, + .sym_pcnt_filter = 5, + }; + struct perf_record_opts *opts = &top.record_opts; + struct perf_target *target = &opts->target; + const struct option options[] = { + OPT_CALLBACK('e', "event", &top.evlist, "event", + "event selector. use 'perf list' to list available events", + parse_events_option), + OPT_U64('c', "count", &opts->user_interval, "event period to sample"), + OPT_STRING('p', "pid", &target->pid, "pid", + "profile events on existing process id"), + OPT_STRING('t', "tid", &target->tid, "tid", + "profile events on existing thread id"), + OPT_BOOLEAN('a', "all-cpus", &target->system_wide, + "system-wide collection from all CPUs"), + OPT_STRING('C', "cpu", &target->cpu_list, "cpu", + "list of cpus to monitor"), + OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name, + "file", "vmlinux pathname"), + OPT_BOOLEAN('K', "hide_kernel_symbols", &top.hide_kernel_symbols, + "hide kernel symbols"), + OPT_UINTEGER('m', "mmap-pages", &opts->mmap_pages, + "number of mmap data pages"), + OPT_INTEGER('r', "realtime", &top.realtime_prio, + "collect data with this RT SCHED_FIFO priority"), + OPT_INTEGER('d', "delay", &top.delay_secs, + "number of seconds to delay between refreshes"), + OPT_BOOLEAN('D', "dump-symtab", &top.dump_symtab, + "dump the symbol table used for profiling"), + OPT_INTEGER('f', "count-filter", &top.count_filter, + "only display functions with more events than this"), + OPT_BOOLEAN('g', "group", &opts->group, + "put the counters into a counter group"), + OPT_BOOLEAN('i', "no-inherit", &opts->no_inherit, + "child tasks do not inherit counters"), + OPT_STRING(0, "sym-annotate", &top.sym_filter, "symbol name", + "symbol to annotate"), + OPT_BOOLEAN('z', "zero", &top.zero, "zero history across updates"), + OPT_UINTEGER('F', "freq", &opts->user_freq, "profile at this frequency"), + OPT_INTEGER('E', "entries", &top.print_entries, + "display this many functions"), + OPT_BOOLEAN('U', "hide_user_symbols", &top.hide_user_symbols, + "hide user symbols"), + OPT_BOOLEAN(0, "tui", &top.use_tui, "Use the TUI interface"), + OPT_BOOLEAN(0, "stdio", &top.use_stdio, "Use the stdio interface"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show counter open errors, etc)"), + OPT_STRING('s', "sort", &sort_order, "key[,key2...]", + "sort by key(s): pid, comm, dso, symbol, parent"), + OPT_BOOLEAN('n', "show-nr-samples", &symbol_conf.show_nr_samples, + "Show a column with the number of samples"), + OPT_CALLBACK_DEFAULT('G', "call-graph", &top.record_opts, + "mode[,dump_size]", record_callchain_help, + &parse_callchain_opt, "fp"), + OPT_BOOLEAN(0, "show-total-period", &symbol_conf.show_total_period, + "Show a column with the sum of periods"), + OPT_STRING(0, "dsos", &symbol_conf.dso_list_str, "dso[,dso...]", + "only consider symbols in these dsos"), + OPT_STRING(0, "comms", &symbol_conf.comm_list_str, "comm[,comm...]", + "only consider symbols in these comms"), + OPT_STRING(0, "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]", + "only consider these symbols"), + OPT_BOOLEAN(0, "source", &symbol_conf.annotate_src, + "Interleave source code with assembly code (default)"), + OPT_BOOLEAN(0, "asm-raw", &symbol_conf.annotate_asm_raw, + "Display raw encoding of assembly instructions (default)"), + OPT_STRING('M', "disassembler-style", &disassembler_style, "disassembler style", + "Specify disassembler style (e.g. -M intel for intel syntax)"), + OPT_STRING('u', "uid", &target->uid_str, "user", "user to profile"), + OPT_END() + }; + const char * const top_usage[] = { + "perf top [<options>]", + NULL + }; + + top.evlist = perf_evlist__new(NULL, NULL); + if (top.evlist == NULL) + return -ENOMEM; + + symbol_conf.exclude_other = false; + + argc = parse_options(argc, argv, options, top_usage, 0); + if (argc) + usage_with_options(top_usage, options); + + if (sort_order == default_sort_order) + sort_order = "dso,symbol"; + + if (setup_sorting() < 0) + usage_with_options(top_usage, options); + + if (top.use_stdio) + use_browser = 0; + else if (top.use_tui) + use_browser = 1; + + setup_browser(false); + + status = perf_target__validate(target); + if (status) { + perf_target__strerror(target, status, errbuf, BUFSIZ); + ui__warning("%s", errbuf); + } + + status = perf_target__parse_uid(target); + if (status) { + int saved_errno = errno; + + perf_target__strerror(target, status, errbuf, BUFSIZ); + ui__error("%s", errbuf); + + status = -saved_errno; + goto out_delete_evlist; + } + + if (perf_target__none(target)) + target->system_wide = true; + + if (perf_evlist__create_maps(top.evlist, target) < 0) + usage_with_options(top_usage, options); + + if (!top.evlist->nr_entries && + perf_evlist__add_default(top.evlist) < 0) { + ui__error("Not enough memory for event selector list\n"); + goto out_delete_maps; + } + + symbol_conf.nr_events = top.evlist->nr_entries; + + if (top.delay_secs < 1) + top.delay_secs = 1; + + if (opts->user_interval != ULLONG_MAX) + opts->default_interval = opts->user_interval; + if (opts->user_freq != UINT_MAX) + opts->freq = opts->user_freq; + + /* + * User specified count overrides default frequency. + */ + if (opts->default_interval) + opts->freq = 0; + else if (opts->freq) { + opts->default_interval = opts->freq; + } else { + ui__error("frequency and count are zero, aborting\n"); + status = -EINVAL; + goto out_delete_maps; + } + + top.sym_evsel = perf_evlist__first(top.evlist); + + symbol_conf.priv_size = sizeof(struct annotation); + + symbol_conf.try_vmlinux_path = (symbol_conf.vmlinux_name == NULL); + if (symbol__init() < 0) + return -1; + + sort_entry__setup_elide(&sort_dso, symbol_conf.dso_list, "dso", stdout); + sort_entry__setup_elide(&sort_comm, symbol_conf.comm_list, "comm", stdout); + sort_entry__setup_elide(&sort_sym, symbol_conf.sym_list, "symbol", stdout); + + /* + * Avoid annotation data structures overhead when symbols aren't on the + * sort list. + */ + top.sort_has_symbols = sort_sym.list.next != NULL; + + get_term_dimensions(&top.winsize); + if (top.print_entries == 0) { + struct sigaction act = { + .sa_sigaction = perf_top__sig_winch, + .sa_flags = SA_SIGINFO, + }; + perf_top__update_print_entries(&top); + sigaction(SIGWINCH, &act, NULL); + } + + status = __cmd_top(&top); + +out_delete_maps: + perf_evlist__delete_maps(top.evlist); +out_delete_evlist: + perf_evlist__delete(top.evlist); + + return status; +} diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c new file mode 100644 index 00000000..d222d7fc --- /dev/null +++ b/tools/perf/builtin-trace.c @@ -0,0 +1,685 @@ +#include "builtin.h" +#include "util/color.h" +#include "util/evlist.h" +#include "util/machine.h" +#include "util/thread.h" +#include "util/parse-options.h" +#include "util/thread_map.h" +#include "event-parse.h" + +#include <libaudit.h> +#include <stdlib.h> + +static struct syscall_fmt { + const char *name; + const char *alias; + bool errmsg; + bool timeout; +} syscall_fmts[] = { + { .name = "access", .errmsg = true, }, + { .name = "arch_prctl", .errmsg = true, .alias = "prctl", }, + { .name = "fstat", .errmsg = true, .alias = "newfstat", }, + { .name = "fstatat", .errmsg = true, .alias = "newfstatat", }, + { .name = "futex", .errmsg = true, }, + { .name = "open", .errmsg = true, }, + { .name = "poll", .errmsg = true, .timeout = true, }, + { .name = "ppoll", .errmsg = true, .timeout = true, }, + { .name = "read", .errmsg = true, }, + { .name = "recvfrom", .errmsg = true, }, + { .name = "select", .errmsg = true, .timeout = true, }, + { .name = "socket", .errmsg = true, }, + { .name = "stat", .errmsg = true, .alias = "newstat", }, +}; + +static int syscall_fmt__cmp(const void *name, const void *fmtp) +{ + const struct syscall_fmt *fmt = fmtp; + return strcmp(name, fmt->name); +} + +static struct syscall_fmt *syscall_fmt__find(const char *name) +{ + const int nmemb = ARRAY_SIZE(syscall_fmts); + return bsearch(name, syscall_fmts, nmemb, sizeof(struct syscall_fmt), syscall_fmt__cmp); +} + +struct syscall { + struct event_format *tp_format; + const char *name; + struct syscall_fmt *fmt; +}; + +static size_t fprintf_duration(unsigned long t, FILE *fp) +{ + double duration = (double)t / NSEC_PER_MSEC; + size_t printed = fprintf(fp, "("); + + if (duration >= 1.0) + printed += color_fprintf(fp, PERF_COLOR_RED, "%6.3f ms", duration); + else if (duration >= 0.01) + printed += color_fprintf(fp, PERF_COLOR_YELLOW, "%6.3f ms", duration); + else + printed += color_fprintf(fp, PERF_COLOR_NORMAL, "%6.3f ms", duration); + return printed + fprintf(stdout, "): "); +} + +struct thread_trace { + u64 entry_time; + u64 exit_time; + bool entry_pending; + unsigned long nr_events; + char *entry_str; + double runtime_ms; +}; + +static struct thread_trace *thread_trace__new(void) +{ + return zalloc(sizeof(struct thread_trace)); +} + +static struct thread_trace *thread__trace(struct thread *thread) +{ + struct thread_trace *ttrace; + + if (thread == NULL) + goto fail; + + if (thread->priv == NULL) + thread->priv = thread_trace__new(); + + if (thread->priv == NULL) + goto fail; + + ttrace = thread->priv; + ++ttrace->nr_events; + + return ttrace; +fail: + color_fprintf(stdout, PERF_COLOR_RED, + "WARNING: not enough memory, dropping samples!\n"); + return NULL; +} + +struct trace { + int audit_machine; + struct { + int max; + struct syscall *table; + } syscalls; + struct perf_record_opts opts; + struct machine host; + u64 base_time; + unsigned long nr_events; + bool sched; + bool multiple_threads; + double duration_filter; + double runtime_ms; +}; + +static bool trace__filter_duration(struct trace *trace, double t) +{ + return t < (trace->duration_filter * NSEC_PER_MSEC); +} + +static size_t trace__fprintf_tstamp(struct trace *trace, u64 tstamp, FILE *fp) +{ + double ts = (double)(tstamp - trace->base_time) / NSEC_PER_MSEC; + + return fprintf(fp, "%10.3f ", ts); +} + +static bool done = false; + +static void sig_handler(int sig __maybe_unused) +{ + done = true; +} + +static size_t trace__fprintf_entry_head(struct trace *trace, struct thread *thread, + u64 duration, u64 tstamp, FILE *fp) +{ + size_t printed = trace__fprintf_tstamp(trace, tstamp, fp); + printed += fprintf_duration(duration, fp); + + if (trace->multiple_threads) + printed += fprintf(fp, "%d ", thread->pid); + + return printed; +} + +static int trace__process_event(struct machine *machine, union perf_event *event) +{ + int ret = 0; + + switch (event->header.type) { + case PERF_RECORD_LOST: + color_fprintf(stdout, PERF_COLOR_RED, + "LOST %" PRIu64 " events!\n", event->lost.lost); + ret = machine__process_lost_event(machine, event); + default: + ret = machine__process_event(machine, event); + break; + } + + return ret; +} + +static int trace__tool_process(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine) +{ + return trace__process_event(machine, event); +} + +static int trace__symbols_init(struct trace *trace, struct perf_evlist *evlist) +{ + int err = symbol__init(); + + if (err) + return err; + + machine__init(&trace->host, "", HOST_KERNEL_ID); + machine__create_kernel_maps(&trace->host); + + if (perf_target__has_task(&trace->opts.target)) { + err = perf_event__synthesize_thread_map(NULL, evlist->threads, + trace__tool_process, + &trace->host); + } else { + err = perf_event__synthesize_threads(NULL, trace__tool_process, + &trace->host); + } + + if (err) + symbol__exit(); + + return err; +} + +static int trace__read_syscall_info(struct trace *trace, int id) +{ + char tp_name[128]; + struct syscall *sc; + const char *name = audit_syscall_to_name(id, trace->audit_machine); + + if (name == NULL) + return -1; + + if (id > trace->syscalls.max) { + struct syscall *nsyscalls = realloc(trace->syscalls.table, (id + 1) * sizeof(*sc)); + + if (nsyscalls == NULL) + return -1; + + if (trace->syscalls.max != -1) { + memset(nsyscalls + trace->syscalls.max + 1, 0, + (id - trace->syscalls.max) * sizeof(*sc)); + } else { + memset(nsyscalls, 0, (id + 1) * sizeof(*sc)); + } + + trace->syscalls.table = nsyscalls; + trace->syscalls.max = id; + } + + sc = trace->syscalls.table + id; + sc->name = name; + sc->fmt = syscall_fmt__find(sc->name); + + snprintf(tp_name, sizeof(tp_name), "sys_enter_%s", sc->name); + sc->tp_format = event_format__new("syscalls", tp_name); + + if (sc->tp_format == NULL && sc->fmt && sc->fmt->alias) { + snprintf(tp_name, sizeof(tp_name), "sys_enter_%s", sc->fmt->alias); + sc->tp_format = event_format__new("syscalls", tp_name); + } + + return sc->tp_format != NULL ? 0 : -1; +} + +static size_t syscall__scnprintf_args(struct syscall *sc, char *bf, size_t size, + unsigned long *args) +{ + int i = 0; + size_t printed = 0; + + if (sc->tp_format != NULL) { + struct format_field *field; + + for (field = sc->tp_format->format.fields->next; field; field = field->next) { + printed += scnprintf(bf + printed, size - printed, + "%s%s: %ld", printed ? ", " : "", + field->name, args[i++]); + } + } else { + while (i < 6) { + printed += scnprintf(bf + printed, size - printed, + "%sarg%d: %ld", + printed ? ", " : "", i, args[i]); + ++i; + } + } + + return printed; +} + +typedef int (*tracepoint_handler)(struct trace *trace, struct perf_evsel *evsel, + struct perf_sample *sample); + +static struct syscall *trace__syscall_info(struct trace *trace, + struct perf_evsel *evsel, + struct perf_sample *sample) +{ + int id = perf_evsel__intval(evsel, sample, "id"); + + if (id < 0) { + printf("Invalid syscall %d id, skipping...\n", id); + return NULL; + } + + if ((id > trace->syscalls.max || trace->syscalls.table[id].name == NULL) && + trace__read_syscall_info(trace, id)) + goto out_cant_read; + + if ((id > trace->syscalls.max || trace->syscalls.table[id].name == NULL)) + goto out_cant_read; + + return &trace->syscalls.table[id]; + +out_cant_read: + printf("Problems reading syscall %d", id); + if (id <= trace->syscalls.max && trace->syscalls.table[id].name != NULL) + printf("(%s)", trace->syscalls.table[id].name); + puts(" information"); + return NULL; +} + +static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel, + struct perf_sample *sample) +{ + char *msg; + void *args; + size_t printed = 0; + struct thread *thread = machine__findnew_thread(&trace->host, sample->tid); + struct syscall *sc = trace__syscall_info(trace, evsel, sample); + struct thread_trace *ttrace = thread__trace(thread); + + if (ttrace == NULL || sc == NULL) + return -1; + + args = perf_evsel__rawptr(evsel, sample, "args"); + if (args == NULL) { + printf("Problems reading syscall arguments\n"); + return -1; + } + + ttrace = thread->priv; + + if (ttrace->entry_str == NULL) { + ttrace->entry_str = malloc(1024); + if (!ttrace->entry_str) + return -1; + } + + ttrace->entry_time = sample->time; + msg = ttrace->entry_str; + printed += scnprintf(msg + printed, 1024 - printed, "%s(", sc->name); + + printed += syscall__scnprintf_args(sc, msg + printed, 1024 - printed, args); + + if (!strcmp(sc->name, "exit_group") || !strcmp(sc->name, "exit")) { + if (!trace->duration_filter) { + trace__fprintf_entry_head(trace, thread, 1, sample->time, stdout); + printf("%-70s\n", ttrace->entry_str); + } + } else + ttrace->entry_pending = true; + + return 0; +} + +static int trace__sys_exit(struct trace *trace, struct perf_evsel *evsel, + struct perf_sample *sample) +{ + int ret; + u64 duration = 0; + struct thread *thread = machine__findnew_thread(&trace->host, sample->tid); + struct thread_trace *ttrace = thread__trace(thread); + struct syscall *sc = trace__syscall_info(trace, evsel, sample); + + if (ttrace == NULL || sc == NULL) + return -1; + + ret = perf_evsel__intval(evsel, sample, "ret"); + + ttrace = thread->priv; + + ttrace->exit_time = sample->time; + + if (ttrace->entry_time) { + duration = sample->time - ttrace->entry_time; + if (trace__filter_duration(trace, duration)) + goto out; + } else if (trace->duration_filter) + goto out; + + trace__fprintf_entry_head(trace, thread, duration, sample->time, stdout); + + if (ttrace->entry_pending) { + printf("%-70s", ttrace->entry_str); + } else { + printf(" ... ["); + color_fprintf(stdout, PERF_COLOR_YELLOW, "continued"); + printf("]: %s()", sc->name); + } + + if (ret < 0 && sc->fmt && sc->fmt->errmsg) { + char bf[256]; + const char *emsg = strerror_r(-ret, bf, sizeof(bf)), + *e = audit_errno_to_name(-ret); + + printf(") = -1 %s %s", e, emsg); + } else if (ret == 0 && sc->fmt && sc->fmt->timeout) + printf(") = 0 Timeout"); + else + printf(") = %d", ret); + + putchar('\n'); +out: + ttrace->entry_pending = false; + + return 0; +} + +static int trace__sched_stat_runtime(struct trace *trace, struct perf_evsel *evsel, + struct perf_sample *sample) +{ + u64 runtime = perf_evsel__intval(evsel, sample, "runtime"); + double runtime_ms = (double)runtime / NSEC_PER_MSEC; + struct thread *thread = machine__findnew_thread(&trace->host, sample->tid); + struct thread_trace *ttrace = thread__trace(thread); + + if (ttrace == NULL) + goto out_dump; + + ttrace->runtime_ms += runtime_ms; + trace->runtime_ms += runtime_ms; + return 0; + +out_dump: + printf("%s: comm=%s,pid=%u,runtime=%" PRIu64 ",vruntime=%" PRIu64 ")\n", + evsel->name, + perf_evsel__strval(evsel, sample, "comm"), + (pid_t)perf_evsel__intval(evsel, sample, "pid"), + runtime, + perf_evsel__intval(evsel, sample, "vruntime")); + return 0; +} + +static int trace__run(struct trace *trace, int argc, const char **argv) +{ + struct perf_evlist *evlist = perf_evlist__new(NULL, NULL); + struct perf_evsel *evsel; + int err = -1, i; + unsigned long before; + const bool forks = argc > 0; + + if (evlist == NULL) { + printf("Not enough memory to run!\n"); + goto out; + } + + if (perf_evlist__add_newtp(evlist, "raw_syscalls", "sys_enter", trace__sys_enter) || + perf_evlist__add_newtp(evlist, "raw_syscalls", "sys_exit", trace__sys_exit)) { + printf("Couldn't read the raw_syscalls tracepoints information!\n"); + goto out_delete_evlist; + } + + if (trace->sched && + perf_evlist__add_newtp(evlist, "sched", "sched_stat_runtime", + trace__sched_stat_runtime)) { + printf("Couldn't read the sched_stat_runtime tracepoint information!\n"); + goto out_delete_evlist; + } + + err = perf_evlist__create_maps(evlist, &trace->opts.target); + if (err < 0) { + printf("Problems parsing the target to trace, check your options!\n"); + goto out_delete_evlist; + } + + err = trace__symbols_init(trace, evlist); + if (err < 0) { + printf("Problems initializing symbol libraries!\n"); + goto out_delete_evlist; + } + + perf_evlist__config(evlist, &trace->opts); + + signal(SIGCHLD, sig_handler); + signal(SIGINT, sig_handler); + + if (forks) { + err = perf_evlist__prepare_workload(evlist, &trace->opts, argv); + if (err < 0) { + printf("Couldn't run the workload!\n"); + goto out_delete_evlist; + } + } + + err = perf_evlist__open(evlist); + if (err < 0) { + printf("Couldn't create the events: %s\n", strerror(errno)); + goto out_delete_evlist; + } + + err = perf_evlist__mmap(evlist, UINT_MAX, false); + if (err < 0) { + printf("Couldn't mmap the events: %s\n", strerror(errno)); + goto out_delete_evlist; + } + + perf_evlist__enable(evlist); + + if (forks) + perf_evlist__start_workload(evlist); + + trace->multiple_threads = evlist->threads->map[0] == -1 || evlist->threads->nr > 1; +again: + before = trace->nr_events; + + for (i = 0; i < evlist->nr_mmaps; i++) { + union perf_event *event; + + while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) { + const u32 type = event->header.type; + tracepoint_handler handler; + struct perf_sample sample; + + ++trace->nr_events; + + err = perf_evlist__parse_sample(evlist, event, &sample); + if (err) { + printf("Can't parse sample, err = %d, skipping...\n", err); + continue; + } + + if (trace->base_time == 0) + trace->base_time = sample.time; + + if (type != PERF_RECORD_SAMPLE) { + trace__process_event(&trace->host, event); + continue; + } + + evsel = perf_evlist__id2evsel(evlist, sample.id); + if (evsel == NULL) { + printf("Unknown tp ID %" PRIu64 ", skipping...\n", sample.id); + continue; + } + + if (sample.raw_data == NULL) { + printf("%s sample with no payload for tid: %d, cpu %d, raw_size=%d, skipping...\n", + perf_evsel__name(evsel), sample.tid, + sample.cpu, sample.raw_size); + continue; + } + + if (sample.raw_data == NULL) { + printf("%s sample with no payload for tid: %d, cpu %d, raw_size=%d, skipping...\n", + perf_evsel__name(evsel), sample.tid, + sample.cpu, sample.raw_size); + continue; + } + + handler = evsel->handler.func; + handler(trace, evsel, &sample); + } + } + + if (trace->nr_events == before) { + if (done) + goto out_delete_evlist; + + poll(evlist->pollfd, evlist->nr_fds, -1); + } + + if (done) + perf_evlist__disable(evlist); + + goto again; + +out_delete_evlist: + perf_evlist__delete(evlist); +out: + return err; +} + +static size_t trace__fprintf_threads_header(FILE *fp) +{ + size_t printed; + + printed = fprintf(fp, "\n _____________________________________________________________________\n"); + printed += fprintf(fp," __) Summary of events (__\n\n"); + printed += fprintf(fp," [ task - pid ] [ events ] [ ratio ] [ runtime ]\n"); + printed += fprintf(fp," _____________________________________________________________________\n\n"); + + return printed; +} + +static size_t trace__fprintf_thread_summary(struct trace *trace, FILE *fp) +{ + size_t printed = trace__fprintf_threads_header(fp); + struct rb_node *nd; + + for (nd = rb_first(&trace->host.threads); nd; nd = rb_next(nd)) { + struct thread *thread = rb_entry(nd, struct thread, rb_node); + struct thread_trace *ttrace = thread->priv; + const char *color; + double ratio; + + if (ttrace == NULL) + continue; + + ratio = (double)ttrace->nr_events / trace->nr_events * 100.0; + + color = PERF_COLOR_NORMAL; + if (ratio > 50.0) + color = PERF_COLOR_RED; + else if (ratio > 25.0) + color = PERF_COLOR_GREEN; + else if (ratio > 5.0) + color = PERF_COLOR_YELLOW; + + printed += color_fprintf(fp, color, "%20s", thread->comm); + printed += fprintf(fp, " - %-5d :%11lu [", thread->pid, ttrace->nr_events); + printed += color_fprintf(fp, color, "%5.1f%%", ratio); + printed += fprintf(fp, " ] %10.3f ms\n", ttrace->runtime_ms); + } + + return printed; +} + +static int trace__set_duration(const struct option *opt, const char *str, + int unset __maybe_unused) +{ + struct trace *trace = opt->value; + + trace->duration_filter = atof(str); + return 0; +} + +int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused) +{ + const char * const trace_usage[] = { + "perf trace [<options>] [<command>]", + "perf trace [<options>] -- <command> [<options>]", + NULL + }; + struct trace trace = { + .audit_machine = audit_detect_machine(), + .syscalls = { + . max = -1, + }, + .opts = { + .target = { + .uid = UINT_MAX, + .uses_mmap = true, + }, + .user_freq = UINT_MAX, + .user_interval = ULLONG_MAX, + .no_delay = true, + .mmap_pages = 1024, + }, + }; + const struct option trace_options[] = { + OPT_STRING('p', "pid", &trace.opts.target.pid, "pid", + "trace events on existing process id"), + OPT_STRING(0, "tid", &trace.opts.target.tid, "tid", + "trace events on existing thread id"), + OPT_BOOLEAN(0, "all-cpus", &trace.opts.target.system_wide, + "system-wide collection from all CPUs"), + OPT_STRING(0, "cpu", &trace.opts.target.cpu_list, "cpu", + "list of cpus to monitor"), + OPT_BOOLEAN(0, "no-inherit", &trace.opts.no_inherit, + "child tasks do not inherit counters"), + OPT_UINTEGER(0, "mmap-pages", &trace.opts.mmap_pages, + "number of mmap data pages"), + OPT_STRING(0, "uid", &trace.opts.target.uid_str, "user", + "user to profile"), + OPT_CALLBACK(0, "duration", &trace, "float", + "show only events with duration > N.M ms", + trace__set_duration), + OPT_BOOLEAN(0, "sched", &trace.sched, "show blocking scheduler events"), + OPT_END() + }; + int err; + char bf[BUFSIZ]; + + argc = parse_options(argc, argv, trace_options, trace_usage, 0); + + err = perf_target__validate(&trace.opts.target); + if (err) { + perf_target__strerror(&trace.opts.target, err, bf, sizeof(bf)); + printf("%s", bf); + return err; + } + + err = perf_target__parse_uid(&trace.opts.target); + if (err) { + perf_target__strerror(&trace.opts.target, err, bf, sizeof(bf)); + printf("%s", bf); + return err; + } + + if (!argc && perf_target__none(&trace.opts.target)) + trace.opts.target.system_wide = true; + + err = trace__run(&trace, argc, argv); + + if (trace.sched && !err) + trace__fprintf_thread_summary(&trace, stdout); + + return err; +} diff --git a/tools/perf/builtin.h b/tools/perf/builtin.h new file mode 100644 index 00000000..08143bd8 --- /dev/null +++ b/tools/perf/builtin.h @@ -0,0 +1,41 @@ +#ifndef BUILTIN_H +#define BUILTIN_H + +#include "util/util.h" +#include "util/strbuf.h" + +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_bench(int argc, const char **argv, const char *prefix); +extern int cmd_buildid_cache(int argc, const char **argv, const char *prefix); +extern int cmd_buildid_list(int argc, const char **argv, const char *prefix); +extern int cmd_diff(int argc, const char **argv, const char *prefix); +extern int cmd_evlist(int argc, const char **argv, const char *prefix); +extern int cmd_help(int argc, const char **argv, const char *prefix); +extern int cmd_sched(int argc, const char **argv, const char *prefix); +extern int cmd_list(int argc, const char **argv, const char *prefix); +extern int cmd_record(int argc, const char **argv, const char *prefix); +extern int cmd_report(int argc, const char **argv, const char *prefix); +extern int cmd_stat(int argc, const char **argv, const char *prefix); +extern int cmd_timechart(int argc, const char **argv, const char *prefix); +extern int cmd_top(int argc, const char **argv, const char *prefix); +extern int cmd_script(int argc, const char **argv, const char *prefix); +extern int cmd_version(int argc, const char **argv, const char *prefix); +extern int cmd_probe(int argc, const char **argv, const char *prefix); +extern int cmd_kmem(int argc, const char **argv, const char *prefix); +extern int cmd_lock(int argc, const char **argv, const char *prefix); +extern int cmd_kvm(int argc, const char **argv, const char *prefix); +extern int cmd_test(int argc, const char **argv, const char *prefix); +extern int cmd_trace(int argc, const char **argv, const char *prefix); +extern int cmd_inject(int argc, const char **argv, const char *prefix); + +extern int find_scripts(char **scripts_array, char **scripts_path_array); +#endif diff --git a/tools/perf/command-list.txt b/tools/perf/command-list.txt new file mode 100644 index 00000000..3e86bbd8 --- /dev/null +++ b/tools/perf/command-list.txt @@ -0,0 +1,26 @@ +# +# List of known perf commands. +# command name category [deprecated] [common] +# +perf-annotate mainporcelain common +perf-archive mainporcelain common +perf-bench mainporcelain common +perf-buildid-cache mainporcelain common +perf-buildid-list mainporcelain common +perf-diff mainporcelain common +perf-evlist mainporcelain common +perf-inject mainporcelain common +perf-list mainporcelain common +perf-sched mainporcelain common +perf-record mainporcelain common +perf-report mainporcelain common +perf-stat mainporcelain common +perf-timechart mainporcelain common +perf-top mainporcelain common +perf-trace mainporcelain common +perf-script mainporcelain common +perf-probe mainporcelain full +perf-kmem mainporcelain common +perf-lock mainporcelain common +perf-kvm mainporcelain common +perf-test mainporcelain common diff --git a/tools/perf/compat-android.h b/tools/perf/compat-android.h new file mode 100644 index 00000000..b76adae2 --- /dev/null +++ b/tools/perf/compat-android.h @@ -0,0 +1,133 @@ +/* Android compatibility header + * Provides missing bits in Bionic on Android, ignored + * on regular Linux. + * + * Written by Bernhard.Rosenkranzer@linaro.org + * + * Released into the public domain. Do with this file + * whatever you want. + */ +#ifdef ANDROID +/* Bionic has its own idea about ALIGN, and kills other definitions. + * Done outside the multiple-inclusion wrapper to make sure we + * can override Bionic's ALIGN by simply including compat-android.h + * again after including Bionic headers. + */ +#undef ALIGN +#undef __ALIGN_MASK +#define ALIGN(x,a) __ALIGN_MASK(x,(typeof(x))(a)-1) +#define __ALIGN_MASK(x,mask) (((x)+(mask))&~(mask)) + +#ifndef _COMPAT_ANDROID_H_ +#define _COMPAT_ANDROID_H_ 1 +/* Stuff Bionic assumes to be present, but that doesn't exist + * anymore after the uabi kernel header reorg + */ +#include <stdint.h> +#include <stdbool.h> +typedef unsigned short __kernel_nlink_t; +typedef intptr_t phys_addr_t; +#include <linux/types.h> +typedef uint32_t u32; +typedef uint64_t u64; +#ifndef CONFIG_DRAM_BASEUL +#ifdef CONFIG_DRAM_BASE +#define CONFIG_DRAM_BASEUL UL(CONFIG_DRAM_BASE) +#else +#define CONFIG_DRAM_BASEUL 0 +#endif +#endif +#define __deprecated + +#include <linux/bitops.h> +#undef BITS_PER_LONG /* Something seems to define this incorrectly */ +#define BITS_PER_LONG _BITSIZE + +#include <stdio.h> +#include <signal.h> +#include <asm/page.h> /* for PAGE_SIZE */ +#include <asm/termios.h> /* for winsize */ + +#ifndef __WORDSIZE +#define __WORDSIZE _BITSIZE +#endif + +#ifndef roundup +#define roundup(x, y) ((((x)+((y)-1))/(y))*(y)) +#endif + +#ifndef __force +#define __force +#endif + +#ifndef __le32 +#define __le32 uint32_t +#endif + +#ifndef FD_SET +#define FD_SET(fd, fdsetp) (((fd_set *)(fdsetp))->fds_bits[(fd) >> 5] |= (1<<((fd) & 31))) +#define FD_ZERO(fdsetp) (memset (fdsetp, 0, sizeof (*(fd_set *)(fdsetp)))) +#endif + +/* Assorted functions that are missing from Bionic */ +/* Android prior to 4.2 lacks psignal(). + * What we're doing here is fairly evil - but necessary since + * Bionic doesn't export any version identifier or the likes. + * We do know that 4.2 is the version introducing psignal() and + * also KLOG_CONSOLE_OFF -- completely unrelated, but something + * we can check for... + */ +#include <sys/klog.h> +#ifndef KLOG_CONSOLE_OFF +static void psignal(int sig, const char *s) +{ + if(sig >= 0 && sig < NSIG) { + if(s) + fprintf(stderr, "%s: %s\n", s, sys_siglist[sig]); + else + fprintf(stderr, "%s\n", sys_siglist[sig]); + } else { + if(s) + fprintf(stderr, "%s: invalid signal\n", s); + else + fputs("invalid signal\n", stderr); + } +} +#endif + +static ssize_t getline(char **lineptr, size_t *n, FILE *stream) +{ + size_t ret = 0; + + if (!lineptr || !n || !stream) + return -1; + + if(!*lineptr) { + *n = 128; + *lineptr = (char*)malloc(*n); + if(!*lineptr) + return -1; + } + + while(!feof(stream) && !ferror(stream)) { + int c; + if(ret == *n) { + *n += 128; + *lineptr = (char*)realloc(*lineptr, *n); + if(!*lineptr) { + *n = 0; + return -1; + } + } + c = fgetc(stream); + if(c == EOF) + break; + *lineptr[ret++] = c; + if(c == '\n') + break; + } + *lineptr[ret] = 0; + return ret; +} +#endif +#endif diff --git a/tools/perf/config/feature-tests.mak b/tools/perf/config/feature-tests.mak new file mode 100644 index 00000000..b4eabb44 --- /dev/null +++ b/tools/perf/config/feature-tests.mak @@ -0,0 +1,238 @@ +define SOURCE_HELLO +#include <stdio.h> +int main(void) +{ + return puts(\"hi\"); +} +endef + +ifndef NO_DWARF +define SOURCE_DWARF +#include <dwarf.h> +#include <elfutils/libdw.h> +#include <elfutils/version.h> +#ifndef _ELFUTILS_PREREQ +#error +#endif + +int main(void) +{ + Dwarf *dbg = dwarf_begin(0, DWARF_C_READ); + return (long)dbg; +} +endef +endif + +define SOURCE_LIBELF +#include <libelf.h> + +int main(void) +{ + Elf *elf = elf_begin(0, ELF_C_READ, 0); + return (long)elf; +} +endef + +define SOURCE_GLIBC +#include <gnu/libc-version.h> + +int main(void) +{ + const char *version = gnu_get_libc_version(); + return (long)version; +} +endef + +define SOURCE_BIONIC +#include <android/api-level.h> + +int main(void) +{ + return __ANDROID_API__; +} +endef + +define SOURCE_ELF_MMAP +#include <libelf.h> +int main(void) +{ + Elf *elf = elf_begin(0, ELF_C_READ_MMAP, 0); + return (long)elf; +} +endef + +ifndef NO_NEWT +define SOURCE_NEWT +#include <newt.h> + +int main(void) +{ + newtInit(); + newtCls(); + return newtFinished(); +} +endef +endif + +ifndef NO_GTK2 +define SOURCE_GTK2 +#pragma GCC diagnostic ignored \"-Wstrict-prototypes\" +#include <gtk/gtk.h> +#pragma GCC diagnostic error \"-Wstrict-prototypes\" + +int main(int argc, char *argv[]) +{ + gtk_init(&argc, &argv); + + return 0; +} +endef + +define SOURCE_GTK2_INFOBAR +#pragma GCC diagnostic ignored \"-Wstrict-prototypes\" +#include <gtk/gtk.h> +#pragma GCC diagnostic error \"-Wstrict-prototypes\" + +int main(void) +{ + gtk_info_bar_new(); + + return 0; +} +endef +endif + +ifndef NO_LIBPERL +define SOURCE_PERL_EMBED +#include <EXTERN.h> +#include <perl.h> + +int main(void) +{ +perl_alloc(); +return 0; +} +endef +endif + +ifndef NO_LIBPYTHON +define SOURCE_PYTHON_VERSION +#include <Python.h> +#if PY_VERSION_HEX >= 0x03000000 + #error +#endif +int main(void) +{ + return 0; +} +endef +define SOURCE_PYTHON_EMBED +#include <Python.h> +int main(void) +{ + Py_Initialize(); + return 0; +} +endef +endif + +define SOURCE_BFD +#include <bfd.h> + +int main(void) +{ + bfd_demangle(0, 0, 0); + return 0; +} +endef + +define SOURCE_CPLUS_DEMANGLE +extern char *cplus_demangle(const char *, int); + +int main(void) +{ + cplus_demangle(0, 0); + return 0; +} +endef + +define SOURCE_STRLCPY +#include <stdlib.h> +extern size_t strlcpy(char *dest, const char *src, size_t size); + +int main(void) +{ + strlcpy(NULL, NULL, 0); + return 0; +} +endef + +ifndef NO_LIBUNWIND +define SOURCE_LIBUNWIND +#include <libunwind.h> +#include <stdlib.h> + +extern int UNW_OBJ(dwarf_search_unwind_table) (unw_addr_space_t as, + unw_word_t ip, + unw_dyn_info_t *di, + unw_proc_info_t *pi, + int need_unwind_info, void *arg); + + +#define dwarf_search_unwind_table UNW_OBJ(dwarf_search_unwind_table) + +int main(void) +{ + unw_addr_space_t addr_space; + addr_space = unw_create_addr_space(NULL, 0); + unw_init_remote(NULL, addr_space, NULL); + dwarf_search_unwind_table(addr_space, 0, NULL, NULL, 0, NULL); + return 0; +} +endef +endif + +ifndef NO_BACKTRACE +define SOURCE_BACKTRACE +#include <execinfo.h> +#include <stdio.h> + +int main(void) +{ + backtrace(NULL, 0); + backtrace_symbols(NULL, 0); + return 0; +} +endef +endif + +ifndef NO_LIBAUDIT +define SOURCE_LIBAUDIT +#include <libaudit.h> + +int main(void) +{ + return audit_open(); +} +endef +endif + +define SOURCE_ON_EXIT +#include <stdio.h> + +int main(void) +{ + return on_exit(NULL, NULL); +} +endef + +define SOURCE_LIBNUMA +#include <numa.h> +#include <numaif.h> + +int main(void) +{ + numa_available(); + return 0; +} +endef
\ No newline at end of file diff --git a/tools/perf/config/utilities.mak b/tools/perf/config/utilities.mak new file mode 100644 index 00000000..3e897198 --- /dev/null +++ b/tools/perf/config/utilities.mak @@ -0,0 +1,194 @@ +# This allows us to work with the newline character: +define newline + + +endef +newline := $(newline) + +# nl-escape +# +# Usage: escape = $(call nl-escape[,escape]) +# +# This is used as the common way to specify +# what should replace a newline when escaping +# newlines; the default is a bizarre string. +# +nl-escape = $(if $(1),$(1),m822df3020w6a44id34bt574ctac44eb9f4n) + +# escape-nl +# +# Usage: escaped-text = $(call escape-nl,text[,escape]) +# +# GNU make's $(shell ...) function converts to a +# single space each newline character in the output +# produced during the expansion; this may not be +# desirable. +# +# The only solution is to change each newline into +# something that won't be converted, so that the +# information can be recovered later with +# $(call unescape-nl...) +# +escape-nl = $(subst $(newline),$(call nl-escape,$(2)),$(1)) + +# unescape-nl +# +# Usage: text = $(call unescape-nl,escaped-text[,escape]) +# +# See escape-nl. +# +unescape-nl = $(subst $(call nl-escape,$(2)),$(newline),$(1)) + +# shell-escape-nl +# +# Usage: $(shell some-command | $(call shell-escape-nl[,escape])) +# +# Use this to escape newlines from within a shell call; +# the default escape is a bizarre string. +# +# NOTE: The escape is used directly as a string constant +# in an `awk' program that is delimited by shell +# single-quotes, so be wary of the characters +# that are chosen. +# +define shell-escape-nl +awk 'NR==1 {t=$$0} NR>1 {t=t "$(nl-escape)" $$0} END {printf t}' +endef + +# shell-unescape-nl +# +# Usage: $(shell some-command | $(call shell-unescape-nl[,escape])) +# +# Use this to unescape newlines from within a shell call; +# the default escape is a bizarre string. +# +# NOTE: The escape is used directly as an extended regular +# expression constant in an `awk' program that is +# delimited by shell single-quotes, so be wary +# of the characters that are chosen. +# +# (The bash shell has a bug where `{gsub(...),...}' is +# misinterpreted as a brace expansion; this can be +# overcome by putting a space between `{' and `gsub'). +# +define shell-unescape-nl +awk 'NR==1 {t=$$0} NR>1 {t=t "\n" $$0} END { gsub(/$(nl-escape)/,"\n",t); printf t }' +endef + +# escape-for-shell-sq +# +# Usage: embeddable-text = $(call escape-for-shell-sq,text) +# +# This function produces text that is suitable for +# embedding in a shell string that is delimited by +# single-quotes. +# +escape-for-shell-sq = $(subst ','\'',$(1)) + +# shell-sq +# +# Usage: single-quoted-and-escaped-text = $(call shell-sq,text) +# +shell-sq = '$(escape-for-shell-sq)' + +# shell-wordify +# +# Usage: wordified-text = $(call shell-wordify,text) +# +# For instance: +# +# |define text +# |hello +# |world +# |endef +# | +# |target: +# | echo $(call shell-wordify,$(text)) +# +# At least GNU make gets confused by expanding a newline +# within the context of a command line of a makefile rule +# (this is in constrast to a `$(shell ...)' function call, +# which can handle it just fine). +# +# This function avoids the problem by producing a string +# that works as a shell word, regardless of whether or +# not it contains a newline. +# +# If the text to be wordified contains a newline, then +# an intrictate shell command substitution is constructed +# to render the text as a single line; when the shell +# processes the resulting escaped text, it transforms +# it into the original unescaped text. +# +# If the text does not contain a newline, then this function +# produces the same results as the `$(shell-sq)' function. +# +shell-wordify = $(if $(findstring $(newline),$(1)),$(_sw-esc-nl),$(shell-sq)) +define _sw-esc-nl +"$$(echo $(call escape-nl,$(shell-sq),$(2)) | $(call shell-unescape-nl,$(2)))" +endef + +# is-absolute +# +# Usage: bool-value = $(call is-absolute,path) +# +is-absolute = $(shell echo $(shell-sq) | grep ^/ -q && echo y) + +# lookup +# +# Usage: absolute-executable-path-or-empty = $(call lookup,path) +# +# (It's necessary to use `sh -c' because GNU make messes up by +# trying too hard and getting things wrong). +# +lookup = $(call unescape-nl,$(shell sh -c $(_l-sh))) +_l-sh = $(call shell-sq,command -v $(shell-sq) | $(call shell-escape-nl,)) + +# is-executable +# +# Usage: bool-value = $(call is-executable,path) +# +# (It's necessary to use `sh -c' because GNU make messes up by +# trying too hard and getting things wrong). +# +is-executable = $(call _is-executable-helper,$(shell-sq)) +_is-executable-helper = $(shell sh -c $(_is-executable-sh)) +_is-executable-sh = $(call shell-sq,test -f $(1) -a -x $(1) && echo y) + +# get-executable +# +# Usage: absolute-executable-path-or-empty = $(call get-executable,path) +# +# The goal is to get an absolute path for an executable; +# the `command -v' is defined by POSIX, but it's not +# necessarily very portable, so it's only used if +# relative path resolution is requested, as determined +# by the presence of a leading `/'. +# +get-executable = $(if $(1),$(if $(is-absolute),$(_ge-abspath),$(lookup))) +_ge-abspath = $(if $(is-executable),$(1)) + +# get-supplied-or-default-executable +# +# Usage: absolute-executable-path-or-empty = $(call get-executable-or-default,variable,default) +# +define get-executable-or-default +$(if $($(1)),$(call _ge_attempt,$($(1)),$(1)),$(call _ge_attempt,$(2))) +endef +_ge_attempt = $(if $(get-executable),$(get-executable),$(_gea_warn)$(call _gea_err,$(2))) +_gea_warn = $(warning The path '$(1)' is not executable.) +_gea_err = $(if $(1),$(error Please set '$(1)' appropriately)) + +# try-cc +# Usage: option = $(call try-cc, source-to-build, cc-options, msg) +ifndef V +TRY_CC_OUTPUT= > /dev/null 2>&1 +endif +TRY_CC_MSG=echo " CHK $(3)" 1>&2; + +try-cc = $(shell sh -c \ + 'TMP="$(OUTPUT)$(TMPOUT).$$$$"; \ + $(TRY_CC_MSG) \ + echo "$(1)" | \ + $(CC) -x c - $(2) -o "$$TMP" $(TRY_CC_OUTPUT) && echo y; \ + rm -f "$$TMP"') diff --git a/tools/perf/design.txt b/tools/perf/design.txt new file mode 100644 index 00000000..67e5d0ca --- /dev/null +++ b/tools/perf/design.txt @@ -0,0 +1,463 @@ + +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_event_open() +system call: + + int sys_perf_event_open(struct perf_event_attr *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_event_attr' is: + +struct perf_event_attr { + /* + * The MSB of the config word signifies if the rest contains cpu + * specific (raw) counter configuration data, if unset, the next + * 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_event_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, + PERF_COUNT_SW_ALIGNMENT_FAULTS = 7, + PERF_COUNT_SW_EMULATION_FAULTS = 8, +}; + +Counters of the type PERF_TYPE_TRACEPOINT are available when the ftrace event +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_event_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_event_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_event_open() system call allows the +counter to be specific to a task: + + pid == 0: if the pid parameter is zero, the counter is attached to the + 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_event_open call +that creates it. The rest of the group members are created +subsequently, with group_fd giving the fd of the group leader. +(A single counter on its own is created with group_fd = -1 and is +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_event_mmap_page) that contains various bits of information such +as where the ring-buffer head is. + +/* + * Structure of the page that can be mapped via mmap + */ +struct perf_event_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_event_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_RECORD_MISC_KERNEL (1 << 0) +#define PERF_RECORD_MISC_USER (1 << 1) +#define PERF_RECORD_MISC_OVERFLOW (1 << 2) + +struct perf_event_header { + __u32 type; + __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_RECORD_MMAP = 1, + PERF_RECORD_MUNMAP = 2, + + /* + * struct { + * struct perf_event_header header; + * + * u32 pid, tid; + * char comm[]; + * }; + */ + PERF_RECORD_COMM = 3, + + /* + * When header.misc & PERF_RECORD_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_event_attr.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 can be enabled with + + ioctl(fd, PERF_EVENT_IOC_ENABLE, 0); + +or disabled with + + ioctl(fd, PERF_EVENT_IOC_DISABLE, 0); + +For a counter group, pass PERF_IOC_FLAG_GROUP as the third argument. +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_EVENT_IOC_REFRESH, nr); + +to enable a counter for 'nr' events, after which it gets disabled again. + +A process can enable or disable all the counter groups that are +attached to it, using prctl: + + prctl(PR_TASK_PERF_EVENTS_ENABLE); + + prctl(PR_TASK_PERF_EVENTS_DISABLE); + +This applies to all counters on the current process, whether created +by this process or by another, and doesn't affect any counters that +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_EVENTS to your Kconfig, you +will need at least this: + - asm/perf_event.h - a basic stub will suffice at first + - support for atomic64 types (and associated helper functions) + - set_perf_event_pending() implemented + +If your architecture does have hardware capabilities, you can override the +weak stub hw_perf_event_init() to register hardware counters. + +Architectures that have d-cache aliassing issues, such as Sparc and ARM, +should select PERF_USE_VMALLOC in order to avoid these for perf mmap(). diff --git a/tools/perf/perf-archive.sh b/tools/perf/perf-archive.sh new file mode 100644 index 00000000..e9193062 --- /dev/null +++ b/tools/perf/perf-archive.sh @@ -0,0 +1,47 @@ +#!/bin/bash +# perf archive +# Arnaldo Carvalho de Melo <acme@redhat.com> + +PERF_DATA=perf.data +if [ $# -ne 0 ] ; then + PERF_DATA=$1 +fi + +# +# PERF_BUILDID_DIR environment variable set by perf +# path to buildid directory, default to $HOME/.debug +# +if [ -z $PERF_BUILDID_DIR ]; then + PERF_BUILDID_DIR=~/.debug/ +else + # append / to make substitutions work + PERF_BUILDID_DIR=$PERF_BUILDID_DIR/ +fi + +BUILDIDS=$(mktemp /tmp/perf-archive-buildids.XXXXXX) +NOBUILDID=0000000000000000000000000000000000000000 + +perf buildid-list -i $PERF_DATA --with-hits | grep -v "^$NOBUILDID " > $BUILDIDS +if [ ! -s $BUILDIDS ] ; then + echo "perf archive: no build-ids found" + rm $BUILDIDS || true + exit 1 +fi + +MANIFEST=$(mktemp /tmp/perf-archive-manifest.XXXXXX) +PERF_BUILDID_LINKDIR=$(readlink -f $PERF_BUILDID_DIR)/ + +cut -d ' ' -f 1 $BUILDIDS | \ +while read build_id ; do + linkname=$PERF_BUILDID_DIR.build-id/${build_id:0:2}/${build_id:2} + filename=$(readlink -f $linkname) + echo ${linkname#$PERF_BUILDID_DIR} >> $MANIFEST + echo ${filename#$PERF_BUILDID_LINKDIR} >> $MANIFEST +done + +tar cjf $PERF_DATA.tar.bz2 -C $PERF_BUILDID_DIR -T $MANIFEST +rm $MANIFEST $BUILDIDS || true +echo -e "Now please run:\n" +echo -e "$ tar xvf $PERF_DATA.tar.bz2 -C ~/.debug\n" +echo "wherever you need to run 'perf report' on." +exit 0 diff --git a/tools/perf/perf.c b/tools/perf/perf.c new file mode 100644 index 00000000..095b8820 --- /dev/null +++ b/tools/perf/perf.c @@ -0,0 +1,543 @@ +/* + * 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" +#include "util/parse-events.h" +#include "util/debugfs.h" +#include <pthread.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."; + +int use_browser = -1; +static int use_pager = -1; +const char *input_name; + +struct cmd_struct { + const char *cmd; + int (*fn)(int, const char **, const char *); + int option; +}; + +static struct cmd_struct commands[] = { + { "buildid-cache", cmd_buildid_cache, 0 }, + { "buildid-list", cmd_buildid_list, 0 }, + { "diff", cmd_diff, 0 }, + { "evlist", cmd_evlist, 0 }, + { "help", cmd_help, 0 }, + { "list", cmd_list, 0 }, + { "record", cmd_record, 0 }, + { "report", cmd_report, 0 }, + { "bench", cmd_bench, 0 }, + { "stat", cmd_stat, 0 }, + { "timechart", cmd_timechart, 0 }, + { "top", cmd_top, 0 }, + { "annotate", cmd_annotate, 0 }, + { "version", cmd_version, 0 }, + { "script", cmd_script, 0 }, + { "sched", cmd_sched, 0 }, +#ifdef LIBELF_SUPPORT + { "probe", cmd_probe, 0 }, +#endif + { "kmem", cmd_kmem, 0 }, + { "lock", cmd_lock, 0 }, + { "kvm", cmd_kvm, 0 }, + { "test", cmd_test, 0 }, +#ifdef LIBAUDIT_SUPPORT + { "trace", cmd_trace, 0 }, +#endif + { "inject", cmd_inject, 0 }, +}; + +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 int browser_command_config(const char *var, const char *value, void *data) +{ + struct pager_config *c = data; + if (!prefixcmp(var, "tui.") && !strcmp(var + 4, c->cmd)) + c->val = perf_config_bool(var, value); + if (!prefixcmp(var, "gtk.") && !strcmp(var + 4, c->cmd)) + c->val = perf_config_bool(var, value) ? 2 : 0; + return 0; +} + +/* + * returns 0 for "no tui", 1 for "use tui", 2 for "use gtk", + * and -1 for "not specified" + */ +static int check_browser_config(const char *cmd) +{ + struct pager_config c; + c.cmd = cmd; + c.val = -1; + perf_config(browser_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, CMD_EXEC_PATH)) { + cmd += strlen(CMD_EXEC_PATH); + 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, CMD_PERF_DIR)) { + setenv(PERF_DIR_ENVIRONMENT, cmd + strlen(CMD_PERF_DIR), 1); + if (envchanged) + *envchanged = 1; + } else if (!strcmp(cmd, "--work-tree")) { + if (*argc < 2) { + fprintf(stderr, "No directory given for --work-tree.\n"); + usage(perf_usage_string); + } + setenv(PERF_WORK_TREE_ENVIRONMENT, (*argv)[1], 1); + if (envchanged) + *envchanged = 1; + (*argv)++; + (*argc)--; + } else if (!prefixcmp(cmd, CMD_WORK_TREE)) { + setenv(PERF_WORK_TREE_ENVIRONMENT, cmd + strlen(CMD_WORK_TREE), 1); + if (envchanged) + *envchanged = 1; + } else if (!strcmp(cmd, "--debugfs-dir")) { + if (*argc < 2) { + fprintf(stderr, "No directory given for --debugfs-dir.\n"); + usage(perf_usage_string); + } + debugfs_set_path((*argv)[1]); + if (envchanged) + *envchanged = 1; + (*argv)++; + (*argc)--; + } else if (!prefixcmp(cmd, CMD_DEBUGFS_DIR)) { + debugfs_set_path(cmd + strlen(CMD_DEBUGFS_DIR)); + fprintf(stderr, "dir: %s\n", debugfs_mountpoint); + if (envchanged) + *envchanged = 1; + } else if (!strcmp(cmd, "--list-cmds")) { + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(commands); i++) { + struct cmd_struct *p = commands+i; + printf("%s ", p->cmd); + } + exit(0); + } 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) + +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_browser == -1) + use_browser = check_browser_config(p->cmd); + + if (use_pager == -1 && p->option & RUN_SETUP) + use_pager = check_pager_config(p->cmd); + if (use_pager == -1 && p->option & USE_PAGER) + use_pager = 1; + commit_pager_choice(); + + status = p->fn(argc, argv, prefix); + exit_browser(status); + + 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; + + status = 1; + /* Check for ENOSPC and EIO errors.. */ + if (fflush(stdout)) { + fprintf(stderr, "write failure on standard output: %s", strerror(errno)); + goto out; + } + if (ferror(stdout)) { + fprintf(stderr, "unknown write failure on standard output"); + goto out; + } + if (fclose(stdout)) { + fprintf(stderr, "close failed on standard output: %s", strerror(errno)); + goto out; + } + status = 0; +out: + return status; +} + +static void handle_internal_command(int argc, const char **argv) +{ + const char *cmd = argv[0]; + unsigned 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; +} + +static void pthread__block_sigwinch(void) +{ + sigset_t set; + + sigemptyset(&set); + sigaddset(&set, SIGWINCH); + pthread_sigmask(SIG_BLOCK, &set, NULL); +} + +void pthread__unblock_sigwinch(void) +{ + sigset_t set; + + sigemptyset(&set); + sigaddset(&set, SIGWINCH); + pthread_sigmask(SIG_UNBLOCK, &set, NULL); +} + +int main(int argc, const char **argv) +{ + const char *cmd; + + page_size = sysconf(_SC_PAGE_SIZE); + + cmd = perf_extract_argv0_path(argv[0]); + if (!cmd) + cmd = "perf-help"; + /* get debugfs mount point from /proc/mounts */ + debugfs_mount(NULL); + /* + * "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); + fprintf(stderr, "cannot handle %s internally", cmd); + goto out; + } + + /* Look for flags.. */ + argv++; + argc--; + handle_options(&argv, &argc, NULL); + commit_pager_choice(); + set_buildid_dir(); + + 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); + goto out; + } + cmd = argv[0]; + + test_attr__init(); + + /* + * We use PATH to find perf commands, but we prepend some higher + * precedence paths: the "--exec-path" option, the PERF_EXEC_PATH + * environment, and the $(perfexecdir) from the Makefile at build + * time. + */ + setup_path(); + /* + * Block SIGWINCH notifications so that the thread that wants it can + * unblock and get syscalls like select interrupted instead of waiting + * forever while the signal goes to some other non interested thread. + */ + pthread__block_sigwinch(); + + while (1) { + static int done_help; + static int was_alias; + + 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]); + goto out; + } + 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)); +out: + return 1; +} diff --git a/tools/perf/perf.h b/tools/perf/perf.h new file mode 100644 index 00000000..74659ecf --- /dev/null +++ b/tools/perf/perf.h @@ -0,0 +1,232 @@ +#ifndef _PERF_PERF_H +#define _PERF_PERF_H + +#include <asm/unistd.h> + +#if defined(__i386__) +#define rmb() asm volatile("lock; addl $0,0(%%esp)" ::: "memory") +#define cpu_relax() asm volatile("rep; nop" ::: "memory"); +#define CPUINFO_PROC "model name" +#ifndef __NR_perf_event_open +# define __NR_perf_event_open 336 +#endif +#endif + +#if defined(__x86_64__) +#define rmb() asm volatile("lfence" ::: "memory") +#define cpu_relax() asm volatile("rep; nop" ::: "memory"); +#define CPUINFO_PROC "model name" +#ifndef __NR_perf_event_open +# define __NR_perf_event_open 298 +#endif +#endif + +#ifdef __powerpc__ +#include "../../arch/powerpc/include/uapi/asm/unistd.h" +#define rmb() asm volatile ("sync" ::: "memory") +#define cpu_relax() asm volatile ("" ::: "memory"); +#define CPUINFO_PROC "cpu" +#endif + +#ifdef __s390__ +#define rmb() asm volatile("bcr 15,0" ::: "memory") +#define cpu_relax() asm volatile("" ::: "memory"); +#endif + +#ifdef __sh__ +#if defined(__SH4A__) || defined(__SH5__) +# define rmb() asm volatile("synco" ::: "memory") +#else +# define rmb() asm volatile("" ::: "memory") +#endif +#define cpu_relax() asm volatile("" ::: "memory") +#define CPUINFO_PROC "cpu type" +#endif + +#ifdef __hppa__ +#define rmb() asm volatile("" ::: "memory") +#define cpu_relax() asm volatile("" ::: "memory"); +#define CPUINFO_PROC "cpu" +#endif + +#ifdef __sparc__ +#define rmb() asm volatile("":::"memory") +#define cpu_relax() asm volatile("":::"memory") +#define CPUINFO_PROC "cpu" +#endif + +#ifdef __alpha__ +#define rmb() asm volatile("mb" ::: "memory") +#define cpu_relax() asm volatile("" ::: "memory") +#define CPUINFO_PROC "cpu model" +#endif + +#ifdef __ia64__ +#define rmb() asm volatile ("mf" ::: "memory") +#define cpu_relax() asm volatile ("hint @pause" ::: "memory") +#define CPUINFO_PROC "model name" +#endif + +#ifdef __arm__ +/* + * Use the __kuser_memory_barrier helper in the CPU helper page. See + * arch/arm/kernel/entry-armv.S in the kernel source for details. + */ +#define rmb() ((void(*)(void))0xffff0fa0)() +#define cpu_relax() asm volatile("":::"memory") +#define CPUINFO_PROC "Processor" +#endif + +#ifdef __aarch64__ +#define rmb() asm volatile("dmb ld" ::: "memory") +#define cpu_relax() asm volatile("yield" ::: "memory") +#endif + +#ifdef __mips__ +#define rmb() asm volatile( \ + ".set mips2\n\t" \ + "sync\n\t" \ + ".set mips0" \ + : /* no output */ \ + : /* no input */ \ + : "memory") +#define cpu_relax() asm volatile("" ::: "memory") +#define CPUINFO_PROC "cpu model" +#endif + +#ifdef __arc__ +#define rmb() asm volatile("" ::: "memory") +#define cpu_relax() rmb() +#define CPUINFO_PROC "Processor" +#endif + +#ifdef __metag__ +#define rmb() asm volatile("" ::: "memory") +#define cpu_relax() asm volatile("" ::: "memory") +#define CPUINFO_PROC "CPU" +#endif + +#include <time.h> +#include <unistd.h> +#include <sys/types.h> +#include <sys/syscall.h> + +#include <linux/perf_event.h> +#include "util/types.h" +#include <stdbool.h> + +/* + * prctl(PR_TASK_PERF_EVENTS_DISABLE) will (cheaply) disable all + * counters in the current task. + */ +#define PR_TASK_PERF_EVENTS_DISABLE 31 +#define PR_TASK_PERF_EVENTS_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; }) + +extern bool test_attr__enabled; +void test_attr__init(void); +void test_attr__open(struct perf_event_attr *attr, pid_t pid, int cpu, + int fd, int group_fd, unsigned long flags); + +static inline int +sys_perf_event_open(struct perf_event_attr *attr, + pid_t pid, int cpu, int group_fd, + unsigned long flags) +{ + int fd; + + fd = syscall(__NR_perf_event_open, attr, pid, cpu, + group_fd, flags); + + if (unlikely(test_attr__enabled)) + test_attr__open(attr, pid, cpu, fd, group_fd, flags); + + return fd; +} + +#define MAX_COUNTERS 256 +#define MAX_NR_CPUS 256 + +struct ip_callchain { + u64 nr; + u64 ips[0]; +}; + +struct branch_flags { + u64 mispred:1; + u64 predicted:1; + u64 reserved:62; +}; + +struct branch_entry { + u64 from; + u64 to; + struct branch_flags flags; +}; + +struct branch_stack { + u64 nr; + struct branch_entry entries[0]; +}; + +extern const char *input_name; +extern bool perf_host, perf_guest; +extern const char perf_version_string[]; + +void pthread__unblock_sigwinch(void); + +#include "util/target.h" + +enum perf_call_graph_mode { + CALLCHAIN_NONE, + CALLCHAIN_FP, + CALLCHAIN_DWARF +}; + +struct perf_record_opts { + struct perf_target target; + int call_graph; + bool group; + bool inherit_stat; + bool no_delay; + bool no_inherit; + bool no_samples; + bool pipe_output; + bool raw_samples; + bool sample_address; + bool sample_time; + bool period; + unsigned int freq; + unsigned int mmap_pages; + unsigned int user_freq; + u64 branch_stack; + u64 default_interval; + u64 user_interval; + u16 stack_dump_size; +}; + +#endif diff --git a/tools/perf/python/twatch.py b/tools/perf/python/twatch.py new file mode 100755 index 00000000..b11cca58 --- /dev/null +++ b/tools/perf/python/twatch.py @@ -0,0 +1,41 @@ +#! /usr/bin/python +# -*- python -*- +# -*- coding: utf-8 -*- +# twatch - Experimental use of the perf python interface +# Copyright (C) 2011 Arnaldo Carvalho de Melo <acme@redhat.com> +# +# This application is free software; you can redistribute it and/or +# modify it under the terms of the GNU General Public License +# as published by the Free Software Foundation; version 2. +# +# This application 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. + +import perf + +def main(): + cpus = perf.cpu_map() + threads = perf.thread_map() + evsel = perf.evsel(task = 1, comm = 1, mmap = 0, + wakeup_events = 1, watermark = 1, + sample_id_all = 1, + sample_type = perf.SAMPLE_PERIOD | perf.SAMPLE_TID | perf.SAMPLE_CPU | perf.SAMPLE_TID) + evsel.open(cpus = cpus, threads = threads); + evlist = perf.evlist(cpus, threads) + evlist.add(evsel) + evlist.mmap() + while True: + evlist.poll(timeout = -1) + for cpu in cpus: + event = evlist.read_on_cpu(cpu) + if not event: + continue + print "cpu: %2d, pid: %4d, tid: %4d" % (event.sample_cpu, + event.sample_pid, + event.sample_tid), + print event + +if __name__ == '__main__': + main() diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Context.c b/tools/perf/scripts/perl/Perf-Trace-Util/Context.c new file mode 100644 index 00000000..790ceba6 --- /dev/null +++ b/tools/perf/scripts/perl/Perf-Trace-Util/Context.c @@ -0,0 +1,135 @@ +/* + * This file was generated automatically by ExtUtils::ParseXS version 2.18_02 from the + * contents of Context.xs. Do not edit this file, edit Context.xs instead. + * + * ANY CHANGES MADE HERE WILL BE LOST! + * + */ + +#line 1 "Context.xs" +/* + * Context.xs. XS interfaces for perf script. + * + * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include "EXTERN.h" +#include "perl.h" +#include "XSUB.h" +#include "../../../perf.h" +#include "../../../util/trace-event.h" + +#ifndef PERL_UNUSED_VAR +# define PERL_UNUSED_VAR(var) if (0) var = var +#endif + +#line 42 "Context.c" + +XS(XS_Perf__Trace__Context_common_pc); /* prototype to pass -Wmissing-prototypes */ +XS(XS_Perf__Trace__Context_common_pc) +{ +#ifdef dVAR + dVAR; dXSARGS; +#else + dXSARGS; +#endif + if (items != 1) + Perl_croak(aTHX_ "Usage: %s(%s)", "Perf::Trace::Context::common_pc", "context"); + PERL_UNUSED_VAR(cv); /* -W */ + { + struct scripting_context * context = INT2PTR(struct scripting_context *,SvIV(ST(0))); + int RETVAL; + dXSTARG; + + RETVAL = common_pc(context); + XSprePUSH; PUSHi((IV)RETVAL); + } + XSRETURN(1); +} + + +XS(XS_Perf__Trace__Context_common_flags); /* prototype to pass -Wmissing-prototypes */ +XS(XS_Perf__Trace__Context_common_flags) +{ +#ifdef dVAR + dVAR; dXSARGS; +#else + dXSARGS; +#endif + if (items != 1) + Perl_croak(aTHX_ "Usage: %s(%s)", "Perf::Trace::Context::common_flags", "context"); + PERL_UNUSED_VAR(cv); /* -W */ + { + struct scripting_context * context = INT2PTR(struct scripting_context *,SvIV(ST(0))); + int RETVAL; + dXSTARG; + + RETVAL = common_flags(context); + XSprePUSH; PUSHi((IV)RETVAL); + } + XSRETURN(1); +} + + +XS(XS_Perf__Trace__Context_common_lock_depth); /* prototype to pass -Wmissing-prototypes */ +XS(XS_Perf__Trace__Context_common_lock_depth) +{ +#ifdef dVAR + dVAR; dXSARGS; +#else + dXSARGS; +#endif + if (items != 1) + Perl_croak(aTHX_ "Usage: %s(%s)", "Perf::Trace::Context::common_lock_depth", "context"); + PERL_UNUSED_VAR(cv); /* -W */ + { + struct scripting_context * context = INT2PTR(struct scripting_context *,SvIV(ST(0))); + int RETVAL; + dXSTARG; + + RETVAL = common_lock_depth(context); + XSprePUSH; PUSHi((IV)RETVAL); + } + XSRETURN(1); +} + +#ifdef __cplusplus +extern "C" +#endif +XS(boot_Perf__Trace__Context); /* prototype to pass -Wmissing-prototypes */ +XS(boot_Perf__Trace__Context) +{ +#ifdef dVAR + dVAR; dXSARGS; +#else + dXSARGS; +#endif + const char* file = __FILE__; + + PERL_UNUSED_VAR(cv); /* -W */ + PERL_UNUSED_VAR(items); /* -W */ + XS_VERSION_BOOTCHECK ; + + newXSproto("Perf::Trace::Context::common_pc", XS_Perf__Trace__Context_common_pc, file, "$"); + newXSproto("Perf::Trace::Context::common_flags", XS_Perf__Trace__Context_common_flags, file, "$"); + newXSproto("Perf::Trace::Context::common_lock_depth", XS_Perf__Trace__Context_common_lock_depth, file, "$"); + if (PL_unitcheckav) + call_list(PL_scopestack_ix, PL_unitcheckav); + XSRETURN_YES; +} + diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs b/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs new file mode 100644 index 00000000..c1e2ed1e --- /dev/null +++ b/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs @@ -0,0 +1,42 @@ +/* + * Context.xs. XS interfaces for perf script. + * + * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include "EXTERN.h" +#include "perl.h" +#include "XSUB.h" +#include "../../../perf.h" +#include "../../../util/script-event.h" + +MODULE = Perf::Trace::Context PACKAGE = Perf::Trace::Context +PROTOTYPES: ENABLE + +int +common_pc(context) + struct scripting_context * context + +int +common_flags(context) + struct scripting_context * context + +int +common_lock_depth(context) + struct scripting_context * context + diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Makefile.PL b/tools/perf/scripts/perl/Perf-Trace-Util/Makefile.PL new file mode 100644 index 00000000..decdeb0f --- /dev/null +++ b/tools/perf/scripts/perl/Perf-Trace-Util/Makefile.PL @@ -0,0 +1,17 @@ +use 5.010000; +use ExtUtils::MakeMaker; +# See lib/ExtUtils/MakeMaker.pm for details of how to influence +# the contents of the Makefile that is written. +WriteMakefile( + NAME => 'Perf::Trace::Context', + VERSION_FROM => 'lib/Perf/Trace/Context.pm', # finds $VERSION + PREREQ_PM => {}, # e.g., Module::Name => 1.1 + ($] >= 5.005 ? ## Add these new keywords supported since 5.005 + (ABSTRACT_FROM => 'lib/Perf/Trace/Context.pm', # retrieve abstract from module + AUTHOR => 'Tom Zanussi <tzanussi@gmail.com>') : ()), + LIBS => [''], # e.g., '-lm' + DEFINE => '-I ../..', # e.g., '-DHAVE_SOMETHING' + INC => '-I.', # e.g., '-I. -I/usr/include/other' + # Un-comment this if you add C files to link with later: + OBJECT => 'Context.o', # link all the C files too +); diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/README b/tools/perf/scripts/perl/Perf-Trace-Util/README new file mode 100644 index 00000000..2f0c7f30 --- /dev/null +++ b/tools/perf/scripts/perl/Perf-Trace-Util/README @@ -0,0 +1,59 @@ +Perf-Trace-Util version 0.01 +============================ + +This module contains utility functions for use with perf script. + +Core.pm and Util.pm are pure Perl modules; Core.pm contains routines +that the core perf support for Perl calls on and should always be +'used', while Util.pm contains useful but optional utility functions +that scripts may want to use. Context.pm contains the Perl->C +interface that allows scripts to access data in the embedding perf +executable; scripts wishing to do that should 'use Context.pm'. + +The Perl->C perf interface is completely driven by Context.xs. If you +want to add new Perl functions that end up accessing C data in the +perf executable, you add desciptions of the new functions here. +scripting_context is a pointer to the perf data in the perf executable +that you want to access - it's passed as the second parameter, +$context, to all handler functions. + +After you do that: + + perl Makefile.PL # to create a Makefile for the next step + make # to create Context.c + + edit Context.c to add const to the char* file = __FILE__ line in + XS(boot_Perf__Trace__Context) to silence a warning/error. + + You can delete the Makefile, object files and anything else that was + generated e.g. blib and shared library, etc, except for of course + Context.c + + You should then be able to run the normal perf make as usual. + +INSTALLATION + +Building perf with perf script Perl scripting should install this +module in the right place. + +You should make sure libperl and ExtUtils/Embed.pm are installed first +e.g. apt-get install libperl-dev or yum install perl-ExtUtils-Embed. + +DEPENDENCIES + +This module requires these other modules and libraries: + + None + +COPYRIGHT AND LICENCE + +Copyright (C) 2009 by Tom Zanussi <tzanussi@gmail.com> + +This library is free software; you can redistribute it and/or modify +it under the same terms as Perl itself, either Perl version 5.10.0 or, +at your option, any later version of Perl 5 you may have available. + +Alternatively, this software may be distributed under the terms of the +GNU General Public License ("GPL") version 2 as published by the Free +Software Foundation. + diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm new file mode 100644 index 00000000..4e2f6039 --- /dev/null +++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm @@ -0,0 +1,55 @@ +package Perf::Trace::Context; + +use 5.010000; +use strict; +use warnings; + +require Exporter; + +our @ISA = qw(Exporter); + +our %EXPORT_TAGS = ( 'all' => [ qw( +) ] ); + +our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } ); + +our @EXPORT = qw( + common_pc common_flags common_lock_depth +); + +our $VERSION = '0.01'; + +require XSLoader; +XSLoader::load('Perf::Trace::Context', $VERSION); + +1; +__END__ +=head1 NAME + +Perf::Trace::Context - Perl extension for accessing functions in perf. + +=head1 SYNOPSIS + + use Perf::Trace::Context; + +=head1 SEE ALSO + +Perf (script) documentation + +=head1 AUTHOR + +Tom Zanussi, E<lt>tzanussi@gmail.com<gt> + +=head1 COPYRIGHT AND LICENSE + +Copyright (C) 2009 by Tom Zanussi + +This library is free software; you can redistribute it and/or modify +it under the same terms as Perl itself, either Perl version 5.10.0 or, +at your option, any later version of Perl 5 you may have available. + +Alternatively, this software may be distributed under the terms of the +GNU General Public License ("GPL") version 2 as published by the Free +Software Foundation. + +=cut diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm new file mode 100644 index 00000000..9158458d --- /dev/null +++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm @@ -0,0 +1,192 @@ +package Perf::Trace::Core; + +use 5.010000; +use strict; +use warnings; + +require Exporter; + +our @ISA = qw(Exporter); + +our %EXPORT_TAGS = ( 'all' => [ qw( +) ] ); + +our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } ); + +our @EXPORT = qw( +define_flag_field define_flag_value flag_str dump_flag_fields +define_symbolic_field define_symbolic_value symbol_str dump_symbolic_fields +trace_flag_str +); + +our $VERSION = '0.01'; + +my %trace_flags = (0x00 => "NONE", + 0x01 => "IRQS_OFF", + 0x02 => "IRQS_NOSUPPORT", + 0x04 => "NEED_RESCHED", + 0x08 => "HARDIRQ", + 0x10 => "SOFTIRQ"); + +sub trace_flag_str +{ + my ($value) = @_; + + my $string; + + my $print_delim = 0; + + foreach my $idx (sort {$a <=> $b} keys %trace_flags) { + if (!$value && !$idx) { + $string .= "NONE"; + last; + } + + if ($idx && ($value & $idx) == $idx) { + if ($print_delim) { + $string .= " | "; + } + $string .= "$trace_flags{$idx}"; + $print_delim = 1; + $value &= ~$idx; + } + } + + return $string; +} + +my %flag_fields; +my %symbolic_fields; + +sub flag_str +{ + my ($event_name, $field_name, $value) = @_; + + my $string; + + if ($flag_fields{$event_name}{$field_name}) { + my $print_delim = 0; + foreach my $idx (sort {$a <=> $b} keys %{$flag_fields{$event_name}{$field_name}{"values"}}) { + if (!$value && !$idx) { + $string .= "$flag_fields{$event_name}{$field_name}{'values'}{$idx}"; + last; + } + if ($idx && ($value & $idx) == $idx) { + if ($print_delim && $flag_fields{$event_name}{$field_name}{'delim'}) { + $string .= " $flag_fields{$event_name}{$field_name}{'delim'} "; + } + $string .= "$flag_fields{$event_name}{$field_name}{'values'}{$idx}"; + $print_delim = 1; + $value &= ~$idx; + } + } + } + + return $string; +} + +sub define_flag_field +{ + my ($event_name, $field_name, $delim) = @_; + + $flag_fields{$event_name}{$field_name}{"delim"} = $delim; +} + +sub define_flag_value +{ + my ($event_name, $field_name, $value, $field_str) = @_; + + $flag_fields{$event_name}{$field_name}{"values"}{$value} = $field_str; +} + +sub dump_flag_fields +{ + for my $event (keys %flag_fields) { + print "event $event:\n"; + for my $field (keys %{$flag_fields{$event}}) { + print " field: $field:\n"; + print " delim: $flag_fields{$event}{$field}{'delim'}\n"; + foreach my $idx (sort {$a <=> $b} keys %{$flag_fields{$event}{$field}{"values"}}) { + print " value $idx: $flag_fields{$event}{$field}{'values'}{$idx}\n"; + } + } + } +} + +sub symbol_str +{ + my ($event_name, $field_name, $value) = @_; + + if ($symbolic_fields{$event_name}{$field_name}) { + foreach my $idx (sort {$a <=> $b} keys %{$symbolic_fields{$event_name}{$field_name}{"values"}}) { + if (!$value && !$idx) { + return "$symbolic_fields{$event_name}{$field_name}{'values'}{$idx}"; + last; + } + if ($value == $idx) { + return "$symbolic_fields{$event_name}{$field_name}{'values'}{$idx}"; + } + } + } + + return undef; +} + +sub define_symbolic_field +{ + my ($event_name, $field_name) = @_; + + # nothing to do, really +} + +sub define_symbolic_value +{ + my ($event_name, $field_name, $value, $field_str) = @_; + + $symbolic_fields{$event_name}{$field_name}{"values"}{$value} = $field_str; +} + +sub dump_symbolic_fields +{ + for my $event (keys %symbolic_fields) { + print "event $event:\n"; + for my $field (keys %{$symbolic_fields{$event}}) { + print " field: $field:\n"; + foreach my $idx (sort {$a <=> $b} keys %{$symbolic_fields{$event}{$field}{"values"}}) { + print " value $idx: $symbolic_fields{$event}{$field}{'values'}{$idx}\n"; + } + } + } +} + +1; +__END__ +=head1 NAME + +Perf::Trace::Core - Perl extension for perf script + +=head1 SYNOPSIS + + use Perf::Trace::Core + +=head1 SEE ALSO + +Perf (script) documentation + +=head1 AUTHOR + +Tom Zanussi, E<lt>tzanussi@gmail.com<gt> + +=head1 COPYRIGHT AND LICENSE + +Copyright (C) 2009 by Tom Zanussi + +This library is free software; you can redistribute it and/or modify +it under the same terms as Perl itself, either Perl version 5.10.0 or, +at your option, any later version of Perl 5 you may have available. + +Alternatively, this software may be distributed under the terms of the +GNU General Public License ("GPL") version 2 as published by the Free +Software Foundation. + +=cut diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm new file mode 100644 index 00000000..05350011 --- /dev/null +++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm @@ -0,0 +1,94 @@ +package Perf::Trace::Util; + +use 5.010000; +use strict; +use warnings; + +require Exporter; + +our @ISA = qw(Exporter); + +our %EXPORT_TAGS = ( 'all' => [ qw( +) ] ); + +our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } ); + +our @EXPORT = qw( +avg nsecs nsecs_secs nsecs_nsecs nsecs_usecs print_nsecs +clear_term +); + +our $VERSION = '0.01'; + +sub avg +{ + my ($total, $n) = @_; + + return $total / $n; +} + +my $NSECS_PER_SEC = 1000000000; + +sub nsecs +{ + my ($secs, $nsecs) = @_; + + return $secs * $NSECS_PER_SEC + $nsecs; +} + +sub nsecs_secs { + my ($nsecs) = @_; + + return $nsecs / $NSECS_PER_SEC; +} + +sub nsecs_nsecs { + my ($nsecs) = @_; + + return $nsecs % $NSECS_PER_SEC; +} + +sub nsecs_str { + my ($nsecs) = @_; + + my $str = sprintf("%5u.%09u", nsecs_secs($nsecs), nsecs_nsecs($nsecs)); + + return $str; +} + +sub clear_term +{ + print "\x1b[H\x1b[2J"; +} + +1; +__END__ +=head1 NAME + +Perf::Trace::Util - Perl extension for perf script + +=head1 SYNOPSIS + + use Perf::Trace::Util; + +=head1 SEE ALSO + +Perf (script) documentation + +=head1 AUTHOR + +Tom Zanussi, E<lt>tzanussi@gmail.com<gt> + +=head1 COPYRIGHT AND LICENSE + +Copyright (C) 2009 by Tom Zanussi + +This library is free software; you can redistribute it and/or modify +it under the same terms as Perl itself, either Perl version 5.10.0 or, +at your option, any later version of Perl 5 you may have available. + +Alternatively, this software may be distributed under the terms of the +GNU General Public License ("GPL") version 2 as published by the Free +Software Foundation. + +=cut diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/typemap b/tools/perf/scripts/perl/Perf-Trace-Util/typemap new file mode 100644 index 00000000..84083680 --- /dev/null +++ b/tools/perf/scripts/perl/Perf-Trace-Util/typemap @@ -0,0 +1 @@ +struct scripting_context * T_PTR diff --git a/tools/perf/scripts/perl/bin/check-perf-trace-record b/tools/perf/scripts/perl/bin/check-perf-trace-record new file mode 100644 index 00000000..423ad6ae --- /dev/null +++ b/tools/perf/scripts/perl/bin/check-perf-trace-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -a -e kmem:kmalloc -e irq:softirq_entry -e kmem:kfree diff --git a/tools/perf/scripts/perl/bin/failed-syscalls-record b/tools/perf/scripts/perl/bin/failed-syscalls-record new file mode 100644 index 00000000..8104895a --- /dev/null +++ b/tools/perf/scripts/perl/bin/failed-syscalls-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e raw_syscalls:sys_exit $@ diff --git a/tools/perf/scripts/perl/bin/failed-syscalls-report b/tools/perf/scripts/perl/bin/failed-syscalls-report new file mode 100644 index 00000000..9f83cc1a --- /dev/null +++ b/tools/perf/scripts/perl/bin/failed-syscalls-report @@ -0,0 +1,10 @@ +#!/bin/bash +# description: system-wide failed syscalls +# args: [comm] +if [ $# -gt 0 ] ; then + if ! expr match "$1" "-" > /dev/null ; then + comm=$1 + shift + fi +fi +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/failed-syscalls.pl $comm diff --git a/tools/perf/scripts/perl/bin/rw-by-file-record b/tools/perf/scripts/perl/bin/rw-by-file-record new file mode 100644 index 00000000..33efc867 --- /dev/null +++ b/tools/perf/scripts/perl/bin/rw-by-file-record @@ -0,0 +1,3 @@ +#!/bin/bash +perf record -e syscalls:sys_enter_read -e syscalls:sys_enter_write $@ + diff --git a/tools/perf/scripts/perl/bin/rw-by-file-report b/tools/perf/scripts/perl/bin/rw-by-file-report new file mode 100644 index 00000000..77200b3f --- /dev/null +++ b/tools/perf/scripts/perl/bin/rw-by-file-report @@ -0,0 +1,10 @@ +#!/bin/bash +# description: r/w activity for a program, by file +# args: <comm> +if [ $# -lt 1 ] ; then + echo "usage: rw-by-file <comm>" + exit +fi +comm=$1 +shift +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rw-by-file.pl $comm diff --git a/tools/perf/scripts/perl/bin/rw-by-pid-record b/tools/perf/scripts/perl/bin/rw-by-pid-record new file mode 100644 index 00000000..7cb9db23 --- /dev/null +++ b/tools/perf/scripts/perl/bin/rw-by-pid-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@ diff --git a/tools/perf/scripts/perl/bin/rw-by-pid-report b/tools/perf/scripts/perl/bin/rw-by-pid-report new file mode 100644 index 00000000..a27b9f31 --- /dev/null +++ b/tools/perf/scripts/perl/bin/rw-by-pid-report @@ -0,0 +1,3 @@ +#!/bin/bash +# description: system-wide r/w activity +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rw-by-pid.pl diff --git a/tools/perf/scripts/perl/bin/rwtop-record b/tools/perf/scripts/perl/bin/rwtop-record new file mode 100644 index 00000000..7cb9db23 --- /dev/null +++ b/tools/perf/scripts/perl/bin/rwtop-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@ diff --git a/tools/perf/scripts/perl/bin/rwtop-report b/tools/perf/scripts/perl/bin/rwtop-report new file mode 100644 index 00000000..83e11ec2 --- /dev/null +++ b/tools/perf/scripts/perl/bin/rwtop-report @@ -0,0 +1,20 @@ +#!/bin/bash +# description: system-wide r/w top +# args: [interval] +n_args=0 +for i in "$@" +do + if expr match "$i" "-" > /dev/null ; then + break + fi + n_args=$(( $n_args + 1 )) +done +if [ "$n_args" -gt 1 ] ; then + echo "usage: rwtop-report [interval]" + exit +fi +if [ "$n_args" -gt 0 ] ; then + interval=$1 + shift +fi +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rwtop.pl $interval diff --git a/tools/perf/scripts/perl/bin/wakeup-latency-record b/tools/perf/scripts/perl/bin/wakeup-latency-record new file mode 100644 index 00000000..464251a1 --- /dev/null +++ b/tools/perf/scripts/perl/bin/wakeup-latency-record @@ -0,0 +1,6 @@ +#!/bin/bash +perf record -e sched:sched_switch -e sched:sched_wakeup $@ + + + + diff --git a/tools/perf/scripts/perl/bin/wakeup-latency-report b/tools/perf/scripts/perl/bin/wakeup-latency-report new file mode 100644 index 00000000..889e8130 --- /dev/null +++ b/tools/perf/scripts/perl/bin/wakeup-latency-report @@ -0,0 +1,3 @@ +#!/bin/bash +# description: system-wide min/max/avg wakeup latency +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/wakeup-latency.pl diff --git a/tools/perf/scripts/perl/check-perf-trace.pl b/tools/perf/scripts/perl/check-perf-trace.pl new file mode 100644 index 00000000..4e7076c2 --- /dev/null +++ b/tools/perf/scripts/perl/check-perf-trace.pl @@ -0,0 +1,106 @@ +# perf script event handlers, generated by perf script -g perl +# (c) 2009, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 + +# This script tests basic functionality such as flag and symbol +# strings, common_xxx() calls back into perf, begin, end, unhandled +# events, etc. Basically, if this script runs successfully and +# displays expected results, perl scripting support should be ok. + +use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib"; +use lib "./Perf-Trace-Util/lib"; +use Perf::Trace::Core; +use Perf::Trace::Context; +use Perf::Trace::Util; + +sub trace_begin +{ + print "trace_begin\n"; +} + +sub trace_end +{ + print "trace_end\n"; + + print_unhandled(); +} + +sub irq::softirq_entry +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $vec) = @_; + + print_header($event_name, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm); + + print_uncommon($context); + + printf("vec=%s\n", + symbol_str("irq::softirq_entry", "vec", $vec)); +} + +sub kmem::kmalloc +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $call_site, $ptr, $bytes_req, $bytes_alloc, + $gfp_flags) = @_; + + print_header($event_name, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm); + + print_uncommon($context); + + printf("call_site=%p, ptr=%p, bytes_req=%u, bytes_alloc=%u, ". + "gfp_flags=%s\n", + $call_site, $ptr, $bytes_req, $bytes_alloc, + + flag_str("kmem::kmalloc", "gfp_flags", $gfp_flags)); +} + +# print trace fields not included in handler args +sub print_uncommon +{ + my ($context) = @_; + + printf("common_preempt_count=%d, common_flags=%s, common_lock_depth=%d, ", + common_pc($context), trace_flag_str(common_flags($context)), + common_lock_depth($context)); + +} + +my %unhandled; + +sub print_unhandled +{ + if ((scalar keys %unhandled) == 0) { + return; + } + + print "\nunhandled events:\n\n"; + + printf("%-40s %10s\n", "event", "count"); + printf("%-40s %10s\n", "----------------------------------------", + "-----------"); + + foreach my $event_name (keys %unhandled) { + printf("%-40s %10d\n", $event_name, $unhandled{$event_name}); + } +} + +sub trace_unhandled +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm) = @_; + + $unhandled{$event_name}++; +} + +sub print_header +{ + my ($event_name, $cpu, $secs, $nsecs, $pid, $comm) = @_; + + printf("%-20s %5u %05u.%09u %8u %-20s ", + $event_name, $cpu, $secs, $nsecs, $pid, $comm); +} diff --git a/tools/perf/scripts/perl/failed-syscalls.pl b/tools/perf/scripts/perl/failed-syscalls.pl new file mode 100644 index 00000000..94bc25a3 --- /dev/null +++ b/tools/perf/scripts/perl/failed-syscalls.pl @@ -0,0 +1,42 @@ +# failed system call counts +# (c) 2010, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 +# +# Displays system-wide failed system call totals +# If a [comm] arg is specified, only syscalls called by [comm] are displayed. + +use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib"; +use lib "./Perf-Trace-Util/lib"; +use Perf::Trace::Core; +use Perf::Trace::Context; +use Perf::Trace::Util; + +my $for_comm = shift; + +my %failed_syscalls; + +sub raw_syscalls::sys_exit +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $id, $ret) = @_; + + if ($ret < 0) { + $failed_syscalls{$common_comm}++; + } +} + +sub trace_end +{ + printf("\nfailed syscalls by comm:\n\n"); + + printf("%-20s %10s\n", "comm", "# errors"); + printf("%-20s %6s %10s\n", "--------------------", "----------"); + + foreach my $comm (sort {$failed_syscalls{$b} <=> $failed_syscalls{$a}} + keys %failed_syscalls) { + next if ($for_comm && $comm ne $for_comm); + + printf("%-20s %10s\n", $comm, $failed_syscalls{$comm}); + } +} diff --git a/tools/perf/scripts/perl/rw-by-file.pl b/tools/perf/scripts/perl/rw-by-file.pl new file mode 100644 index 00000000..74844ee2 --- /dev/null +++ b/tools/perf/scripts/perl/rw-by-file.pl @@ -0,0 +1,106 @@ +#!/usr/bin/perl -w +# (c) 2009, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 + +# Display r/w activity for files read/written to for a given program + +# The common_* event handler fields are the most useful fields common to +# all events. They don't necessarily correspond to the 'common_*' fields +# in the status files. Those fields not available as handler params can +# be retrieved via script functions of the form get_common_*(). + +use 5.010000; +use strict; +use warnings; + +use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib"; +use lib "./Perf-Trace-Util/lib"; +use Perf::Trace::Core; +use Perf::Trace::Util; + +my $usage = "perf script -s rw-by-file.pl <comm>\n"; + +my $for_comm = shift or die $usage; + +my %reads; +my %writes; + +sub syscalls::sys_enter_read +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, $nr, $fd, $buf, $count) = @_; + + if ($common_comm eq $for_comm) { + $reads{$fd}{bytes_requested} += $count; + $reads{$fd}{total_reads}++; + } +} + +sub syscalls::sys_enter_write +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, $nr, $fd, $buf, $count) = @_; + + if ($common_comm eq $for_comm) { + $writes{$fd}{bytes_written} += $count; + $writes{$fd}{total_writes}++; + } +} + +sub trace_end +{ + printf("file read counts for $for_comm:\n\n"); + + printf("%6s %10s %10s\n", "fd", "# reads", "bytes_requested"); + printf("%6s %10s %10s\n", "------", "----------", "-----------"); + + foreach my $fd (sort {$reads{$b}{bytes_requested} <=> + $reads{$a}{bytes_requested}} keys %reads) { + my $total_reads = $reads{$fd}{total_reads}; + my $bytes_requested = $reads{$fd}{bytes_requested}; + printf("%6u %10u %10u\n", $fd, $total_reads, $bytes_requested); + } + + printf("\nfile write counts for $for_comm:\n\n"); + + printf("%6s %10s %10s\n", "fd", "# writes", "bytes_written"); + printf("%6s %10s %10s\n", "------", "----------", "-----------"); + + foreach my $fd (sort {$writes{$b}{bytes_written} <=> + $writes{$a}{bytes_written}} keys %writes) { + my $total_writes = $writes{$fd}{total_writes}; + my $bytes_written = $writes{$fd}{bytes_written}; + printf("%6u %10u %10u\n", $fd, $total_writes, $bytes_written); + } + + print_unhandled(); +} + +my %unhandled; + +sub print_unhandled +{ + if ((scalar keys %unhandled) == 0) { + return; + } + + print "\nunhandled events:\n\n"; + + printf("%-40s %10s\n", "event", "count"); + printf("%-40s %10s\n", "----------------------------------------", + "-----------"); + + foreach my $event_name (keys %unhandled) { + printf("%-40s %10d\n", $event_name, $unhandled{$event_name}); + } +} + +sub trace_unhandled +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm) = @_; + + $unhandled{$event_name}++; +} + + diff --git a/tools/perf/scripts/perl/rw-by-pid.pl b/tools/perf/scripts/perl/rw-by-pid.pl new file mode 100644 index 00000000..9db23c9d --- /dev/null +++ b/tools/perf/scripts/perl/rw-by-pid.pl @@ -0,0 +1,184 @@ +#!/usr/bin/perl -w +# (c) 2009, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 + +# Display r/w activity for all processes + +# The common_* event handler fields are the most useful fields common to +# all events. They don't necessarily correspond to the 'common_*' fields +# in the status files. Those fields not available as handler params can +# be retrieved via script functions of the form get_common_*(). + +use 5.010000; +use strict; +use warnings; + +use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib"; +use lib "./Perf-Trace-Util/lib"; +use Perf::Trace::Core; +use Perf::Trace::Util; + +my %reads; +my %writes; + +sub syscalls::sys_exit_read +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $ret) = @_; + + if ($ret > 0) { + $reads{$common_pid}{bytes_read} += $ret; + } else { + if (!defined ($reads{$common_pid}{bytes_read})) { + $reads{$common_pid}{bytes_read} = 0; + } + $reads{$common_pid}{errors}{$ret}++; + } +} + +sub syscalls::sys_enter_read +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $fd, $buf, $count) = @_; + + $reads{$common_pid}{bytes_requested} += $count; + $reads{$common_pid}{total_reads}++; + $reads{$common_pid}{comm} = $common_comm; +} + +sub syscalls::sys_exit_write +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $ret) = @_; + + if ($ret <= 0) { + $writes{$common_pid}{errors}{$ret}++; + } +} + +sub syscalls::sys_enter_write +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $fd, $buf, $count) = @_; + + $writes{$common_pid}{bytes_written} += $count; + $writes{$common_pid}{total_writes}++; + $writes{$common_pid}{comm} = $common_comm; +} + +sub trace_end +{ + printf("read counts by pid:\n\n"); + + printf("%6s %20s %10s %10s %10s\n", "pid", "comm", + "# reads", "bytes_requested", "bytes_read"); + printf("%6s %-20s %10s %10s %10s\n", "------", "--------------------", + "-----------", "----------", "----------"); + + foreach my $pid (sort { ($reads{$b}{bytes_read} || 0) <=> + ($reads{$a}{bytes_read} || 0) } keys %reads) { + my $comm = $reads{$pid}{comm} || ""; + my $total_reads = $reads{$pid}{total_reads} || 0; + my $bytes_requested = $reads{$pid}{bytes_requested} || 0; + my $bytes_read = $reads{$pid}{bytes_read} || 0; + + printf("%6s %-20s %10s %10s %10s\n", $pid, $comm, + $total_reads, $bytes_requested, $bytes_read); + } + + printf("\nfailed reads by pid:\n\n"); + + printf("%6s %20s %6s %10s\n", "pid", "comm", "error #", "# errors"); + printf("%6s %20s %6s %10s\n", "------", "--------------------", + "------", "----------"); + + my @errcounts = (); + + foreach my $pid (keys %reads) { + foreach my $error (keys %{$reads{$pid}{errors}}) { + my $comm = $reads{$pid}{comm} || ""; + my $errcount = $reads{$pid}{errors}{$error} || 0; + push @errcounts, [$pid, $comm, $error, $errcount]; + } + } + + @errcounts = sort { $b->[3] <=> $a->[3] } @errcounts; + + for my $i (0 .. $#errcounts) { + printf("%6d %-20s %6d %10s\n", $errcounts[$i][0], + $errcounts[$i][1], $errcounts[$i][2], $errcounts[$i][3]); + } + + printf("\nwrite counts by pid:\n\n"); + + printf("%6s %20s %10s %10s\n", "pid", "comm", + "# writes", "bytes_written"); + printf("%6s %-20s %10s %10s\n", "------", "--------------------", + "-----------", "----------"); + + foreach my $pid (sort { ($writes{$b}{bytes_written} || 0) <=> + ($writes{$a}{bytes_written} || 0)} keys %writes) { + my $comm = $writes{$pid}{comm} || ""; + my $total_writes = $writes{$pid}{total_writes} || 0; + my $bytes_written = $writes{$pid}{bytes_written} || 0; + + printf("%6s %-20s %10s %10s\n", $pid, $comm, + $total_writes, $bytes_written); + } + + printf("\nfailed writes by pid:\n\n"); + + printf("%6s %20s %6s %10s\n", "pid", "comm", "error #", "# errors"); + printf("%6s %20s %6s %10s\n", "------", "--------------------", + "------", "----------"); + + @errcounts = (); + + foreach my $pid (keys %writes) { + foreach my $error (keys %{$writes{$pid}{errors}}) { + my $comm = $writes{$pid}{comm} || ""; + my $errcount = $writes{$pid}{errors}{$error} || 0; + push @errcounts, [$pid, $comm, $error, $errcount]; + } + } + + @errcounts = sort { $b->[3] <=> $a->[3] } @errcounts; + + for my $i (0 .. $#errcounts) { + printf("%6d %-20s %6d %10s\n", $errcounts[$i][0], + $errcounts[$i][1], $errcounts[$i][2], $errcounts[$i][3]); + } + + print_unhandled(); +} + +my %unhandled; + +sub print_unhandled +{ + if ((scalar keys %unhandled) == 0) { + return; + } + + print "\nunhandled events:\n\n"; + + printf("%-40s %10s\n", "event", "count"); + printf("%-40s %10s\n", "----------------------------------------", + "-----------"); + + foreach my $event_name (keys %unhandled) { + printf("%-40s %10d\n", $event_name, $unhandled{$event_name}); + } +} + +sub trace_unhandled +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm) = @_; + + $unhandled{$event_name}++; +} diff --git a/tools/perf/scripts/perl/rwtop.pl b/tools/perf/scripts/perl/rwtop.pl new file mode 100644 index 00000000..8b207870 --- /dev/null +++ b/tools/perf/scripts/perl/rwtop.pl @@ -0,0 +1,203 @@ +#!/usr/bin/perl -w +# (c) 2010, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 + +# read/write top +# +# Periodically displays system-wide r/w call activity, broken down by +# pid. If an [interval] arg is specified, the display will be +# refreshed every [interval] seconds. The default interval is 3 +# seconds. + +use 5.010000; +use strict; +use warnings; + +use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib"; +use lib "./Perf-Trace-Util/lib"; +use Perf::Trace::Core; +use Perf::Trace::Util; +use POSIX qw/SIGALRM SA_RESTART/; + +my $default_interval = 3; +my $nlines = 20; +my $print_thread; +my $print_pending = 0; + +my %reads; +my %writes; + +my $interval = shift; +if (!$interval) { + $interval = $default_interval; +} + +sub syscalls::sys_exit_read +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $ret) = @_; + + print_check(); + + if ($ret > 0) { + $reads{$common_pid}{bytes_read} += $ret; + } else { + if (!defined ($reads{$common_pid}{bytes_read})) { + $reads{$common_pid}{bytes_read} = 0; + } + $reads{$common_pid}{errors}{$ret}++; + } +} + +sub syscalls::sys_enter_read +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $fd, $buf, $count) = @_; + + print_check(); + + $reads{$common_pid}{bytes_requested} += $count; + $reads{$common_pid}{total_reads}++; + $reads{$common_pid}{comm} = $common_comm; +} + +sub syscalls::sys_exit_write +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $ret) = @_; + + print_check(); + + if ($ret <= 0) { + $writes{$common_pid}{errors}{$ret}++; + } +} + +sub syscalls::sys_enter_write +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $nr, $fd, $buf, $count) = @_; + + print_check(); + + $writes{$common_pid}{bytes_written} += $count; + $writes{$common_pid}{total_writes}++; + $writes{$common_pid}{comm} = $common_comm; +} + +sub trace_begin +{ + my $sa = POSIX::SigAction->new(\&set_print_pending); + $sa->flags(SA_RESTART); + $sa->safe(1); + POSIX::sigaction(SIGALRM, $sa) or die "Can't set SIGALRM handler: $!\n"; + alarm 1; +} + +sub trace_end +{ + print_unhandled(); + print_totals(); +} + +sub print_check() +{ + if ($print_pending == 1) { + $print_pending = 0; + print_totals(); + } +} + +sub set_print_pending() +{ + $print_pending = 1; + alarm $interval; +} + +sub print_totals +{ + my $count; + + $count = 0; + + clear_term(); + + printf("\nread counts by pid:\n\n"); + + printf("%6s %20s %10s %10s %10s\n", "pid", "comm", + "# reads", "bytes_req", "bytes_read"); + printf("%6s %-20s %10s %10s %10s\n", "------", "--------------------", + "----------", "----------", "----------"); + + foreach my $pid (sort { ($reads{$b}{bytes_read} || 0) <=> + ($reads{$a}{bytes_read} || 0) } keys %reads) { + my $comm = $reads{$pid}{comm} || ""; + my $total_reads = $reads{$pid}{total_reads} || 0; + my $bytes_requested = $reads{$pid}{bytes_requested} || 0; + my $bytes_read = $reads{$pid}{bytes_read} || 0; + + printf("%6s %-20s %10s %10s %10s\n", $pid, $comm, + $total_reads, $bytes_requested, $bytes_read); + + if (++$count == $nlines) { + last; + } + } + + $count = 0; + + printf("\nwrite counts by pid:\n\n"); + + printf("%6s %20s %10s %13s\n", "pid", "comm", + "# writes", "bytes_written"); + printf("%6s %-20s %10s %13s\n", "------", "--------------------", + "----------", "-------------"); + + foreach my $pid (sort { ($writes{$b}{bytes_written} || 0) <=> + ($writes{$a}{bytes_written} || 0)} keys %writes) { + my $comm = $writes{$pid}{comm} || ""; + my $total_writes = $writes{$pid}{total_writes} || 0; + my $bytes_written = $writes{$pid}{bytes_written} || 0; + + printf("%6s %-20s %10s %13s\n", $pid, $comm, + $total_writes, $bytes_written); + + if (++$count == $nlines) { + last; + } + } + + %reads = (); + %writes = (); +} + +my %unhandled; + +sub print_unhandled +{ + if ((scalar keys %unhandled) == 0) { + return; + } + + print "\nunhandled events:\n\n"; + + printf("%-40s %10s\n", "event", "count"); + printf("%-40s %10s\n", "----------------------------------------", + "-----------"); + + foreach my $event_name (keys %unhandled) { + printf("%-40s %10d\n", $event_name, $unhandled{$event_name}); + } +} + +sub trace_unhandled +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm) = @_; + + $unhandled{$event_name}++; +} diff --git a/tools/perf/scripts/perl/wakeup-latency.pl b/tools/perf/scripts/perl/wakeup-latency.pl new file mode 100644 index 00000000..d9143dce --- /dev/null +++ b/tools/perf/scripts/perl/wakeup-latency.pl @@ -0,0 +1,107 @@ +#!/usr/bin/perl -w +# (c) 2009, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 + +# Display avg/min/max wakeup latency + +# The common_* event handler fields are the most useful fields common to +# all events. They don't necessarily correspond to the 'common_*' fields +# in the status files. Those fields not available as handler params can +# be retrieved via script functions of the form get_common_*(). + +use 5.010000; +use strict; +use warnings; + +use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib"; +use lib "./Perf-Trace-Util/lib"; +use Perf::Trace::Core; +use Perf::Trace::Util; + +my %last_wakeup; + +my $max_wakeup_latency; +my $min_wakeup_latency; +my $total_wakeup_latency = 0; +my $total_wakeups = 0; + +sub sched::sched_switch +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $prev_comm, $prev_pid, $prev_prio, $prev_state, $next_comm, $next_pid, + $next_prio) = @_; + + my $wakeup_ts = $last_wakeup{$common_cpu}{ts}; + if ($wakeup_ts) { + my $switch_ts = nsecs($common_secs, $common_nsecs); + my $wakeup_latency = $switch_ts - $wakeup_ts; + if ($wakeup_latency > $max_wakeup_latency) { + $max_wakeup_latency = $wakeup_latency; + } + if ($wakeup_latency < $min_wakeup_latency) { + $min_wakeup_latency = $wakeup_latency; + } + $total_wakeup_latency += $wakeup_latency; + $total_wakeups++; + } + $last_wakeup{$common_cpu}{ts} = 0; +} + +sub sched::sched_wakeup +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm, + $comm, $pid, $prio, $success, $target_cpu) = @_; + + $last_wakeup{$target_cpu}{ts} = nsecs($common_secs, $common_nsecs); +} + +sub trace_begin +{ + $min_wakeup_latency = 1000000000; + $max_wakeup_latency = 0; +} + +sub trace_end +{ + printf("wakeup_latency stats:\n\n"); + print "total_wakeups: $total_wakeups\n"; + if ($total_wakeups) { + printf("avg_wakeup_latency (ns): %u\n", + avg($total_wakeup_latency, $total_wakeups)); + } else { + printf("avg_wakeup_latency (ns): N/A\n"); + } + printf("min_wakeup_latency (ns): %u\n", $min_wakeup_latency); + printf("max_wakeup_latency (ns): %u\n", $max_wakeup_latency); + + print_unhandled(); +} + +my %unhandled; + +sub print_unhandled +{ + if ((scalar keys %unhandled) == 0) { + return; + } + + print "\nunhandled events:\n\n"; + + printf("%-40s %10s\n", "event", "count"); + printf("%-40s %10s\n", "----------------------------------------", + "-----------"); + + foreach my $event_name (keys %unhandled) { + printf("%-40s %10d\n", $event_name, $unhandled{$event_name}); + } +} + +sub trace_unhandled +{ + my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs, + $common_pid, $common_comm) = @_; + + $unhandled{$event_name}++; +} diff --git a/tools/perf/scripts/python/Perf-Trace-Util/Context.c b/tools/perf/scripts/python/Perf-Trace-Util/Context.c new file mode 100644 index 00000000..315067b8 --- /dev/null +++ b/tools/perf/scripts/python/Perf-Trace-Util/Context.c @@ -0,0 +1,88 @@ +/* + * Context.c. Python interfaces for perf script. + * + * Copyright (C) 2010 Tom Zanussi <tzanussi@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <Python.h> +#include "../../../perf.h" +#include "../../../util/trace-event.h" + +PyMODINIT_FUNC initperf_trace_context(void); + +static PyObject *perf_trace_context_common_pc(PyObject *self, PyObject *args) +{ + static struct scripting_context *scripting_context; + PyObject *context; + int retval; + + if (!PyArg_ParseTuple(args, "O", &context)) + return NULL; + + scripting_context = PyCObject_AsVoidPtr(context); + retval = common_pc(scripting_context); + + return Py_BuildValue("i", retval); +} + +static PyObject *perf_trace_context_common_flags(PyObject *self, + PyObject *args) +{ + static struct scripting_context *scripting_context; + PyObject *context; + int retval; + + if (!PyArg_ParseTuple(args, "O", &context)) + return NULL; + + scripting_context = PyCObject_AsVoidPtr(context); + retval = common_flags(scripting_context); + + return Py_BuildValue("i", retval); +} + +static PyObject *perf_trace_context_common_lock_depth(PyObject *self, + PyObject *args) +{ + static struct scripting_context *scripting_context; + PyObject *context; + int retval; + + if (!PyArg_ParseTuple(args, "O", &context)) + return NULL; + + scripting_context = PyCObject_AsVoidPtr(context); + retval = common_lock_depth(scripting_context); + + return Py_BuildValue("i", retval); +} + +static PyMethodDef ContextMethods[] = { + { "common_pc", perf_trace_context_common_pc, METH_VARARGS, + "Get the common preempt count event field value."}, + { "common_flags", perf_trace_context_common_flags, METH_VARARGS, + "Get the common flags event field value."}, + { "common_lock_depth", perf_trace_context_common_lock_depth, + METH_VARARGS, "Get the common lock depth event field value."}, + { NULL, NULL, 0, NULL} +}; + +PyMODINIT_FUNC initperf_trace_context(void) +{ + (void) Py_InitModule("perf_trace_context", ContextMethods); +} diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py new file mode 100644 index 00000000..de7211e4 --- /dev/null +++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py @@ -0,0 +1,121 @@ +# Core.py - Python extension for perf script, core functions +# +# Copyright (C) 2010 by Tom Zanussi <tzanussi@gmail.com> +# +# This software may be distributed under the terms of the GNU General +# Public License ("GPL") version 2 as published by the Free Software +# Foundation. + +from collections import defaultdict + +def autodict(): + return defaultdict(autodict) + +flag_fields = autodict() +symbolic_fields = autodict() + +def define_flag_field(event_name, field_name, delim): + flag_fields[event_name][field_name]['delim'] = delim + +def define_flag_value(event_name, field_name, value, field_str): + flag_fields[event_name][field_name]['values'][value] = field_str + +def define_symbolic_field(event_name, field_name): + # nothing to do, really + pass + +def define_symbolic_value(event_name, field_name, value, field_str): + symbolic_fields[event_name][field_name]['values'][value] = field_str + +def flag_str(event_name, field_name, value): + string = "" + + if flag_fields[event_name][field_name]: + print_delim = 0 + keys = flag_fields[event_name][field_name]['values'].keys() + keys.sort() + for idx in keys: + if not value and not idx: + string += flag_fields[event_name][field_name]['values'][idx] + break + if idx and (value & idx) == idx: + if print_delim and flag_fields[event_name][field_name]['delim']: + string += " " + flag_fields[event_name][field_name]['delim'] + " " + string += flag_fields[event_name][field_name]['values'][idx] + print_delim = 1 + value &= ~idx + + return string + +def symbol_str(event_name, field_name, value): + string = "" + + if symbolic_fields[event_name][field_name]: + keys = symbolic_fields[event_name][field_name]['values'].keys() + keys.sort() + for idx in keys: + if not value and not idx: + string = symbolic_fields[event_name][field_name]['values'][idx] + break + if (value == idx): + string = symbolic_fields[event_name][field_name]['values'][idx] + break + + return string + +trace_flags = { 0x00: "NONE", \ + 0x01: "IRQS_OFF", \ + 0x02: "IRQS_NOSUPPORT", \ + 0x04: "NEED_RESCHED", \ + 0x08: "HARDIRQ", \ + 0x10: "SOFTIRQ" } + +def trace_flag_str(value): + string = "" + print_delim = 0 + + keys = trace_flags.keys() + + for idx in keys: + if not value and not idx: + string += "NONE" + break + + if idx and (value & idx) == idx: + if print_delim: + string += " | "; + string += trace_flags[idx] + print_delim = 1 + value &= ~idx + + return string + + +def taskState(state): + states = { + 0 : "R", + 1 : "S", + 2 : "D", + 64: "DEAD" + } + + if state not in states: + return "Unknown" + + return states[state] + + +class EventHeaders: + def __init__(self, common_cpu, common_secs, common_nsecs, + common_pid, common_comm): + self.cpu = common_cpu + self.secs = common_secs + self.nsecs = common_nsecs + self.pid = common_pid + self.comm = common_comm + + def ts(self): + return (self.secs * (10 ** 9)) + self.nsecs + + def ts_format(self): + return "%d.%d" % (self.secs, int(self.nsecs / 1000)) diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/EventClass.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/EventClass.py new file mode 100755 index 00000000..9e098579 --- /dev/null +++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/EventClass.py @@ -0,0 +1,94 @@ +# EventClass.py +# +# This is a library defining some events types classes, which could +# be used by other scripts to analyzing the perf samples. +# +# Currently there are just a few classes defined for examples, +# PerfEvent is the base class for all perf event sample, PebsEvent +# is a HW base Intel x86 PEBS event, and user could add more SW/HW +# event classes based on requirements. + +import struct + +# Event types, user could add more here +EVTYPE_GENERIC = 0 +EVTYPE_PEBS = 1 # Basic PEBS event +EVTYPE_PEBS_LL = 2 # PEBS event with load latency info +EVTYPE_IBS = 3 + +# +# Currently we don't have good way to tell the event type, but by +# the size of raw buffer, raw PEBS event with load latency data's +# size is 176 bytes, while the pure PEBS event's size is 144 bytes. +# +def create_event(name, comm, dso, symbol, raw_buf): + if (len(raw_buf) == 144): + event = PebsEvent(name, comm, dso, symbol, raw_buf) + elif (len(raw_buf) == 176): + event = PebsNHM(name, comm, dso, symbol, raw_buf) + else: + event = PerfEvent(name, comm, dso, symbol, raw_buf) + + return event + +class PerfEvent(object): + event_num = 0 + def __init__(self, name, comm, dso, symbol, raw_buf, ev_type=EVTYPE_GENERIC): + self.name = name + self.comm = comm + self.dso = dso + self.symbol = symbol + self.raw_buf = raw_buf + self.ev_type = ev_type + PerfEvent.event_num += 1 + + def show(self): + print "PMU event: name=%12s, symbol=%24s, comm=%8s, dso=%12s" % (self.name, self.symbol, self.comm, self.dso) + +# +# Basic Intel PEBS (Precise Event-based Sampling) event, whose raw buffer +# contains the context info when that event happened: the EFLAGS and +# linear IP info, as well as all the registers. +# +class PebsEvent(PerfEvent): + pebs_num = 0 + def __init__(self, name, comm, dso, symbol, raw_buf, ev_type=EVTYPE_PEBS): + tmp_buf=raw_buf[0:80] + flags, ip, ax, bx, cx, dx, si, di, bp, sp = struct.unpack('QQQQQQQQQQ', tmp_buf) + self.flags = flags + self.ip = ip + self.ax = ax + self.bx = bx + self.cx = cx + self.dx = dx + self.si = si + self.di = di + self.bp = bp + self.sp = sp + + PerfEvent.__init__(self, name, comm, dso, symbol, raw_buf, ev_type) + PebsEvent.pebs_num += 1 + del tmp_buf + +# +# Intel Nehalem and Westmere support PEBS plus Load Latency info which lie +# in the four 64 bit words write after the PEBS data: +# Status: records the IA32_PERF_GLOBAL_STATUS register value +# DLA: Data Linear Address (EIP) +# DSE: Data Source Encoding, where the latency happens, hit or miss +# in L1/L2/L3 or IO operations +# LAT: the actual latency in cycles +# +class PebsNHM(PebsEvent): + pebs_nhm_num = 0 + def __init__(self, name, comm, dso, symbol, raw_buf, ev_type=EVTYPE_PEBS_LL): + tmp_buf=raw_buf[144:176] + status, dla, dse, lat = struct.unpack('QQQQ', tmp_buf) + self.status = status + self.dla = dla + self.dse = dse + self.lat = lat + + PebsEvent.__init__(self, name, comm, dso, symbol, raw_buf, ev_type) + PebsNHM.pebs_nhm_num += 1 + del tmp_buf diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py new file mode 100644 index 00000000..fdd92f69 --- /dev/null +++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py @@ -0,0 +1,184 @@ +# SchedGui.py - Python extension for perf script, basic GUI code for +# traces drawing and overview. +# +# Copyright (C) 2010 by Frederic Weisbecker <fweisbec@gmail.com> +# +# This software is distributed under the terms of the GNU General +# Public License ("GPL") version 2 as published by the Free Software +# Foundation. + + +try: + import wx +except ImportError: + raise ImportError, "You need to install the wxpython lib for this script" + + +class RootFrame(wx.Frame): + Y_OFFSET = 100 + RECT_HEIGHT = 100 + RECT_SPACE = 50 + EVENT_MARKING_WIDTH = 5 + + def __init__(self, sched_tracer, title, parent = None, id = -1): + wx.Frame.__init__(self, parent, id, title) + + (self.screen_width, self.screen_height) = wx.GetDisplaySize() + self.screen_width -= 10 + self.screen_height -= 10 + self.zoom = 0.5 + self.scroll_scale = 20 + self.sched_tracer = sched_tracer + self.sched_tracer.set_root_win(self) + (self.ts_start, self.ts_end) = sched_tracer.interval() + self.update_width_virtual() + self.nr_rects = sched_tracer.nr_rectangles() + 1 + self.height_virtual = RootFrame.Y_OFFSET + (self.nr_rects * (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE)) + + # whole window panel + self.panel = wx.Panel(self, size=(self.screen_width, self.screen_height)) + + # scrollable container + self.scroll = wx.ScrolledWindow(self.panel) + self.scroll.SetScrollbars(self.scroll_scale, self.scroll_scale, self.width_virtual / self.scroll_scale, self.height_virtual / self.scroll_scale) + self.scroll.EnableScrolling(True, True) + self.scroll.SetFocus() + + # scrollable drawing area + self.scroll_panel = wx.Panel(self.scroll, size=(self.screen_width - 15, self.screen_height / 2)) + self.scroll_panel.Bind(wx.EVT_PAINT, self.on_paint) + self.scroll_panel.Bind(wx.EVT_KEY_DOWN, self.on_key_press) + self.scroll_panel.Bind(wx.EVT_LEFT_DOWN, self.on_mouse_down) + self.scroll.Bind(wx.EVT_PAINT, self.on_paint) + self.scroll.Bind(wx.EVT_KEY_DOWN, self.on_key_press) + self.scroll.Bind(wx.EVT_LEFT_DOWN, self.on_mouse_down) + + self.scroll.Fit() + self.Fit() + + self.scroll_panel.SetDimensions(-1, -1, self.width_virtual, self.height_virtual, wx.SIZE_USE_EXISTING) + + self.txt = None + + self.Show(True) + + def us_to_px(self, val): + return val / (10 ** 3) * self.zoom + + def px_to_us(self, val): + return (val / self.zoom) * (10 ** 3) + + def scroll_start(self): + (x, y) = self.scroll.GetViewStart() + return (x * self.scroll_scale, y * self.scroll_scale) + + def scroll_start_us(self): + (x, y) = self.scroll_start() + return self.px_to_us(x) + + def paint_rectangle_zone(self, nr, color, top_color, start, end): + offset_px = self.us_to_px(start - self.ts_start) + width_px = self.us_to_px(end - self.ts_start) + + offset_py = RootFrame.Y_OFFSET + (nr * (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE)) + width_py = RootFrame.RECT_HEIGHT + + dc = self.dc + + if top_color is not None: + (r, g, b) = top_color + top_color = wx.Colour(r, g, b) + brush = wx.Brush(top_color, wx.SOLID) + dc.SetBrush(brush) + dc.DrawRectangle(offset_px, offset_py, width_px, RootFrame.EVENT_MARKING_WIDTH) + width_py -= RootFrame.EVENT_MARKING_WIDTH + offset_py += RootFrame.EVENT_MARKING_WIDTH + + (r ,g, b) = color + color = wx.Colour(r, g, b) + brush = wx.Brush(color, wx.SOLID) + dc.SetBrush(brush) + dc.DrawRectangle(offset_px, offset_py, width_px, width_py) + + def update_rectangles(self, dc, start, end): + start += self.ts_start + end += self.ts_start + self.sched_tracer.fill_zone(start, end) + + def on_paint(self, event): + dc = wx.PaintDC(self.scroll_panel) + self.dc = dc + + width = min(self.width_virtual, self.screen_width) + (x, y) = self.scroll_start() + start = self.px_to_us(x) + end = self.px_to_us(x + width) + self.update_rectangles(dc, start, end) + + def rect_from_ypixel(self, y): + y -= RootFrame.Y_OFFSET + rect = y / (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE) + height = y % (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE) + + if rect < 0 or rect > self.nr_rects - 1 or height > RootFrame.RECT_HEIGHT: + return -1 + + return rect + + def update_summary(self, txt): + if self.txt: + self.txt.Destroy() + self.txt = wx.StaticText(self.panel, -1, txt, (0, (self.screen_height / 2) + 50)) + + + def on_mouse_down(self, event): + (x, y) = event.GetPositionTuple() + rect = self.rect_from_ypixel(y) + if rect == -1: + return + + t = self.px_to_us(x) + self.ts_start + + self.sched_tracer.mouse_down(rect, t) + + + def update_width_virtual(self): + self.width_virtual = self.us_to_px(self.ts_end - self.ts_start) + + def __zoom(self, x): + self.update_width_virtual() + (xpos, ypos) = self.scroll.GetViewStart() + xpos = self.us_to_px(x) / self.scroll_scale + self.scroll.SetScrollbars(self.scroll_scale, self.scroll_scale, self.width_virtual / self.scroll_scale, self.height_virtual / self.scroll_scale, xpos, ypos) + self.Refresh() + + def zoom_in(self): + x = self.scroll_start_us() + self.zoom *= 2 + self.__zoom(x) + + def zoom_out(self): + x = self.scroll_start_us() + self.zoom /= 2 + self.__zoom(x) + + + def on_key_press(self, event): + key = event.GetRawKeyCode() + if key == ord("+"): + self.zoom_in() + return + if key == ord("-"): + self.zoom_out() + return + + key = event.GetKeyCode() + (x, y) = self.scroll.GetViewStart() + if key == wx.WXK_RIGHT: + self.scroll.Scroll(x + 1, y) + elif key == wx.WXK_LEFT: + self.scroll.Scroll(x - 1, y) + elif key == wx.WXK_DOWN: + self.scroll.Scroll(x, y + 1) + elif key == wx.WXK_UP: + self.scroll.Scroll(x, y - 1) diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py new file mode 100644 index 00000000..15c84002 --- /dev/null +++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py @@ -0,0 +1,86 @@ +# Util.py - Python extension for perf script, miscellaneous utility code +# +# Copyright (C) 2010 by Tom Zanussi <tzanussi@gmail.com> +# +# This software may be distributed under the terms of the GNU General +# Public License ("GPL") version 2 as published by the Free Software +# Foundation. + +import errno, os + +FUTEX_WAIT = 0 +FUTEX_WAKE = 1 +FUTEX_PRIVATE_FLAG = 128 +FUTEX_CLOCK_REALTIME = 256 +FUTEX_CMD_MASK = ~(FUTEX_PRIVATE_FLAG | FUTEX_CLOCK_REALTIME) + +NSECS_PER_SEC = 1000000000 + +def avg(total, n): + return total / n + +def nsecs(secs, nsecs): + return secs * NSECS_PER_SEC + nsecs + +def nsecs_secs(nsecs): + return nsecs / NSECS_PER_SEC + +def nsecs_nsecs(nsecs): + return nsecs % NSECS_PER_SEC + +def nsecs_str(nsecs): + str = "%5u.%09u" % (nsecs_secs(nsecs), nsecs_nsecs(nsecs)), + return str + +def add_stats(dict, key, value): + if not dict.has_key(key): + dict[key] = (value, value, value, 1) + else: + min, max, avg, count = dict[key] + if value < min: + min = value + if value > max: + max = value + avg = (avg + value) / 2 + dict[key] = (min, max, avg, count + 1) + +def clear_term(): + print("\x1b[H\x1b[2J") + +audit_package_warned = False + +try: + import audit + machine_to_id = { + 'x86_64': audit.MACH_86_64, + 'alpha' : audit.MACH_ALPHA, + 'ia64' : audit.MACH_IA64, + 'ppc' : audit.MACH_PPC, + 'ppc64' : audit.MACH_PPC64, + 's390' : audit.MACH_S390, + 's390x' : audit.MACH_S390X, + 'i386' : audit.MACH_X86, + 'i586' : audit.MACH_X86, + 'i686' : audit.MACH_X86, + } + try: + machine_to_id['armeb'] = audit.MACH_ARMEB + except: + pass + machine_id = machine_to_id[os.uname()[4]] +except: + if not audit_package_warned: + audit_package_warned = True + print "Install the audit-libs-python package to get syscall names" + +def syscall_name(id): + try: + return audit.audit_syscall_to_name(id, machine_id) + except: + return str(id) + +def strerror(nr): + try: + return errno.errorcode[abs(nr)] + except: + return "Unknown %d errno" % nr diff --git a/tools/perf/scripts/python/bin/event_analyzing_sample-record b/tools/perf/scripts/python/bin/event_analyzing_sample-record new file mode 100644 index 00000000..5ce652da --- /dev/null +++ b/tools/perf/scripts/python/bin/event_analyzing_sample-record @@ -0,0 +1,8 @@ +#!/bin/bash + +# +# event_analyzing_sample.py can cover all type of perf samples including +# the tracepoints, so no special record requirements, just record what +# you want to analyze. +# +perf record $@ diff --git a/tools/perf/scripts/python/bin/event_analyzing_sample-report b/tools/perf/scripts/python/bin/event_analyzing_sample-report new file mode 100644 index 00000000..0941fc94 --- /dev/null +++ b/tools/perf/scripts/python/bin/event_analyzing_sample-report @@ -0,0 +1,3 @@ +#!/bin/bash +# description: analyze all perf samples +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/event_analyzing_sample.py diff --git a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record new file mode 100644 index 00000000..8104895a --- /dev/null +++ b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e raw_syscalls:sys_exit $@ diff --git a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report new file mode 100644 index 00000000..fda5096d --- /dev/null +++ b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report @@ -0,0 +1,10 @@ +#!/bin/bash +# description: system-wide failed syscalls, by pid +# args: [comm] +if [ $# -gt 0 ] ; then + if ! expr match "$1" "-" > /dev/null ; then + comm=$1 + shift + fi +fi +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/failed-syscalls-by-pid.py $comm diff --git a/tools/perf/scripts/python/bin/futex-contention-record b/tools/perf/scripts/python/bin/futex-contention-record new file mode 100644 index 00000000..b1495c9a --- /dev/null +++ b/tools/perf/scripts/python/bin/futex-contention-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e syscalls:sys_enter_futex -e syscalls:sys_exit_futex $@ diff --git a/tools/perf/scripts/python/bin/futex-contention-report b/tools/perf/scripts/python/bin/futex-contention-report new file mode 100644 index 00000000..6c442710 --- /dev/null +++ b/tools/perf/scripts/python/bin/futex-contention-report @@ -0,0 +1,4 @@ +#!/bin/bash +# description: futext contention measurement + +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/futex-contention.py diff --git a/tools/perf/scripts/python/bin/net_dropmonitor-record b/tools/perf/scripts/python/bin/net_dropmonitor-record new file mode 100755 index 00000000..423fb81d --- /dev/null +++ b/tools/perf/scripts/python/bin/net_dropmonitor-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e skb:kfree_skb $@ diff --git a/tools/perf/scripts/python/bin/net_dropmonitor-report b/tools/perf/scripts/python/bin/net_dropmonitor-report new file mode 100755 index 00000000..8d698f5a --- /dev/null +++ b/tools/perf/scripts/python/bin/net_dropmonitor-report @@ -0,0 +1,4 @@ +#!/bin/bash +# description: display a table of dropped frames + +perf script -s "$PERF_EXEC_PATH"/scripts/python/net_dropmonitor.py $@ diff --git a/tools/perf/scripts/python/bin/netdev-times-record b/tools/perf/scripts/python/bin/netdev-times-record new file mode 100644 index 00000000..558754b8 --- /dev/null +++ b/tools/perf/scripts/python/bin/netdev-times-record @@ -0,0 +1,8 @@ +#!/bin/bash +perf record -e net:net_dev_xmit -e net:net_dev_queue \ + -e net:netif_receive_skb -e net:netif_rx \ + -e skb:consume_skb -e skb:kfree_skb \ + -e skb:skb_copy_datagram_iovec -e napi:napi_poll \ + -e irq:irq_handler_entry -e irq:irq_handler_exit \ + -e irq:softirq_entry -e irq:softirq_exit \ + -e irq:softirq_raise $@ diff --git a/tools/perf/scripts/python/bin/netdev-times-report b/tools/perf/scripts/python/bin/netdev-times-report new file mode 100644 index 00000000..8f759291 --- /dev/null +++ b/tools/perf/scripts/python/bin/netdev-times-report @@ -0,0 +1,5 @@ +#!/bin/bash +# description: display a process of packet and processing time +# args: [tx] [rx] [dev=] [debug] + +perf script -s "$PERF_EXEC_PATH"/scripts/python/netdev-times.py $@ diff --git a/tools/perf/scripts/python/bin/sched-migration-record b/tools/perf/scripts/python/bin/sched-migration-record new file mode 100644 index 00000000..7493fddb --- /dev/null +++ b/tools/perf/scripts/python/bin/sched-migration-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -m 16384 -e sched:sched_wakeup -e sched:sched_wakeup_new -e sched:sched_switch -e sched:sched_migrate_task $@ diff --git a/tools/perf/scripts/python/bin/sched-migration-report b/tools/perf/scripts/python/bin/sched-migration-report new file mode 100644 index 00000000..68b037a1 --- /dev/null +++ b/tools/perf/scripts/python/bin/sched-migration-report @@ -0,0 +1,3 @@ +#!/bin/bash +# description: sched migration overview +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/sched-migration.py diff --git a/tools/perf/scripts/python/bin/sctop-record b/tools/perf/scripts/python/bin/sctop-record new file mode 100644 index 00000000..4efbfaa7 --- /dev/null +++ b/tools/perf/scripts/python/bin/sctop-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e raw_syscalls:sys_enter $@ diff --git a/tools/perf/scripts/python/bin/sctop-report b/tools/perf/scripts/python/bin/sctop-report new file mode 100644 index 00000000..c32db294 --- /dev/null +++ b/tools/perf/scripts/python/bin/sctop-report @@ -0,0 +1,24 @@ +#!/bin/bash +# description: syscall top +# args: [comm] [interval] +n_args=0 +for i in "$@" +do + if expr match "$i" "-" > /dev/null ; then + break + fi + n_args=$(( $n_args + 1 )) +done +if [ "$n_args" -gt 2 ] ; then + echo "usage: sctop-report [comm] [interval]" + exit +fi +if [ "$n_args" -gt 1 ] ; then + comm=$1 + interval=$2 + shift 2 +elif [ "$n_args" -gt 0 ] ; then + interval=$1 + shift +fi +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/sctop.py $comm $interval diff --git a/tools/perf/scripts/python/bin/syscall-counts-by-pid-record b/tools/perf/scripts/python/bin/syscall-counts-by-pid-record new file mode 100644 index 00000000..4efbfaa7 --- /dev/null +++ b/tools/perf/scripts/python/bin/syscall-counts-by-pid-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e raw_syscalls:sys_enter $@ diff --git a/tools/perf/scripts/python/bin/syscall-counts-by-pid-report b/tools/perf/scripts/python/bin/syscall-counts-by-pid-report new file mode 100644 index 00000000..16eb8d65 --- /dev/null +++ b/tools/perf/scripts/python/bin/syscall-counts-by-pid-report @@ -0,0 +1,10 @@ +#!/bin/bash +# description: system-wide syscall counts, by pid +# args: [comm] +if [ $# -gt 0 ] ; then + if ! expr match "$1" "-" > /dev/null ; then + comm=$1 + shift + fi +fi +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/syscall-counts-by-pid.py $comm diff --git a/tools/perf/scripts/python/bin/syscall-counts-record b/tools/perf/scripts/python/bin/syscall-counts-record new file mode 100644 index 00000000..4efbfaa7 --- /dev/null +++ b/tools/perf/scripts/python/bin/syscall-counts-record @@ -0,0 +1,2 @@ +#!/bin/bash +perf record -e raw_syscalls:sys_enter $@ diff --git a/tools/perf/scripts/python/bin/syscall-counts-report b/tools/perf/scripts/python/bin/syscall-counts-report new file mode 100644 index 00000000..0f0e9d45 --- /dev/null +++ b/tools/perf/scripts/python/bin/syscall-counts-report @@ -0,0 +1,10 @@ +#!/bin/bash +# description: system-wide syscall counts +# args: [comm] +if [ $# -gt 0 ] ; then + if ! expr match "$1" "-" > /dev/null ; then + comm=$1 + shift + fi +fi +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/syscall-counts.py $comm diff --git a/tools/perf/scripts/python/check-perf-trace.py b/tools/perf/scripts/python/check-perf-trace.py new file mode 100644 index 00000000..4647a769 --- /dev/null +++ b/tools/perf/scripts/python/check-perf-trace.py @@ -0,0 +1,82 @@ +# perf script event handlers, generated by perf script -g python +# (c) 2010, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 +# +# This script tests basic functionality such as flag and symbol +# strings, common_xxx() calls back into perf, begin, end, unhandled +# events, etc. Basically, if this script runs successfully and +# displays expected results, Python scripting support should be ok. + +import os +import sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from Core import * +from perf_trace_context import * + +unhandled = autodict() + +def trace_begin(): + print "trace_begin" + pass + +def trace_end(): + print_unhandled() + +def irq__softirq_entry(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + vec): + print_header(event_name, common_cpu, common_secs, common_nsecs, + common_pid, common_comm) + + print_uncommon(context) + + print "vec=%s\n" % \ + (symbol_str("irq__softirq_entry", "vec", vec)), + +def kmem__kmalloc(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + call_site, ptr, bytes_req, bytes_alloc, + gfp_flags): + print_header(event_name, common_cpu, common_secs, common_nsecs, + common_pid, common_comm) + + print_uncommon(context) + + print "call_site=%u, ptr=%u, bytes_req=%u, " \ + "bytes_alloc=%u, gfp_flags=%s\n" % \ + (call_site, ptr, bytes_req, bytes_alloc, + + flag_str("kmem__kmalloc", "gfp_flags", gfp_flags)), + +def trace_unhandled(event_name, context, event_fields_dict): + try: + unhandled[event_name] += 1 + except TypeError: + unhandled[event_name] = 1 + +def print_header(event_name, cpu, secs, nsecs, pid, comm): + print "%-20s %5u %05u.%09u %8u %-20s " % \ + (event_name, cpu, secs, nsecs, pid, comm), + +# print trace fields not included in handler args +def print_uncommon(context): + print "common_preempt_count=%d, common_flags=%s, common_lock_depth=%d, " \ + % (common_pc(context), trace_flag_str(common_flags(context)), \ + common_lock_depth(context)) + +def print_unhandled(): + keys = unhandled.keys() + if not keys: + return + + print "\nunhandled events:\n\n", + + print "%-40s %10s\n" % ("event", "count"), + print "%-40s %10s\n" % ("----------------------------------------", \ + "-----------"), + + for event_name in keys: + print "%-40s %10d\n" % (event_name, unhandled[event_name]) diff --git a/tools/perf/scripts/python/event_analyzing_sample.py b/tools/perf/scripts/python/event_analyzing_sample.py new file mode 100644 index 00000000..163c39fa --- /dev/null +++ b/tools/perf/scripts/python/event_analyzing_sample.py @@ -0,0 +1,189 @@ +# event_analyzing_sample.py: general event handler in python +# +# Current perf report is already very powerful with the annotation integrated, +# and this script is not trying to be as powerful as perf report, but +# providing end user/developer a flexible way to analyze the events other +# than trace points. +# +# The 2 database related functions in this script just show how to gather +# the basic information, and users can modify and write their own functions +# according to their specific requirement. +# +# The first function "show_general_events" just does a basic grouping for all +# generic events with the help of sqlite, and the 2nd one "show_pebs_ll" is +# for a x86 HW PMU event: PEBS with load latency data. +# + +import os +import sys +import math +import struct +import sqlite3 + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from EventClass import * + +# +# If the perf.data has a big number of samples, then the insert operation +# will be very time consuming (about 10+ minutes for 10000 samples) if the +# .db database is on disk. Move the .db file to RAM based FS to speedup +# the handling, which will cut the time down to several seconds. +# +con = sqlite3.connect("/dev/shm/perf.db") +con.isolation_level = None + +def trace_begin(): + print "In trace_begin:\n" + + # + # Will create several tables at the start, pebs_ll is for PEBS data with + # load latency info, while gen_events is for general event. + # + con.execute(""" + create table if not exists gen_events ( + name text, + symbol text, + comm text, + dso text + );""") + con.execute(""" + create table if not exists pebs_ll ( + name text, + symbol text, + comm text, + dso text, + flags integer, + ip integer, + status integer, + dse integer, + dla integer, + lat integer + );""") + +# +# Create and insert event object to a database so that user could +# do more analysis with simple database commands. +# +def process_event(param_dict): + event_attr = param_dict["attr"] + sample = param_dict["sample"] + raw_buf = param_dict["raw_buf"] + comm = param_dict["comm"] + name = param_dict["ev_name"] + + # Symbol and dso info are not always resolved + if (param_dict.has_key("dso")): + dso = param_dict["dso"] + else: + dso = "Unknown_dso" + + if (param_dict.has_key("symbol")): + symbol = param_dict["symbol"] + else: + symbol = "Unknown_symbol" + + # Create the event object and insert it to the right table in database + event = create_event(name, comm, dso, symbol, raw_buf) + insert_db(event) + +def insert_db(event): + if event.ev_type == EVTYPE_GENERIC: + con.execute("insert into gen_events values(?, ?, ?, ?)", + (event.name, event.symbol, event.comm, event.dso)) + elif event.ev_type == EVTYPE_PEBS_LL: + event.ip &= 0x7fffffffffffffff + event.dla &= 0x7fffffffffffffff + con.execute("insert into pebs_ll values (?, ?, ?, ?, ?, ?, ?, ?, ?, ?)", + (event.name, event.symbol, event.comm, event.dso, event.flags, + event.ip, event.status, event.dse, event.dla, event.lat)) + +def trace_end(): + print "In trace_end:\n" + # We show the basic info for the 2 type of event classes + show_general_events() + show_pebs_ll() + con.close() + +# +# As the event number may be very big, so we can't use linear way +# to show the histogram in real number, but use a log2 algorithm. +# + +def num2sym(num): + # Each number will have at least one '#' + snum = '#' * (int)(math.log(num, 2) + 1) + return snum + +def show_general_events(): + + # Check the total record number in the table + count = con.execute("select count(*) from gen_events") + for t in count: + print "There is %d records in gen_events table" % t[0] + if t[0] == 0: + return + + print "Statistics about the general events grouped by thread/symbol/dso: \n" + + # Group by thread + commq = con.execute("select comm, count(comm) from gen_events group by comm order by -count(comm)") + print "\n%16s %8s %16s\n%s" % ("comm", "number", "histogram", "="*42) + for row in commq: + print "%16s %8d %s" % (row[0], row[1], num2sym(row[1])) + + # Group by symbol + print "\n%32s %8s %16s\n%s" % ("symbol", "number", "histogram", "="*58) + symbolq = con.execute("select symbol, count(symbol) from gen_events group by symbol order by -count(symbol)") + for row in symbolq: + print "%32s %8d %s" % (row[0], row[1], num2sym(row[1])) + + # Group by dso + print "\n%40s %8s %16s\n%s" % ("dso", "number", "histogram", "="*74) + dsoq = con.execute("select dso, count(dso) from gen_events group by dso order by -count(dso)") + for row in dsoq: + print "%40s %8d %s" % (row[0], row[1], num2sym(row[1])) + +# +# This function just shows the basic info, and we could do more with the +# data in the tables, like checking the function parameters when some +# big latency events happen. +# +def show_pebs_ll(): + + count = con.execute("select count(*) from pebs_ll") + for t in count: + print "There is %d records in pebs_ll table" % t[0] + if t[0] == 0: + return + + print "Statistics about the PEBS Load Latency events grouped by thread/symbol/dse/latency: \n" + + # Group by thread + commq = con.execute("select comm, count(comm) from pebs_ll group by comm order by -count(comm)") + print "\n%16s %8s %16s\n%s" % ("comm", "number", "histogram", "="*42) + for row in commq: + print "%16s %8d %s" % (row[0], row[1], num2sym(row[1])) + + # Group by symbol + print "\n%32s %8s %16s\n%s" % ("symbol", "number", "histogram", "="*58) + symbolq = con.execute("select symbol, count(symbol) from pebs_ll group by symbol order by -count(symbol)") + for row in symbolq: + print "%32s %8d %s" % (row[0], row[1], num2sym(row[1])) + + # Group by dse + dseq = con.execute("select dse, count(dse) from pebs_ll group by dse order by -count(dse)") + print "\n%32s %8s %16s\n%s" % ("dse", "number", "histogram", "="*58) + for row in dseq: + print "%32s %8d %s" % (row[0], row[1], num2sym(row[1])) + + # Group by latency + latq = con.execute("select lat, count(lat) from pebs_ll group by lat order by lat") + print "\n%32s %8s %16s\n%s" % ("latency", "number", "histogram", "="*58) + for row in latq: + print "%32s %8d %s" % (row[0], row[1], num2sym(row[1])) + +def trace_unhandled(event_name, context, event_fields_dict): + print ' '.join(['%s=%s'%(k,str(v))for k,v in sorted(event_fields_dict.items())]) diff --git a/tools/perf/scripts/python/failed-syscalls-by-pid.py b/tools/perf/scripts/python/failed-syscalls-by-pid.py new file mode 100644 index 00000000..85805fac --- /dev/null +++ b/tools/perf/scripts/python/failed-syscalls-by-pid.py @@ -0,0 +1,73 @@ +# failed system call counts, by pid +# (c) 2010, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 +# +# Displays system-wide failed system call totals, broken down by pid. +# If a [comm] arg is specified, only syscalls called by [comm] are displayed. + +import os +import sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * +from Util import * + +usage = "perf script -s syscall-counts-by-pid.py [comm|pid]\n"; + +for_comm = None +for_pid = None + +if len(sys.argv) > 2: + sys.exit(usage) + +if len(sys.argv) > 1: + try: + for_pid = int(sys.argv[1]) + except: + for_comm = sys.argv[1] + +syscalls = autodict() + +def trace_begin(): + print "Press control+C to stop and show the summary" + +def trace_end(): + print_error_totals() + +def raw_syscalls__sys_exit(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + id, ret): + if (for_comm and common_comm != for_comm) or \ + (for_pid and common_pid != for_pid ): + return + + if ret < 0: + try: + syscalls[common_comm][common_pid][id][ret] += 1 + except TypeError: + syscalls[common_comm][common_pid][id][ret] = 1 + +def print_error_totals(): + if for_comm is not None: + print "\nsyscall errors for %s:\n\n" % (for_comm), + else: + print "\nsyscall errors:\n\n", + + print "%-30s %10s\n" % ("comm [pid]", "count"), + print "%-30s %10s\n" % ("------------------------------", \ + "----------"), + + comm_keys = syscalls.keys() + for comm in comm_keys: + pid_keys = syscalls[comm].keys() + for pid in pid_keys: + print "\n%s [%d]\n" % (comm, pid), + id_keys = syscalls[comm][pid].keys() + for id in id_keys: + print " syscall: %-16s\n" % syscall_name(id), + ret_keys = syscalls[comm][pid][id].keys() + for ret, val in sorted(syscalls[comm][pid][id].iteritems(), key = lambda(k, v): (v, k), reverse = True): + print " err = %-20s %10d\n" % (strerror(ret), val), diff --git a/tools/perf/scripts/python/futex-contention.py b/tools/perf/scripts/python/futex-contention.py new file mode 100644 index 00000000..11e70a38 --- /dev/null +++ b/tools/perf/scripts/python/futex-contention.py @@ -0,0 +1,50 @@ +# futex contention +# (c) 2010, Arnaldo Carvalho de Melo <acme@redhat.com> +# Licensed under the terms of the GNU GPL License version 2 +# +# Translation of: +# +# http://sourceware.org/systemtap/wiki/WSFutexContention +# +# to perf python scripting. +# +# Measures futex contention + +import os, sys +sys.path.append(os.environ['PERF_EXEC_PATH'] + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') +from Util import * + +process_names = {} +thread_thislock = {} +thread_blocktime = {} + +lock_waits = {} # long-lived stats on (tid,lock) blockage elapsed time +process_names = {} # long-lived pid-to-execname mapping + +def syscalls__sys_enter_futex(event, ctxt, cpu, s, ns, tid, comm, + nr, uaddr, op, val, utime, uaddr2, val3): + cmd = op & FUTEX_CMD_MASK + if cmd != FUTEX_WAIT: + return # we don't care about originators of WAKE events + + process_names[tid] = comm + thread_thislock[tid] = uaddr + thread_blocktime[tid] = nsecs(s, ns) + +def syscalls__sys_exit_futex(event, ctxt, cpu, s, ns, tid, comm, + nr, ret): + if thread_blocktime.has_key(tid): + elapsed = nsecs(s, ns) - thread_blocktime[tid] + add_stats(lock_waits, (tid, thread_thislock[tid]), elapsed) + del thread_blocktime[tid] + del thread_thislock[tid] + +def trace_begin(): + print "Press control+C to stop and show the summary" + +def trace_end(): + for (tid, lock) in lock_waits: + min, max, avg, count = lock_waits[tid, lock] + print "%s[%d] lock %x contended %d times, %d avg ns" % \ + (process_names[tid], tid, lock, count, avg) + diff --git a/tools/perf/scripts/python/net_dropmonitor.py b/tools/perf/scripts/python/net_dropmonitor.py new file mode 100755 index 00000000..a4ffc950 --- /dev/null +++ b/tools/perf/scripts/python/net_dropmonitor.py @@ -0,0 +1,72 @@ +# Monitor the system for dropped packets and proudce a report of drop locations and counts + +import os +import sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * +from Util import * + +drop_log = {} +kallsyms = [] + +def get_kallsyms_table(): + global kallsyms + try: + f = open("/proc/kallsyms", "r") + linecount = 0 + for line in f: + linecount = linecount+1 + f.seek(0) + except: + return + + + j = 0 + for line in f: + loc = int(line.split()[0], 16) + name = line.split()[2] + j = j +1 + if ((j % 100) == 0): + print "\r" + str(j) + "/" + str(linecount), + kallsyms.append({ 'loc': loc, 'name' : name}) + + print "\r" + str(j) + "/" + str(linecount) + kallsyms.sort() + return + +def get_sym(sloc): + loc = int(sloc) + for i in kallsyms: + if (i['loc'] >= loc): + return (i['name'], i['loc']-loc) + return (None, 0) + +def print_drop_table(): + print "%25s %25s %25s" % ("LOCATION", "OFFSET", "COUNT") + for i in drop_log.keys(): + (sym, off) = get_sym(i) + if sym == None: + sym = i + print "%25s %25s %25s" % (sym, off, drop_log[i]) + + +def trace_begin(): + print "Starting trace (Ctrl-C to dump results)" + +def trace_end(): + print "Gathering kallsyms data" + get_kallsyms_table() + print_drop_table() + +# called from perf, when it finds a correspoinding event +def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm, + skbaddr, protocol, location): + slocation = str(location) + try: + drop_log[slocation] = drop_log[slocation] + 1 + except: + drop_log[slocation] = 1 diff --git a/tools/perf/scripts/python/netdev-times.py b/tools/perf/scripts/python/netdev-times.py new file mode 100644 index 00000000..9aa0a329 --- /dev/null +++ b/tools/perf/scripts/python/netdev-times.py @@ -0,0 +1,464 @@ +# Display a process of packets and processed time. +# It helps us to investigate networking or network device. +# +# options +# tx: show only tx chart +# rx: show only rx chart +# dev=: show only thing related to specified device +# debug: work with debug mode. It shows buffer status. + +import os +import sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * +from Util import * + +all_event_list = []; # insert all tracepoint event related with this script +irq_dic = {}; # key is cpu and value is a list which stacks irqs + # which raise NET_RX softirq +net_rx_dic = {}; # key is cpu and value include time of NET_RX softirq-entry + # and a list which stacks receive +receive_hunk_list = []; # a list which include a sequence of receive events +rx_skb_list = []; # received packet list for matching + # skb_copy_datagram_iovec + +buffer_budget = 65536; # the budget of rx_skb_list, tx_queue_list and + # tx_xmit_list +of_count_rx_skb_list = 0; # overflow count + +tx_queue_list = []; # list of packets which pass through dev_queue_xmit +of_count_tx_queue_list = 0; # overflow count + +tx_xmit_list = []; # list of packets which pass through dev_hard_start_xmit +of_count_tx_xmit_list = 0; # overflow count + +tx_free_list = []; # list of packets which is freed + +# options +show_tx = 0; +show_rx = 0; +dev = 0; # store a name of device specified by option "dev=" +debug = 0; + +# indices of event_info tuple +EINFO_IDX_NAME= 0 +EINFO_IDX_CONTEXT=1 +EINFO_IDX_CPU= 2 +EINFO_IDX_TIME= 3 +EINFO_IDX_PID= 4 +EINFO_IDX_COMM= 5 + +# Calculate a time interval(msec) from src(nsec) to dst(nsec) +def diff_msec(src, dst): + return (dst - src) / 1000000.0 + +# Display a process of transmitting a packet +def print_transmit(hunk): + if dev != 0 and hunk['dev'].find(dev) < 0: + return + print "%7s %5d %6d.%06dsec %12.3fmsec %12.3fmsec" % \ + (hunk['dev'], hunk['len'], + nsecs_secs(hunk['queue_t']), + nsecs_nsecs(hunk['queue_t'])/1000, + diff_msec(hunk['queue_t'], hunk['xmit_t']), + diff_msec(hunk['xmit_t'], hunk['free_t'])) + +# Format for displaying rx packet processing +PF_IRQ_ENTRY= " irq_entry(+%.3fmsec irq=%d:%s)" +PF_SOFT_ENTRY=" softirq_entry(+%.3fmsec)" +PF_NAPI_POLL= " napi_poll_exit(+%.3fmsec %s)" +PF_JOINT= " |" +PF_WJOINT= " | |" +PF_NET_RECV= " |---netif_receive_skb(+%.3fmsec skb=%x len=%d)" +PF_NET_RX= " |---netif_rx(+%.3fmsec skb=%x)" +PF_CPY_DGRAM= " | skb_copy_datagram_iovec(+%.3fmsec %d:%s)" +PF_KFREE_SKB= " | kfree_skb(+%.3fmsec location=%x)" +PF_CONS_SKB= " | consume_skb(+%.3fmsec)" + +# Display a process of received packets and interrputs associated with +# a NET_RX softirq +def print_receive(hunk): + show_hunk = 0 + irq_list = hunk['irq_list'] + cpu = irq_list[0]['cpu'] + base_t = irq_list[0]['irq_ent_t'] + # check if this hunk should be showed + if dev != 0: + for i in range(len(irq_list)): + if irq_list[i]['name'].find(dev) >= 0: + show_hunk = 1 + break + else: + show_hunk = 1 + if show_hunk == 0: + return + + print "%d.%06dsec cpu=%d" % \ + (nsecs_secs(base_t), nsecs_nsecs(base_t)/1000, cpu) + for i in range(len(irq_list)): + print PF_IRQ_ENTRY % \ + (diff_msec(base_t, irq_list[i]['irq_ent_t']), + irq_list[i]['irq'], irq_list[i]['name']) + print PF_JOINT + irq_event_list = irq_list[i]['event_list'] + for j in range(len(irq_event_list)): + irq_event = irq_event_list[j] + if irq_event['event'] == 'netif_rx': + print PF_NET_RX % \ + (diff_msec(base_t, irq_event['time']), + irq_event['skbaddr']) + print PF_JOINT + print PF_SOFT_ENTRY % \ + diff_msec(base_t, hunk['sirq_ent_t']) + print PF_JOINT + event_list = hunk['event_list'] + for i in range(len(event_list)): + event = event_list[i] + if event['event_name'] == 'napi_poll': + print PF_NAPI_POLL % \ + (diff_msec(base_t, event['event_t']), event['dev']) + if i == len(event_list) - 1: + print "" + else: + print PF_JOINT + else: + print PF_NET_RECV % \ + (diff_msec(base_t, event['event_t']), event['skbaddr'], + event['len']) + if 'comm' in event.keys(): + print PF_WJOINT + print PF_CPY_DGRAM % \ + (diff_msec(base_t, event['comm_t']), + event['pid'], event['comm']) + elif 'handle' in event.keys(): + print PF_WJOINT + if event['handle'] == "kfree_skb": + print PF_KFREE_SKB % \ + (diff_msec(base_t, + event['comm_t']), + event['location']) + elif event['handle'] == "consume_skb": + print PF_CONS_SKB % \ + diff_msec(base_t, + event['comm_t']) + print PF_JOINT + +def trace_begin(): + global show_tx + global show_rx + global dev + global debug + + for i in range(len(sys.argv)): + if i == 0: + continue + arg = sys.argv[i] + if arg == 'tx': + show_tx = 1 + elif arg =='rx': + show_rx = 1 + elif arg.find('dev=',0, 4) >= 0: + dev = arg[4:] + elif arg == 'debug': + debug = 1 + if show_tx == 0 and show_rx == 0: + show_tx = 1 + show_rx = 1 + +def trace_end(): + # order all events in time + all_event_list.sort(lambda a,b :cmp(a[EINFO_IDX_TIME], + b[EINFO_IDX_TIME])) + # process all events + for i in range(len(all_event_list)): + event_info = all_event_list[i] + name = event_info[EINFO_IDX_NAME] + if name == 'irq__softirq_exit': + handle_irq_softirq_exit(event_info) + elif name == 'irq__softirq_entry': + handle_irq_softirq_entry(event_info) + elif name == 'irq__softirq_raise': + handle_irq_softirq_raise(event_info) + elif name == 'irq__irq_handler_entry': + handle_irq_handler_entry(event_info) + elif name == 'irq__irq_handler_exit': + handle_irq_handler_exit(event_info) + elif name == 'napi__napi_poll': + handle_napi_poll(event_info) + elif name == 'net__netif_receive_skb': + handle_netif_receive_skb(event_info) + elif name == 'net__netif_rx': + handle_netif_rx(event_info) + elif name == 'skb__skb_copy_datagram_iovec': + handle_skb_copy_datagram_iovec(event_info) + elif name == 'net__net_dev_queue': + handle_net_dev_queue(event_info) + elif name == 'net__net_dev_xmit': + handle_net_dev_xmit(event_info) + elif name == 'skb__kfree_skb': + handle_kfree_skb(event_info) + elif name == 'skb__consume_skb': + handle_consume_skb(event_info) + # display receive hunks + if show_rx: + for i in range(len(receive_hunk_list)): + print_receive(receive_hunk_list[i]) + # display transmit hunks + if show_tx: + print " dev len Qdisc " \ + " netdevice free" + for i in range(len(tx_free_list)): + print_transmit(tx_free_list[i]) + if debug: + print "debug buffer status" + print "----------------------------" + print "xmit Qdisc:remain:%d overflow:%d" % \ + (len(tx_queue_list), of_count_tx_queue_list) + print "xmit netdevice:remain:%d overflow:%d" % \ + (len(tx_xmit_list), of_count_tx_xmit_list) + print "receive:remain:%d overflow:%d" % \ + (len(rx_skb_list), of_count_rx_skb_list) + +# called from perf, when it finds a correspoinding event +def irq__softirq_entry(name, context, cpu, sec, nsec, pid, comm, vec): + if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX": + return + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec) + all_event_list.append(event_info) + +def irq__softirq_exit(name, context, cpu, sec, nsec, pid, comm, vec): + if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX": + return + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec) + all_event_list.append(event_info) + +def irq__softirq_raise(name, context, cpu, sec, nsec, pid, comm, vec): + if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX": + return + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec) + all_event_list.append(event_info) + +def irq__irq_handler_entry(name, context, cpu, sec, nsec, pid, comm, + irq, irq_name): + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, + irq, irq_name) + all_event_list.append(event_info) + +def irq__irq_handler_exit(name, context, cpu, sec, nsec, pid, comm, irq, ret): + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, irq, ret) + all_event_list.append(event_info) + +def napi__napi_poll(name, context, cpu, sec, nsec, pid, comm, napi, dev_name): + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, + napi, dev_name) + all_event_list.append(event_info) + +def net__netif_receive_skb(name, context, cpu, sec, nsec, pid, comm, skbaddr, + skblen, dev_name): + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, + skbaddr, skblen, dev_name) + all_event_list.append(event_info) + +def net__netif_rx(name, context, cpu, sec, nsec, pid, comm, skbaddr, + skblen, dev_name): + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, + skbaddr, skblen, dev_name) + all_event_list.append(event_info) + +def net__net_dev_queue(name, context, cpu, sec, nsec, pid, comm, + skbaddr, skblen, dev_name): + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, + skbaddr, skblen, dev_name) + all_event_list.append(event_info) + +def net__net_dev_xmit(name, context, cpu, sec, nsec, pid, comm, + skbaddr, skblen, rc, dev_name): + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, + skbaddr, skblen, rc ,dev_name) + all_event_list.append(event_info) + +def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm, + skbaddr, protocol, location): + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, + skbaddr, protocol, location) + all_event_list.append(event_info) + +def skb__consume_skb(name, context, cpu, sec, nsec, pid, comm, skbaddr): + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, + skbaddr) + all_event_list.append(event_info) + +def skb__skb_copy_datagram_iovec(name, context, cpu, sec, nsec, pid, comm, + skbaddr, skblen): + event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, + skbaddr, skblen) + all_event_list.append(event_info) + +def handle_irq_handler_entry(event_info): + (name, context, cpu, time, pid, comm, irq, irq_name) = event_info + if cpu not in irq_dic.keys(): + irq_dic[cpu] = [] + irq_record = {'irq':irq, 'name':irq_name, 'cpu':cpu, 'irq_ent_t':time} + irq_dic[cpu].append(irq_record) + +def handle_irq_handler_exit(event_info): + (name, context, cpu, time, pid, comm, irq, ret) = event_info + if cpu not in irq_dic.keys(): + return + irq_record = irq_dic[cpu].pop() + if irq != irq_record['irq']: + return + irq_record.update({'irq_ext_t':time}) + # if an irq doesn't include NET_RX softirq, drop. + if 'event_list' in irq_record.keys(): + irq_dic[cpu].append(irq_record) + +def handle_irq_softirq_raise(event_info): + (name, context, cpu, time, pid, comm, vec) = event_info + if cpu not in irq_dic.keys() \ + or len(irq_dic[cpu]) == 0: + return + irq_record = irq_dic[cpu].pop() + if 'event_list' in irq_record.keys(): + irq_event_list = irq_record['event_list'] + else: + irq_event_list = [] + irq_event_list.append({'time':time, 'event':'sirq_raise'}) + irq_record.update({'event_list':irq_event_list}) + irq_dic[cpu].append(irq_record) + +def handle_irq_softirq_entry(event_info): + (name, context, cpu, time, pid, comm, vec) = event_info + net_rx_dic[cpu] = {'sirq_ent_t':time, 'event_list':[]} + +def handle_irq_softirq_exit(event_info): + (name, context, cpu, time, pid, comm, vec) = event_info + irq_list = [] + event_list = 0 + if cpu in irq_dic.keys(): + irq_list = irq_dic[cpu] + del irq_dic[cpu] + if cpu in net_rx_dic.keys(): + sirq_ent_t = net_rx_dic[cpu]['sirq_ent_t'] + event_list = net_rx_dic[cpu]['event_list'] + del net_rx_dic[cpu] + if irq_list == [] or event_list == 0: + return + rec_data = {'sirq_ent_t':sirq_ent_t, 'sirq_ext_t':time, + 'irq_list':irq_list, 'event_list':event_list} + # merge information realted to a NET_RX softirq + receive_hunk_list.append(rec_data) + +def handle_napi_poll(event_info): + (name, context, cpu, time, pid, comm, napi, dev_name) = event_info + if cpu in net_rx_dic.keys(): + event_list = net_rx_dic[cpu]['event_list'] + rec_data = {'event_name':'napi_poll', + 'dev':dev_name, 'event_t':time} + event_list.append(rec_data) + +def handle_netif_rx(event_info): + (name, context, cpu, time, pid, comm, + skbaddr, skblen, dev_name) = event_info + if cpu not in irq_dic.keys() \ + or len(irq_dic[cpu]) == 0: + return + irq_record = irq_dic[cpu].pop() + if 'event_list' in irq_record.keys(): + irq_event_list = irq_record['event_list'] + else: + irq_event_list = [] + irq_event_list.append({'time':time, 'event':'netif_rx', + 'skbaddr':skbaddr, 'skblen':skblen, 'dev_name':dev_name}) + irq_record.update({'event_list':irq_event_list}) + irq_dic[cpu].append(irq_record) + +def handle_netif_receive_skb(event_info): + global of_count_rx_skb_list + + (name, context, cpu, time, pid, comm, + skbaddr, skblen, dev_name) = event_info + if cpu in net_rx_dic.keys(): + rec_data = {'event_name':'netif_receive_skb', + 'event_t':time, 'skbaddr':skbaddr, 'len':skblen} + event_list = net_rx_dic[cpu]['event_list'] + event_list.append(rec_data) + rx_skb_list.insert(0, rec_data) + if len(rx_skb_list) > buffer_budget: + rx_skb_list.pop() + of_count_rx_skb_list += 1 + +def handle_net_dev_queue(event_info): + global of_count_tx_queue_list + + (name, context, cpu, time, pid, comm, + skbaddr, skblen, dev_name) = event_info + skb = {'dev':dev_name, 'skbaddr':skbaddr, 'len':skblen, 'queue_t':time} + tx_queue_list.insert(0, skb) + if len(tx_queue_list) > buffer_budget: + tx_queue_list.pop() + of_count_tx_queue_list += 1 + +def handle_net_dev_xmit(event_info): + global of_count_tx_xmit_list + + (name, context, cpu, time, pid, comm, + skbaddr, skblen, rc, dev_name) = event_info + if rc == 0: # NETDEV_TX_OK + for i in range(len(tx_queue_list)): + skb = tx_queue_list[i] + if skb['skbaddr'] == skbaddr: + skb['xmit_t'] = time + tx_xmit_list.insert(0, skb) + del tx_queue_list[i] + if len(tx_xmit_list) > buffer_budget: + tx_xmit_list.pop() + of_count_tx_xmit_list += 1 + return + +def handle_kfree_skb(event_info): + (name, context, cpu, time, pid, comm, + skbaddr, protocol, location) = event_info + for i in range(len(tx_queue_list)): + skb = tx_queue_list[i] + if skb['skbaddr'] == skbaddr: + del tx_queue_list[i] + return + for i in range(len(tx_xmit_list)): + skb = tx_xmit_list[i] + if skb['skbaddr'] == skbaddr: + skb['free_t'] = time + tx_free_list.append(skb) + del tx_xmit_list[i] + return + for i in range(len(rx_skb_list)): + rec_data = rx_skb_list[i] + if rec_data['skbaddr'] == skbaddr: + rec_data.update({'handle':"kfree_skb", + 'comm':comm, 'pid':pid, 'comm_t':time}) + del rx_skb_list[i] + return + +def handle_consume_skb(event_info): + (name, context, cpu, time, pid, comm, skbaddr) = event_info + for i in range(len(tx_xmit_list)): + skb = tx_xmit_list[i] + if skb['skbaddr'] == skbaddr: + skb['free_t'] = time + tx_free_list.append(skb) + del tx_xmit_list[i] + return + +def handle_skb_copy_datagram_iovec(event_info): + (name, context, cpu, time, pid, comm, skbaddr, skblen) = event_info + for i in range(len(rx_skb_list)): + rec_data = rx_skb_list[i] + if skbaddr == rec_data['skbaddr']: + rec_data.update({'handle':"skb_copy_datagram_iovec", + 'comm':comm, 'pid':pid, 'comm_t':time}) + del rx_skb_list[i] + return diff --git a/tools/perf/scripts/python/sched-migration.py b/tools/perf/scripts/python/sched-migration.py new file mode 100644 index 00000000..74d55ec0 --- /dev/null +++ b/tools/perf/scripts/python/sched-migration.py @@ -0,0 +1,461 @@ +#!/usr/bin/python +# +# Cpu task migration overview toy +# +# Copyright (C) 2010 Frederic Weisbecker <fweisbec@gmail.com> +# +# perf script event handlers have been generated by perf script -g python +# +# This software is distributed under the terms of the GNU General +# Public License ("GPL") version 2 as published by the Free Software +# Foundation. + + +import os +import sys + +from collections import defaultdict +from UserList import UserList + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') +sys.path.append('scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * +from SchedGui import * + + +threads = { 0 : "idle"} + +def thread_name(pid): + return "%s:%d" % (threads[pid], pid) + +class RunqueueEventUnknown: + @staticmethod + def color(): + return None + + def __repr__(self): + return "unknown" + +class RunqueueEventSleep: + @staticmethod + def color(): + return (0, 0, 0xff) + + def __init__(self, sleeper): + self.sleeper = sleeper + + def __repr__(self): + return "%s gone to sleep" % thread_name(self.sleeper) + +class RunqueueEventWakeup: + @staticmethod + def color(): + return (0xff, 0xff, 0) + + def __init__(self, wakee): + self.wakee = wakee + + def __repr__(self): + return "%s woke up" % thread_name(self.wakee) + +class RunqueueEventFork: + @staticmethod + def color(): + return (0, 0xff, 0) + + def __init__(self, child): + self.child = child + + def __repr__(self): + return "new forked task %s" % thread_name(self.child) + +class RunqueueMigrateIn: + @staticmethod + def color(): + return (0, 0xf0, 0xff) + + def __init__(self, new): + self.new = new + + def __repr__(self): + return "task migrated in %s" % thread_name(self.new) + +class RunqueueMigrateOut: + @staticmethod + def color(): + return (0xff, 0, 0xff) + + def __init__(self, old): + self.old = old + + def __repr__(self): + return "task migrated out %s" % thread_name(self.old) + +class RunqueueSnapshot: + def __init__(self, tasks = [0], event = RunqueueEventUnknown()): + self.tasks = tuple(tasks) + self.event = event + + def sched_switch(self, prev, prev_state, next): + event = RunqueueEventUnknown() + + if taskState(prev_state) == "R" and next in self.tasks \ + and prev in self.tasks: + return self + + if taskState(prev_state) != "R": + event = RunqueueEventSleep(prev) + + next_tasks = list(self.tasks[:]) + if prev in self.tasks: + if taskState(prev_state) != "R": + next_tasks.remove(prev) + elif taskState(prev_state) == "R": + next_tasks.append(prev) + + if next not in next_tasks: + next_tasks.append(next) + + return RunqueueSnapshot(next_tasks, event) + + def migrate_out(self, old): + if old not in self.tasks: + return self + next_tasks = [task for task in self.tasks if task != old] + + return RunqueueSnapshot(next_tasks, RunqueueMigrateOut(old)) + + def __migrate_in(self, new, event): + if new in self.tasks: + self.event = event + return self + next_tasks = self.tasks[:] + tuple([new]) + + return RunqueueSnapshot(next_tasks, event) + + def migrate_in(self, new): + return self.__migrate_in(new, RunqueueMigrateIn(new)) + + def wake_up(self, new): + return self.__migrate_in(new, RunqueueEventWakeup(new)) + + def wake_up_new(self, new): + return self.__migrate_in(new, RunqueueEventFork(new)) + + def load(self): + """ Provide the number of tasks on the runqueue. + Don't count idle""" + return len(self.tasks) - 1 + + def __repr__(self): + ret = self.tasks.__repr__() + ret += self.origin_tostring() + + return ret + +class TimeSlice: + def __init__(self, start, prev): + self.start = start + self.prev = prev + self.end = start + # cpus that triggered the event + self.event_cpus = [] + if prev is not None: + self.total_load = prev.total_load + self.rqs = prev.rqs.copy() + else: + self.rqs = defaultdict(RunqueueSnapshot) + self.total_load = 0 + + def __update_total_load(self, old_rq, new_rq): + diff = new_rq.load() - old_rq.load() + self.total_load += diff + + def sched_switch(self, ts_list, prev, prev_state, next, cpu): + old_rq = self.prev.rqs[cpu] + new_rq = old_rq.sched_switch(prev, prev_state, next) + + if old_rq is new_rq: + return + + self.rqs[cpu] = new_rq + self.__update_total_load(old_rq, new_rq) + ts_list.append(self) + self.event_cpus = [cpu] + + def migrate(self, ts_list, new, old_cpu, new_cpu): + if old_cpu == new_cpu: + return + old_rq = self.prev.rqs[old_cpu] + out_rq = old_rq.migrate_out(new) + self.rqs[old_cpu] = out_rq + self.__update_total_load(old_rq, out_rq) + + new_rq = self.prev.rqs[new_cpu] + in_rq = new_rq.migrate_in(new) + self.rqs[new_cpu] = in_rq + self.__update_total_load(new_rq, in_rq) + + ts_list.append(self) + + if old_rq is not out_rq: + self.event_cpus.append(old_cpu) + self.event_cpus.append(new_cpu) + + def wake_up(self, ts_list, pid, cpu, fork): + old_rq = self.prev.rqs[cpu] + if fork: + new_rq = old_rq.wake_up_new(pid) + else: + new_rq = old_rq.wake_up(pid) + + if new_rq is old_rq: + return + self.rqs[cpu] = new_rq + self.__update_total_load(old_rq, new_rq) + ts_list.append(self) + self.event_cpus = [cpu] + + def next(self, t): + self.end = t + return TimeSlice(t, self) + +class TimeSliceList(UserList): + def __init__(self, arg = []): + self.data = arg + + def get_time_slice(self, ts): + if len(self.data) == 0: + slice = TimeSlice(ts, TimeSlice(-1, None)) + else: + slice = self.data[-1].next(ts) + return slice + + def find_time_slice(self, ts): + start = 0 + end = len(self.data) + found = -1 + searching = True + while searching: + if start == end or start == end - 1: + searching = False + + i = (end + start) / 2 + if self.data[i].start <= ts and self.data[i].end >= ts: + found = i + end = i + continue + + if self.data[i].end < ts: + start = i + + elif self.data[i].start > ts: + end = i + + return found + + def set_root_win(self, win): + self.root_win = win + + def mouse_down(self, cpu, t): + idx = self.find_time_slice(t) + if idx == -1: + return + + ts = self[idx] + rq = ts.rqs[cpu] + raw = "CPU: %d\n" % cpu + raw += "Last event : %s\n" % rq.event.__repr__() + raw += "Timestamp : %d.%06d\n" % (ts.start / (10 ** 9), (ts.start % (10 ** 9)) / 1000) + raw += "Duration : %6d us\n" % ((ts.end - ts.start) / (10 ** 6)) + raw += "Load = %d\n" % rq.load() + for t in rq.tasks: + raw += "%s \n" % thread_name(t) + + self.root_win.update_summary(raw) + + def update_rectangle_cpu(self, slice, cpu): + rq = slice.rqs[cpu] + + if slice.total_load != 0: + load_rate = rq.load() / float(slice.total_load) + else: + load_rate = 0 + + red_power = int(0xff - (0xff * load_rate)) + color = (0xff, red_power, red_power) + + top_color = None + + if cpu in slice.event_cpus: + top_color = rq.event.color() + + self.root_win.paint_rectangle_zone(cpu, color, top_color, slice.start, slice.end) + + def fill_zone(self, start, end): + i = self.find_time_slice(start) + if i == -1: + return + + for i in xrange(i, len(self.data)): + timeslice = self.data[i] + if timeslice.start > end: + return + + for cpu in timeslice.rqs: + self.update_rectangle_cpu(timeslice, cpu) + + def interval(self): + if len(self.data) == 0: + return (0, 0) + + return (self.data[0].start, self.data[-1].end) + + def nr_rectangles(self): + last_ts = self.data[-1] + max_cpu = 0 + for cpu in last_ts.rqs: + if cpu > max_cpu: + max_cpu = cpu + return max_cpu + + +class SchedEventProxy: + def __init__(self): + self.current_tsk = defaultdict(lambda : -1) + self.timeslices = TimeSliceList() + + def sched_switch(self, headers, prev_comm, prev_pid, prev_prio, prev_state, + next_comm, next_pid, next_prio): + """ Ensure the task we sched out this cpu is really the one + we logged. Otherwise we may have missed traces """ + + on_cpu_task = self.current_tsk[headers.cpu] + + if on_cpu_task != -1 and on_cpu_task != prev_pid: + print "Sched switch event rejected ts: %s cpu: %d prev: %s(%d) next: %s(%d)" % \ + (headers.ts_format(), headers.cpu, prev_comm, prev_pid, next_comm, next_pid) + + threads[prev_pid] = prev_comm + threads[next_pid] = next_comm + self.current_tsk[headers.cpu] = next_pid + + ts = self.timeslices.get_time_slice(headers.ts()) + ts.sched_switch(self.timeslices, prev_pid, prev_state, next_pid, headers.cpu) + + def migrate(self, headers, pid, prio, orig_cpu, dest_cpu): + ts = self.timeslices.get_time_slice(headers.ts()) + ts.migrate(self.timeslices, pid, orig_cpu, dest_cpu) + + def wake_up(self, headers, comm, pid, success, target_cpu, fork): + if success == 0: + return + ts = self.timeslices.get_time_slice(headers.ts()) + ts.wake_up(self.timeslices, pid, target_cpu, fork) + + +def trace_begin(): + global parser + parser = SchedEventProxy() + +def trace_end(): + app = wx.App(False) + timeslices = parser.timeslices + frame = RootFrame(timeslices, "Migration") + app.MainLoop() + +def sched__sched_stat_runtime(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, runtime, vruntime): + pass + +def sched__sched_stat_iowait(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, delay): + pass + +def sched__sched_stat_sleep(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, delay): + pass + +def sched__sched_stat_wait(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, delay): + pass + +def sched__sched_process_fork(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + parent_comm, parent_pid, child_comm, child_pid): + pass + +def sched__sched_process_wait(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, prio): + pass + +def sched__sched_process_exit(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, prio): + pass + +def sched__sched_process_free(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, prio): + pass + +def sched__sched_migrate_task(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, prio, orig_cpu, + dest_cpu): + headers = EventHeaders(common_cpu, common_secs, common_nsecs, + common_pid, common_comm) + parser.migrate(headers, pid, prio, orig_cpu, dest_cpu) + +def sched__sched_switch(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + prev_comm, prev_pid, prev_prio, prev_state, + next_comm, next_pid, next_prio): + + headers = EventHeaders(common_cpu, common_secs, common_nsecs, + common_pid, common_comm) + parser.sched_switch(headers, prev_comm, prev_pid, prev_prio, prev_state, + next_comm, next_pid, next_prio) + +def sched__sched_wakeup_new(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, prio, success, + target_cpu): + headers = EventHeaders(common_cpu, common_secs, common_nsecs, + common_pid, common_comm) + parser.wake_up(headers, comm, pid, success, target_cpu, 1) + +def sched__sched_wakeup(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, prio, success, + target_cpu): + headers = EventHeaders(common_cpu, common_secs, common_nsecs, + common_pid, common_comm) + parser.wake_up(headers, comm, pid, success, target_cpu, 0) + +def sched__sched_wait_task(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid, prio): + pass + +def sched__sched_kthread_stop_ret(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + ret): + pass + +def sched__sched_kthread_stop(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + comm, pid): + pass + +def trace_unhandled(event_name, context, common_cpu, common_secs, common_nsecs, + common_pid, common_comm): + pass diff --git a/tools/perf/scripts/python/sctop.py b/tools/perf/scripts/python/sctop.py new file mode 100644 index 00000000..42c267e2 --- /dev/null +++ b/tools/perf/scripts/python/sctop.py @@ -0,0 +1,75 @@ +# system call top +# (c) 2010, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 +# +# Periodically displays system-wide system call totals, broken down by +# syscall. If a [comm] arg is specified, only syscalls called by +# [comm] are displayed. If an [interval] arg is specified, the display +# will be refreshed every [interval] seconds. The default interval is +# 3 seconds. + +import os, sys, thread, time + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * +from Util import * + +usage = "perf script -s sctop.py [comm] [interval]\n"; + +for_comm = None +default_interval = 3 +interval = default_interval + +if len(sys.argv) > 3: + sys.exit(usage) + +if len(sys.argv) > 2: + for_comm = sys.argv[1] + interval = int(sys.argv[2]) +elif len(sys.argv) > 1: + try: + interval = int(sys.argv[1]) + except ValueError: + for_comm = sys.argv[1] + interval = default_interval + +syscalls = autodict() + +def trace_begin(): + thread.start_new_thread(print_syscall_totals, (interval,)) + pass + +def raw_syscalls__sys_enter(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + id, args): + if for_comm is not None: + if common_comm != for_comm: + return + try: + syscalls[id] += 1 + except TypeError: + syscalls[id] = 1 + +def print_syscall_totals(interval): + while 1: + clear_term() + if for_comm is not None: + print "\nsyscall events for %s:\n\n" % (for_comm), + else: + print "\nsyscall events:\n\n", + + print "%-40s %10s\n" % ("event", "count"), + print "%-40s %10s\n" % ("----------------------------------------", \ + "----------"), + + for id, val in sorted(syscalls.iteritems(), key = lambda(k, v): (v, k), \ + reverse = True): + try: + print "%-40s %10d\n" % (syscall_name(id), val), + except TypeError: + pass + syscalls.clear() + time.sleep(interval) diff --git a/tools/perf/scripts/python/syscall-counts-by-pid.py b/tools/perf/scripts/python/syscall-counts-by-pid.py new file mode 100644 index 00000000..c64d1c55 --- /dev/null +++ b/tools/perf/scripts/python/syscall-counts-by-pid.py @@ -0,0 +1,69 @@ +# system call counts, by pid +# (c) 2010, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 +# +# Displays system-wide system call totals, broken down by syscall. +# If a [comm] arg is specified, only syscalls called by [comm] are displayed. + +import os, sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * +from Util import syscall_name + +usage = "perf script -s syscall-counts-by-pid.py [comm]\n"; + +for_comm = None +for_pid = None + +if len(sys.argv) > 2: + sys.exit(usage) + +if len(sys.argv) > 1: + try: + for_pid = int(sys.argv[1]) + except: + for_comm = sys.argv[1] + +syscalls = autodict() + +def trace_begin(): + print "Press control+C to stop and show the summary" + +def trace_end(): + print_syscall_totals() + +def raw_syscalls__sys_enter(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + id, args): + + if (for_comm and common_comm != for_comm) or \ + (for_pid and common_pid != for_pid ): + return + try: + syscalls[common_comm][common_pid][id] += 1 + except TypeError: + syscalls[common_comm][common_pid][id] = 1 + +def print_syscall_totals(): + if for_comm is not None: + print "\nsyscall events for %s:\n\n" % (for_comm), + else: + print "\nsyscall events by comm/pid:\n\n", + + print "%-40s %10s\n" % ("comm [pid]/syscalls", "count"), + print "%-40s %10s\n" % ("----------------------------------------", \ + "----------"), + + comm_keys = syscalls.keys() + for comm in comm_keys: + pid_keys = syscalls[comm].keys() + for pid in pid_keys: + print "\n%s [%d]\n" % (comm, pid), + id_keys = syscalls[comm][pid].keys() + for id, val in sorted(syscalls[comm][pid].iteritems(), \ + key = lambda(k, v): (v, k), reverse = True): + print " %-38s %10d\n" % (syscall_name(id), val), diff --git a/tools/perf/scripts/python/syscall-counts.py b/tools/perf/scripts/python/syscall-counts.py new file mode 100644 index 00000000..b435d3f1 --- /dev/null +++ b/tools/perf/scripts/python/syscall-counts.py @@ -0,0 +1,59 @@ +# system call counts +# (c) 2010, Tom Zanussi <tzanussi@gmail.com> +# Licensed under the terms of the GNU GPL License version 2 +# +# Displays system-wide system call totals, broken down by syscall. +# If a [comm] arg is specified, only syscalls called by [comm] are displayed. + +import os +import sys + +sys.path.append(os.environ['PERF_EXEC_PATH'] + \ + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + +from perf_trace_context import * +from Core import * +from Util import syscall_name + +usage = "perf script -s syscall-counts.py [comm]\n"; + +for_comm = None + +if len(sys.argv) > 2: + sys.exit(usage) + +if len(sys.argv) > 1: + for_comm = sys.argv[1] + +syscalls = autodict() + +def trace_begin(): + print "Press control+C to stop and show the summary" + +def trace_end(): + print_syscall_totals() + +def raw_syscalls__sys_enter(event_name, context, common_cpu, + common_secs, common_nsecs, common_pid, common_comm, + id, args): + if for_comm is not None: + if common_comm != for_comm: + return + try: + syscalls[id] += 1 + except TypeError: + syscalls[id] = 1 + +def print_syscall_totals(): + if for_comm is not None: + print "\nsyscall events for %s:\n\n" % (for_comm), + else: + print "\nsyscall events:\n\n", + + print "%-40s %10s\n" % ("event", "count"), + print "%-40s %10s\n" % ("----------------------------------------", \ + "-----------"), + + for id, val in sorted(syscalls.iteritems(), key = lambda(k, v): (v, k), \ + reverse = True): + print "%-40s %10d\n" % (syscall_name(id), val), diff --git a/tools/perf/tests/attr.c b/tools/perf/tests/attr.c new file mode 100644 index 00000000..bdcceb88 --- /dev/null +++ b/tools/perf/tests/attr.c @@ -0,0 +1,178 @@ + +/* + * The struct perf_event_attr test support. + * + * This test is embedded inside into perf directly and is governed + * by the PERF_TEST_ATTR environment variable and hook inside + * sys_perf_event_open function. + * + * The general idea is to store 'struct perf_event_attr' details for + * each event created within single perf command. Each event details + * are stored into separate text file. Once perf command is finished + * these files can be checked for values we expect for command. + * + * Besides 'struct perf_event_attr' values we also store 'fd' and + * 'group_fd' values to allow checking for groups created. + * + * This all is triggered by setting PERF_TEST_ATTR environment variable. + * It must contain name of existing directory with access and write + * permissions. All the event text files are stored there. + */ + +/* + * Powerpc needs __SANE_USERSPACE_TYPES__ before <linux/types.h> to select + * 'int-ll64.h' and avoid compile warnings when printing __u64 with %llu. + */ +#define __SANE_USERSPACE_TYPES__ +#include <stdlib.h> +#include <stdio.h> +#include <inttypes.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include "../perf.h" +#include "util.h" +#include "exec_cmd.h" +#include "tests.h" + +#define ENV "PERF_TEST_ATTR" + +extern int verbose; + +static char *dir; + +void test_attr__init(void) +{ + dir = getenv(ENV); + test_attr__enabled = (dir != NULL); +} + +#define BUFSIZE 1024 + +#define __WRITE_ASS(str, fmt, data) \ +do { \ + char buf[BUFSIZE]; \ + size_t size; \ + \ + size = snprintf(buf, BUFSIZE, #str "=%"fmt "\n", data); \ + if (1 != fwrite(buf, size, 1, file)) { \ + perror("test attr - failed to write event file"); \ + fclose(file); \ + return -1; \ + } \ + \ +} while (0) + +#define WRITE_ASS(field, fmt) __WRITE_ASS(field, fmt, attr->field) + +static int store_event(struct perf_event_attr *attr, pid_t pid, int cpu, + int fd, int group_fd, unsigned long flags) +{ + FILE *file; + char path[PATH_MAX]; + + snprintf(path, PATH_MAX, "%s/event-%d-%llu-%d", dir, + attr->type, attr->config, fd); + + file = fopen(path, "w+"); + if (!file) { + perror("test attr - failed to open event file"); + return -1; + } + + if (fprintf(file, "[event-%d-%llu-%d]\n", + attr->type, attr->config, fd) < 0) { + perror("test attr - failed to write event file"); + fclose(file); + return -1; + } + + /* syscall arguments */ + __WRITE_ASS(fd, "d", fd); + __WRITE_ASS(group_fd, "d", group_fd); + __WRITE_ASS(cpu, "d", cpu); + __WRITE_ASS(pid, "d", pid); + __WRITE_ASS(flags, "lu", flags); + + /* struct perf_event_attr */ + WRITE_ASS(type, PRIu32); + WRITE_ASS(size, PRIu32); + WRITE_ASS(config, "llu"); + WRITE_ASS(sample_period, "llu"); + WRITE_ASS(sample_type, "llu"); + WRITE_ASS(read_format, "llu"); + WRITE_ASS(disabled, "d"); + WRITE_ASS(inherit, "d"); + WRITE_ASS(pinned, "d"); + WRITE_ASS(exclusive, "d"); + WRITE_ASS(exclude_user, "d"); + WRITE_ASS(exclude_kernel, "d"); + WRITE_ASS(exclude_hv, "d"); + WRITE_ASS(exclude_idle, "d"); + WRITE_ASS(mmap, "d"); + WRITE_ASS(comm, "d"); + WRITE_ASS(freq, "d"); + WRITE_ASS(inherit_stat, "d"); + WRITE_ASS(enable_on_exec, "d"); + WRITE_ASS(task, "d"); + WRITE_ASS(watermark, "d"); + WRITE_ASS(precise_ip, "d"); + WRITE_ASS(mmap_data, "d"); + WRITE_ASS(sample_id_all, "d"); + WRITE_ASS(exclude_host, "d"); + WRITE_ASS(exclude_guest, "d"); + WRITE_ASS(exclude_callchain_kernel, "d"); + WRITE_ASS(exclude_callchain_user, "d"); + WRITE_ASS(wakeup_events, PRIu32); + WRITE_ASS(bp_type, PRIu32); + WRITE_ASS(config1, "llu"); + WRITE_ASS(config2, "llu"); + WRITE_ASS(branch_sample_type, "llu"); + WRITE_ASS(sample_regs_user, "llu"); + WRITE_ASS(sample_stack_user, PRIu32); + + fclose(file); + return 0; +} + +void test_attr__open(struct perf_event_attr *attr, pid_t pid, int cpu, + int fd, int group_fd, unsigned long flags) +{ + int errno_saved = errno; + + if (store_event(attr, pid, cpu, fd, group_fd, flags)) + die("test attr FAILED"); + + errno = errno_saved; +} + +static int run_dir(const char *d, const char *perf) +{ + char cmd[3*PATH_MAX]; + + snprintf(cmd, 3*PATH_MAX, PYTHON " %s/attr.py -d %s/attr/ -p %s %s", + d, d, perf, verbose ? "-v" : ""); + + return system(cmd); +} + +int test__attr(void) +{ + struct stat st; + char path_perf[PATH_MAX]; + char path_dir[PATH_MAX]; + + /* First try developement tree tests. */ + if (!lstat("./tests", &st)) + return run_dir("./tests", "./perf"); + + /* Then installed path. */ + snprintf(path_dir, PATH_MAX, "%s/tests", perf_exec_path()); + snprintf(path_perf, PATH_MAX, "%s/perf", BINDIR); + + if (!lstat(path_dir, &st) && + !lstat(path_perf, &st)) + return run_dir(path_dir, path_perf); + + fprintf(stderr, " (ommitted)"); + return 0; +} diff --git a/tools/perf/tests/attr.py b/tools/perf/tests/attr.py new file mode 100644 index 00000000..2f629ca4 --- /dev/null +++ b/tools/perf/tests/attr.py @@ -0,0 +1,331 @@ +#! /usr/bin/python + +import os +import sys +import glob +import optparse +import tempfile +import logging +import shutil +import ConfigParser + +class Fail(Exception): + def __init__(self, test, msg): + self.msg = msg + self.test = test + def getMsg(self): + return '\'%s\' - %s' % (self.test.path, self.msg) + +class Unsup(Exception): + def __init__(self, test): + self.test = test + def getMsg(self): + return '\'%s\'' % self.test.path + +class Event(dict): + terms = [ + 'flags', + 'type', + 'size', + 'config', + 'sample_period', + 'sample_type', + 'read_format', + 'disabled', + 'inherit', + 'pinned', + 'exclusive', + 'exclude_user', + 'exclude_kernel', + 'exclude_hv', + 'exclude_idle', + 'mmap', + 'comm', + 'freq', + 'inherit_stat', + 'enable_on_exec', + 'task', + 'watermark', + 'precise_ip', + 'mmap_data', + 'sample_id_all', + 'exclude_host', + 'exclude_guest', + 'exclude_callchain_kernel', + 'exclude_callchain_user', + 'wakeup_events', + 'bp_type', + 'config1', + 'config2', + 'branch_sample_type', + 'sample_regs_user', + 'sample_stack_user', + ] + + def add(self, data): + for key, val in data: + log.debug(" %s = %s" % (key, val)) + self[key] = val + + def __init__(self, name, data, base): + log.debug(" Event %s" % name); + self.name = name; + self.group = '' + self.add(base) + self.add(data) + + def compare_data(self, a, b): + # Allow multiple values in assignment separated by '|' + a_list = a.split('|') + b_list = b.split('|') + + for a_item in a_list: + for b_item in b_list: + if (a_item == b_item): + return True + elif (a_item == '*') or (b_item == '*'): + return True + + return False + + def equal(self, other): + for t in Event.terms: + log.debug(" [%s] %s %s" % (t, self[t], other[t])); + if not self.has_key(t) or not other.has_key(t): + return False + if not self.compare_data(self[t], other[t]): + return False + return True + + def diff(self, other): + for t in Event.terms: + if not self.has_key(t) or not other.has_key(t): + continue + if not self.compare_data(self[t], other[t]): + log.warning("expected %s=%s, got %s" % (t, self[t], other[t])) + + +# Test file description needs to have following sections: +# [config] +# - just single instance in file +# - needs to specify: +# 'command' - perf command name +# 'args' - special command arguments +# 'ret' - expected command return value (0 by default) +# +# [eventX:base] +# - one or multiple instances in file +# - expected values assignments +class Test(object): + def __init__(self, path, options): + parser = ConfigParser.SafeConfigParser() + parser.read(path) + + log.debug("running '%s'" % path) + + self.path = path + self.test_dir = options.test_dir + self.perf = options.perf + self.command = parser.get('config', 'command') + self.args = parser.get('config', 'args') + + try: + self.ret = parser.get('config', 'ret') + except: + self.ret = 0 + + self.expect = {} + self.result = {} + log.debug(" loading expected events"); + self.load_events(path, self.expect) + + def is_event(self, name): + if name.find("event") == -1: + return False + else: + return True + + def load_events(self, path, events): + parser_event = ConfigParser.SafeConfigParser() + parser_event.read(path) + + # The event record section header contains 'event' word, + # optionaly followed by ':' allowing to load 'parent + # event' first as a base + for section in filter(self.is_event, parser_event.sections()): + + parser_items = parser_event.items(section); + base_items = {} + + # Read parent event if there's any + if (':' in section): + base = section[section.index(':') + 1:] + parser_base = ConfigParser.SafeConfigParser() + parser_base.read(self.test_dir + '/' + base) + base_items = parser_base.items('event') + + e = Event(section, parser_items, base_items) + events[section] = e + + def run_cmd(self, tempdir): + cmd = "PERF_TEST_ATTR=%s %s %s -o %s/perf.data %s" % (tempdir, + self.perf, self.command, tempdir, self.args) + ret = os.WEXITSTATUS(os.system(cmd)) + + log.warning(" running '%s' ret %d " % (cmd, ret)) + + if ret != int(self.ret): + raise Unsup(self) + + def compare(self, expect, result): + match = {} + + log.debug(" compare"); + + # For each expected event find all matching + # events in result. Fail if there's not any. + for exp_name, exp_event in expect.items(): + exp_list = [] + log.debug(" matching [%s]" % exp_name) + for res_name, res_event in result.items(): + log.debug(" to [%s]" % res_name) + if (exp_event.equal(res_event)): + exp_list.append(res_name) + log.debug(" ->OK") + else: + log.debug(" ->FAIL"); + + log.debug(" match: [%s] matches %s" % (exp_name, str(exp_list))) + + # we did not any matching event - fail + if (not exp_list): + exp_event.diff(res_event) + raise Fail(self, 'match failure'); + + match[exp_name] = exp_list + + # For each defined group in the expected events + # check we match the same group in the result. + for exp_name, exp_event in expect.items(): + group = exp_event.group + + if (group == ''): + continue + + for res_name in match[exp_name]: + res_group = result[res_name].group + if res_group not in match[group]: + raise Fail(self, 'group failure') + + log.debug(" group: [%s] matches group leader %s" % + (exp_name, str(match[group]))) + + log.debug(" matched") + + def resolve_groups(self, events): + for name, event in events.items(): + group_fd = event['group_fd']; + if group_fd == '-1': + continue; + + for iname, ievent in events.items(): + if (ievent['fd'] == group_fd): + event.group = iname + log.debug('[%s] has group leader [%s]' % (name, iname)) + break; + + def run(self): + tempdir = tempfile.mkdtemp(); + + try: + # run the test script + self.run_cmd(tempdir); + + # load events expectation for the test + log.debug(" loading result events"); + for f in glob.glob(tempdir + '/event*'): + self.load_events(f, self.result); + + # resolve group_fd to event names + self.resolve_groups(self.expect); + self.resolve_groups(self.result); + + # do the expectation - results matching - both ways + self.compare(self.expect, self.result) + self.compare(self.result, self.expect) + + finally: + # cleanup + shutil.rmtree(tempdir) + + +def run_tests(options): + for f in glob.glob(options.test_dir + '/' + options.test): + try: + Test(f, options).run() + except Unsup, obj: + log.warning("unsupp %s" % obj.getMsg()) + +def setup_log(verbose): + global log + level = logging.CRITICAL + + if verbose == 1: + level = logging.WARNING + if verbose == 2: + level = logging.INFO + if verbose >= 3: + level = logging.DEBUG + + log = logging.getLogger('test') + log.setLevel(level) + ch = logging.StreamHandler() + ch.setLevel(level) + formatter = logging.Formatter('%(message)s') + ch.setFormatter(formatter) + log.addHandler(ch) + +USAGE = '''%s [OPTIONS] + -d dir # tests dir + -p path # perf binary + -t test # single test + -v # verbose level +''' % sys.argv[0] + +def main(): + parser = optparse.OptionParser(usage=USAGE) + + parser.add_option("-t", "--test", + action="store", type="string", dest="test") + parser.add_option("-d", "--test-dir", + action="store", type="string", dest="test_dir") + parser.add_option("-p", "--perf", + action="store", type="string", dest="perf") + parser.add_option("-v", "--verbose", + action="count", dest="verbose") + + options, args = parser.parse_args() + if args: + parser.error('FAILED wrong arguments %s' % ' '.join(args)) + return -1 + + setup_log(options.verbose) + + if not options.test_dir: + print 'FAILED no -d option specified' + sys.exit(-1) + + if not options.test: + options.test = 'test*' + + try: + run_tests(options) + + except Fail, obj: + print "FAILED %s" % obj.getMsg(); + sys.exit(-1) + + sys.exit(0) + +if __name__ == '__main__': + main() diff --git a/tools/perf/tests/attr/README b/tools/perf/tests/attr/README new file mode 100644 index 00000000..d102957c --- /dev/null +++ b/tools/perf/tests/attr/README @@ -0,0 +1,64 @@ +The struct perf_event_attr test (attr tests) support +==================================================== +This testing support is embedded into perf directly and is governed +by the PERF_TEST_ATTR environment variable and hook inside the +sys_perf_event_open function. + +The general idea is to store 'struct perf_event_attr' details for +each event created within single perf command. Each event details +are stored into separate text file. Once perf command is finished +these files are checked for values we expect for command. + +The attr tests consist of following parts: + +tests/attr.c +------------ +This is the sys_perf_event_open hook implementation. The hook +is triggered when the PERF_TEST_ATTR environment variable is +defined. It must contain name of existing directory with access +and write permissions. + +For each sys_perf_event_open call event details are stored in +separate file. Besides 'struct perf_event_attr' values we also +store 'fd' and 'group_fd' values to allow checking for groups. + +tests/attr.py +------------- +This is the python script that does all the hard work. It reads +the test definition, executes it and checks results. + +tests/attr/ +----------- +Directory containing all attr test definitions. +Following tests are defined (with perf commands): + + perf record kill (test-record-basic) + perf record -b kill (test-record-branch-any) + perf record -j any kill (test-record-branch-filter-any) + perf record -j any_call kill (test-record-branch-filter-any_call) + perf record -j any_ret kill (test-record-branch-filter-any_ret) + perf record -j hv kill (test-record-branch-filter-hv) + perf record -j ind_call kill (test-record-branch-filter-ind_call) + perf record -j k kill (test-record-branch-filter-k) + perf record -j u kill (test-record-branch-filter-u) + perf record -c 123 kill (test-record-count) + perf record -d kill (test-record-data) + perf record -F 100 kill (test-record-freq) + perf record -g -- kill (test-record-graph-default) + perf record -g dwarf -- kill (test-record-graph-dwarf) + perf record -g fp kill (test-record-graph-fp) + perf record --group -e cycles,instructions kill (test-record-group) + perf record -e '{cycles,instructions}' kill (test-record-group1) + perf record -D kill (test-record-no-delay) + perf record -i kill (test-record-no-inherit) + perf record -n kill (test-record-no-samples) + perf record -c 100 -P kill (test-record-period) + perf record -R kill (test-record-raw) + perf stat -e cycles kill (test-stat-basic) + perf stat kill (test-stat-default) + perf stat -d kill (test-stat-detailed-1) + perf stat -dd kill (test-stat-detailed-2) + perf stat -ddd kill (test-stat-detailed-3) + perf stat --group -e cycles,instructions kill (test-stat-group) + perf stat -e '{cycles,instructions}' kill (test-stat-group1) + perf stat -i -e cycles kill (test-stat-no-inherit) diff --git a/tools/perf/tests/attr/base-record b/tools/perf/tests/attr/base-record new file mode 100644 index 00000000..5bc3880f --- /dev/null +++ b/tools/perf/tests/attr/base-record @@ -0,0 +1,39 @@ +[event] +fd=1 +group_fd=-1 +flags=0 +type=0|1 +size=96 +config=0 +sample_period=4000 +sample_type=263 +read_format=0 +disabled=1 +inherit=1 +pinned=0 +exclusive=0 +exclude_user=0 +exclude_kernel=0 +exclude_hv=0 +exclude_idle=0 +mmap=1 +comm=1 +freq=1 +inherit_stat=0 +enable_on_exec=1 +task=0 +watermark=0 +precise_ip=0 +mmap_data=0 +sample_id_all=1 +exclude_host=0 +exclude_guest=1 +exclude_callchain_kernel=0 +exclude_callchain_user=0 +wakeup_events=0 +bp_type=0 +config1=0 +config2=0 +branch_sample_type=0 +sample_regs_user=0 +sample_stack_user=0 diff --git a/tools/perf/tests/attr/base-stat b/tools/perf/tests/attr/base-stat new file mode 100644 index 00000000..4bd79a82 --- /dev/null +++ b/tools/perf/tests/attr/base-stat @@ -0,0 +1,39 @@ +[event] +fd=1 +group_fd=-1 +flags=0 +type=0 +size=96 +config=0 +sample_period=0 +sample_type=0 +read_format=3 +disabled=1 +inherit=1 +pinned=0 +exclusive=0 +exclude_user=0 +exclude_kernel=0 +exclude_hv=0 +exclude_idle=0 +mmap=0 +comm=0 +freq=0 +inherit_stat=0 +enable_on_exec=1 +task=0 +watermark=0 +precise_ip=0 +mmap_data=0 +sample_id_all=0 +exclude_host=0 +exclude_guest=1 +exclude_callchain_kernel=0 +exclude_callchain_user=0 +wakeup_events=0 +bp_type=0 +config1=0 +config2=0 +branch_sample_type=0 +sample_regs_user=0 +sample_stack_user=0 diff --git a/tools/perf/tests/attr/test-record-basic b/tools/perf/tests/attr/test-record-basic new file mode 100644 index 00000000..55c04283 --- /dev/null +++ b/tools/perf/tests/attr/test-record-basic @@ -0,0 +1,5 @@ +[config] +command = record +args = kill >/dev/null 2>&1 + +[event:base-record] diff --git a/tools/perf/tests/attr/test-record-branch-any b/tools/perf/tests/attr/test-record-branch-any new file mode 100644 index 00000000..1421960e --- /dev/null +++ b/tools/perf/tests/attr/test-record-branch-any @@ -0,0 +1,8 @@ +[config] +command = record +args = -b kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=2311 +branch_sample_type=8 diff --git a/tools/perf/tests/attr/test-record-branch-filter-any b/tools/perf/tests/attr/test-record-branch-filter-any new file mode 100644 index 00000000..915c4df0 --- /dev/null +++ b/tools/perf/tests/attr/test-record-branch-filter-any @@ -0,0 +1,8 @@ +[config] +command = record +args = -j any kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=2311 +branch_sample_type=8 diff --git a/tools/perf/tests/attr/test-record-branch-filter-any_call b/tools/perf/tests/attr/test-record-branch-filter-any_call new file mode 100644 index 00000000..8708dbd4 --- /dev/null +++ b/tools/perf/tests/attr/test-record-branch-filter-any_call @@ -0,0 +1,8 @@ +[config] +command = record +args = -j any_call kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=2311 +branch_sample_type=16 diff --git a/tools/perf/tests/attr/test-record-branch-filter-any_ret b/tools/perf/tests/attr/test-record-branch-filter-any_ret new file mode 100644 index 00000000..0d3607a6 --- /dev/null +++ b/tools/perf/tests/attr/test-record-branch-filter-any_ret @@ -0,0 +1,8 @@ +[config] +command = record +args = -j any_ret kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=2311 +branch_sample_type=32 diff --git a/tools/perf/tests/attr/test-record-branch-filter-hv b/tools/perf/tests/attr/test-record-branch-filter-hv new file mode 100644 index 00000000..f2552674 --- /dev/null +++ b/tools/perf/tests/attr/test-record-branch-filter-hv @@ -0,0 +1,8 @@ +[config] +command = record +args = -j hv kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=2311 +branch_sample_type=8 diff --git a/tools/perf/tests/attr/test-record-branch-filter-ind_call b/tools/perf/tests/attr/test-record-branch-filter-ind_call new file mode 100644 index 00000000..e862dd17 --- /dev/null +++ b/tools/perf/tests/attr/test-record-branch-filter-ind_call @@ -0,0 +1,8 @@ +[config] +command = record +args = -j ind_call kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=2311 +branch_sample_type=64 diff --git a/tools/perf/tests/attr/test-record-branch-filter-k b/tools/perf/tests/attr/test-record-branch-filter-k new file mode 100644 index 00000000..182971e8 --- /dev/null +++ b/tools/perf/tests/attr/test-record-branch-filter-k @@ -0,0 +1,8 @@ +[config] +command = record +args = -j k kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=2311 +branch_sample_type=8 diff --git a/tools/perf/tests/attr/test-record-branch-filter-u b/tools/perf/tests/attr/test-record-branch-filter-u new file mode 100644 index 00000000..83449ef9 --- /dev/null +++ b/tools/perf/tests/attr/test-record-branch-filter-u @@ -0,0 +1,8 @@ +[config] +command = record +args = -j u kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=2311 +branch_sample_type=8 diff --git a/tools/perf/tests/attr/test-record-count b/tools/perf/tests/attr/test-record-count new file mode 100644 index 00000000..2f841de5 --- /dev/null +++ b/tools/perf/tests/attr/test-record-count @@ -0,0 +1,8 @@ +[config] +command = record +args = -c 123 kill >/dev/null 2>&1 + +[event:base-record] +sample_period=123 +sample_type=7 +freq=0 diff --git a/tools/perf/tests/attr/test-record-data b/tools/perf/tests/attr/test-record-data new file mode 100644 index 00000000..6627c3e7 --- /dev/null +++ b/tools/perf/tests/attr/test-record-data @@ -0,0 +1,8 @@ +[config] +command = record +args = -d kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=271 +mmap_data=1 diff --git a/tools/perf/tests/attr/test-record-freq b/tools/perf/tests/attr/test-record-freq new file mode 100644 index 00000000..600d0f8f --- /dev/null +++ b/tools/perf/tests/attr/test-record-freq @@ -0,0 +1,6 @@ +[config] +command = record +args = -F 100 kill >/dev/null 2>&1 + +[event:base-record] +sample_period=100 diff --git a/tools/perf/tests/attr/test-record-graph-default b/tools/perf/tests/attr/test-record-graph-default new file mode 100644 index 00000000..833d1849 --- /dev/null +++ b/tools/perf/tests/attr/test-record-graph-default @@ -0,0 +1,6 @@ +[config] +command = record +args = -g -- kill >/dev/null 2>&1 + +[event:base-record] +sample_type=295 diff --git a/tools/perf/tests/attr/test-record-graph-dwarf b/tools/perf/tests/attr/test-record-graph-dwarf new file mode 100644 index 00000000..e93e082f --- /dev/null +++ b/tools/perf/tests/attr/test-record-graph-dwarf @@ -0,0 +1,10 @@ +[config] +command = record +args = -g dwarf -- kill >/dev/null 2>&1 + +[event:base-record] +sample_type=12583 +exclude_callchain_user=1 +sample_stack_user=8192 +# TODO different for each arch, no support for that now +sample_regs_user=* diff --git a/tools/perf/tests/attr/test-record-graph-fp b/tools/perf/tests/attr/test-record-graph-fp new file mode 100644 index 00000000..7cef3743 --- /dev/null +++ b/tools/perf/tests/attr/test-record-graph-fp @@ -0,0 +1,6 @@ +[config] +command = record +args = -g fp kill >/dev/null 2>&1 + +[event:base-record] +sample_type=295 diff --git a/tools/perf/tests/attr/test-record-group b/tools/perf/tests/attr/test-record-group new file mode 100644 index 00000000..57739cac --- /dev/null +++ b/tools/perf/tests/attr/test-record-group @@ -0,0 +1,20 @@ +[config] +command = record +args = --group -e cycles,instructions kill >/dev/null 2>&1 + +[event-1:base-record] +fd=1 +group_fd=-1 +sample_type=327 +read_format=4 + +[event-2:base-record] +fd=2 +group_fd=1 +config=1 +sample_type=327 +read_format=4 +mmap=0 +comm=0 +enable_on_exec=0 +disabled=0 diff --git a/tools/perf/tests/attr/test-record-group1 b/tools/perf/tests/attr/test-record-group1 new file mode 100644 index 00000000..c5548d05 --- /dev/null +++ b/tools/perf/tests/attr/test-record-group1 @@ -0,0 +1,21 @@ +[config] +command = record +args = -e '{cycles,instructions}' kill >/dev/null 2>&1 + +[event-1:base-record] +fd=1 +group_fd=-1 +sample_type=327 +read_format=4 + +[event-2:base-record] +fd=2 +group_fd=1 +type=0 +config=1 +sample_type=327 +read_format=4 +mmap=0 +comm=0 +enable_on_exec=0 +disabled=0 diff --git a/tools/perf/tests/attr/test-record-no-delay b/tools/perf/tests/attr/test-record-no-delay new file mode 100644 index 00000000..f253b78c --- /dev/null +++ b/tools/perf/tests/attr/test-record-no-delay @@ -0,0 +1,9 @@ +[config] +command = record +args = -D kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=263 +watermark=0 +wakeup_events=1 diff --git a/tools/perf/tests/attr/test-record-no-inherit b/tools/perf/tests/attr/test-record-no-inherit new file mode 100644 index 00000000..9079a25c --- /dev/null +++ b/tools/perf/tests/attr/test-record-no-inherit @@ -0,0 +1,7 @@ +[config] +command = record +args = -i kill >/dev/null 2>&1 + +[event:base-record] +sample_type=259 +inherit=0 diff --git a/tools/perf/tests/attr/test-record-no-samples b/tools/perf/tests/attr/test-record-no-samples new file mode 100644 index 00000000..d0141b24 --- /dev/null +++ b/tools/perf/tests/attr/test-record-no-samples @@ -0,0 +1,6 @@ +[config] +command = record +args = -n kill >/dev/null 2>&1 + +[event:base-record] +sample_period=0 diff --git a/tools/perf/tests/attr/test-record-period b/tools/perf/tests/attr/test-record-period new file mode 100644 index 00000000..8abc5314 --- /dev/null +++ b/tools/perf/tests/attr/test-record-period @@ -0,0 +1,7 @@ +[config] +command = record +args = -c 100 -P kill >/dev/null 2>&1 + +[event:base-record] +sample_period=100 +freq=0 diff --git a/tools/perf/tests/attr/test-record-raw b/tools/perf/tests/attr/test-record-raw new file mode 100644 index 00000000..4a8ef25b --- /dev/null +++ b/tools/perf/tests/attr/test-record-raw @@ -0,0 +1,7 @@ +[config] +command = record +args = -R kill >/dev/null 2>&1 + +[event:base-record] +sample_period=4000 +sample_type=1415 diff --git a/tools/perf/tests/attr/test-stat-basic b/tools/perf/tests/attr/test-stat-basic new file mode 100644 index 00000000..74e17881 --- /dev/null +++ b/tools/perf/tests/attr/test-stat-basic @@ -0,0 +1,6 @@ +[config] +command = stat +args = -e cycles kill >/dev/null 2>&1 +ret = 1 + +[event:base-stat] diff --git a/tools/perf/tests/attr/test-stat-default b/tools/perf/tests/attr/test-stat-default new file mode 100644 index 00000000..19270f54 --- /dev/null +++ b/tools/perf/tests/attr/test-stat-default @@ -0,0 +1,64 @@ +[config] +command = stat +args = kill >/dev/null 2>&1 +ret = 1 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_TASK_CLOCK +[event1:base-stat] +fd=1 +type=1 +config=1 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CONTEXT_SWITCHES +[event2:base-stat] +fd=2 +type=1 +config=3 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CPU_MIGRATIONS +[event3:base-stat] +fd=3 +type=1 +config=4 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_PAGE_FAULTS +[event4:base-stat] +fd=4 +type=1 +config=2 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_CPU_CYCLES +[event5:base-stat] +fd=5 +type=0 +config=0 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_FRONTEND +[event6:base-stat] +fd=6 +type=0 +config=7 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_BACKEND +[event7:base-stat] +fd=7 +type=0 +config=8 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_INSTRUCTIONS +[event8:base-stat] +fd=8 +type=0 +config=1 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS +[event9:base-stat] +fd=9 +type=0 +config=4 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES +[event10:base-stat] +fd=10 +type=0 +config=5 diff --git a/tools/perf/tests/attr/test-stat-detailed-1 b/tools/perf/tests/attr/test-stat-detailed-1 new file mode 100644 index 00000000..51426b87 --- /dev/null +++ b/tools/perf/tests/attr/test-stat-detailed-1 @@ -0,0 +1,101 @@ +[config] +command = stat +args = -d kill >/dev/null 2>&1 +ret = 1 + + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_TASK_CLOCK +[event1:base-stat] +fd=1 +type=1 +config=1 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CONTEXT_SWITCHES +[event2:base-stat] +fd=2 +type=1 +config=3 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CPU_MIGRATIONS +[event3:base-stat] +fd=3 +type=1 +config=4 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_PAGE_FAULTS +[event4:base-stat] +fd=4 +type=1 +config=2 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_CPU_CYCLES +[event5:base-stat] +fd=5 +type=0 +config=0 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_FRONTEND +[event6:base-stat] +fd=6 +type=0 +config=7 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_BACKEND +[event7:base-stat] +fd=7 +type=0 +config=8 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_INSTRUCTIONS +[event8:base-stat] +fd=8 +type=0 +config=1 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS +[event9:base-stat] +fd=9 +type=0 +config=4 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES +[event10:base-stat] +fd=10 +type=0 +config=5 + +# PERF_TYPE_HW_CACHE / +# PERF_COUNT_HW_CACHE_L1D << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event11:base-stat] +fd=11 +type=3 +config=0 + +# PERF_TYPE_HW_CACHE / +# PERF_COUNT_HW_CACHE_L1D << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event12:base-stat] +fd=12 +type=3 +config=65536 + +# PERF_TYPE_HW_CACHE / +# PERF_COUNT_HW_CACHE_LL << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event13:base-stat] +fd=13 +type=3 +config=2 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_LL << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event14:base-stat] +fd=14 +type=3 +config=65538 diff --git a/tools/perf/tests/attr/test-stat-detailed-2 b/tools/perf/tests/attr/test-stat-detailed-2 new file mode 100644 index 00000000..8de5acc3 --- /dev/null +++ b/tools/perf/tests/attr/test-stat-detailed-2 @@ -0,0 +1,155 @@ +[config] +command = stat +args = -dd kill >/dev/null 2>&1 +ret = 1 + + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_TASK_CLOCK +[event1:base-stat] +fd=1 +type=1 +config=1 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CONTEXT_SWITCHES +[event2:base-stat] +fd=2 +type=1 +config=3 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CPU_MIGRATIONS +[event3:base-stat] +fd=3 +type=1 +config=4 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_PAGE_FAULTS +[event4:base-stat] +fd=4 +type=1 +config=2 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_CPU_CYCLES +[event5:base-stat] +fd=5 +type=0 +config=0 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_FRONTEND +[event6:base-stat] +fd=6 +type=0 +config=7 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_BACKEND +[event7:base-stat] +fd=7 +type=0 +config=8 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_INSTRUCTIONS +[event8:base-stat] +fd=8 +type=0 +config=1 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS +[event9:base-stat] +fd=9 +type=0 +config=4 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES +[event10:base-stat] +fd=10 +type=0 +config=5 + +# PERF_TYPE_HW_CACHE / +# PERF_COUNT_HW_CACHE_L1D << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event11:base-stat] +fd=11 +type=3 +config=0 + +# PERF_TYPE_HW_CACHE / +# PERF_COUNT_HW_CACHE_L1D << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event12:base-stat] +fd=12 +type=3 +config=65536 + +# PERF_TYPE_HW_CACHE / +# PERF_COUNT_HW_CACHE_LL << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event13:base-stat] +fd=13 +type=3 +config=2 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_LL << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event14:base-stat] +fd=14 +type=3 +config=65538 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_L1I << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event15:base-stat] +fd=15 +type=3 +config=1 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_L1I << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event16:base-stat] +fd=16 +type=3 +config=65537 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_DTLB << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event17:base-stat] +fd=17 +type=3 +config=3 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_DTLB << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event18:base-stat] +fd=18 +type=3 +config=65539 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_ITLB << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event19:base-stat] +fd=19 +type=3 +config=4 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_ITLB << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event20:base-stat] +fd=20 +type=3 +config=65540 diff --git a/tools/perf/tests/attr/test-stat-detailed-3 b/tools/perf/tests/attr/test-stat-detailed-3 new file mode 100644 index 00000000..0a1f45bf --- /dev/null +++ b/tools/perf/tests/attr/test-stat-detailed-3 @@ -0,0 +1,173 @@ +[config] +command = stat +args = -ddd kill >/dev/null 2>&1 +ret = 1 + + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_TASK_CLOCK +[event1:base-stat] +fd=1 +type=1 +config=1 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CONTEXT_SWITCHES +[event2:base-stat] +fd=2 +type=1 +config=3 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CPU_MIGRATIONS +[event3:base-stat] +fd=3 +type=1 +config=4 + +# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_PAGE_FAULTS +[event4:base-stat] +fd=4 +type=1 +config=2 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_CPU_CYCLES +[event5:base-stat] +fd=5 +type=0 +config=0 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_FRONTEND +[event6:base-stat] +fd=6 +type=0 +config=7 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_BACKEND +[event7:base-stat] +fd=7 +type=0 +config=8 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_INSTRUCTIONS +[event8:base-stat] +fd=8 +type=0 +config=1 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS +[event9:base-stat] +fd=9 +type=0 +config=4 + +# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES +[event10:base-stat] +fd=10 +type=0 +config=5 + +# PERF_TYPE_HW_CACHE / +# PERF_COUNT_HW_CACHE_L1D << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event11:base-stat] +fd=11 +type=3 +config=0 + +# PERF_TYPE_HW_CACHE / +# PERF_COUNT_HW_CACHE_L1D << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event12:base-stat] +fd=12 +type=3 +config=65536 + +# PERF_TYPE_HW_CACHE / +# PERF_COUNT_HW_CACHE_LL << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event13:base-stat] +fd=13 +type=3 +config=2 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_LL << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event14:base-stat] +fd=14 +type=3 +config=65538 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_L1I << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event15:base-stat] +fd=15 +type=3 +config=1 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_L1I << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event16:base-stat] +fd=16 +type=3 +config=65537 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_DTLB << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event17:base-stat] +fd=17 +type=3 +config=3 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_DTLB << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event18:base-stat] +fd=18 +type=3 +config=65539 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_ITLB << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event19:base-stat] +fd=19 +type=3 +config=4 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_ITLB << 0 | +# (PERF_COUNT_HW_CACHE_OP_READ << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event20:base-stat] +fd=20 +type=3 +config=65540 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_L1D << 0 | +# (PERF_COUNT_HW_CACHE_OP_PREFETCH << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) +[event21:base-stat] +fd=21 +type=3 +config=512 + +# PERF_TYPE_HW_CACHE, +# PERF_COUNT_HW_CACHE_L1D << 0 | +# (PERF_COUNT_HW_CACHE_OP_PREFETCH << 8) | +# (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) +[event22:base-stat] +fd=22 +type=3 +config=66048 diff --git a/tools/perf/tests/attr/test-stat-group b/tools/perf/tests/attr/test-stat-group new file mode 100644 index 00000000..fdc1596a --- /dev/null +++ b/tools/perf/tests/attr/test-stat-group @@ -0,0 +1,15 @@ +[config] +command = stat +args = --group -e cycles,instructions kill >/dev/null 2>&1 +ret = 1 + +[event-1:base-stat] +fd=1 +group_fd=-1 + +[event-2:base-stat] +fd=2 +group_fd=1 +config=1 +disabled=0 +enable_on_exec=0 diff --git a/tools/perf/tests/attr/test-stat-group1 b/tools/perf/tests/attr/test-stat-group1 new file mode 100644 index 00000000..2a1f86e4 --- /dev/null +++ b/tools/perf/tests/attr/test-stat-group1 @@ -0,0 +1,15 @@ +[config] +command = stat +args = -e '{cycles,instructions}' kill >/dev/null 2>&1 +ret = 1 + +[event-1:base-stat] +fd=1 +group_fd=-1 + +[event-2:base-stat] +fd=2 +group_fd=1 +config=1 +disabled=0 +enable_on_exec=0 diff --git a/tools/perf/tests/attr/test-stat-no-inherit b/tools/perf/tests/attr/test-stat-no-inherit new file mode 100644 index 00000000..d54b2a1e --- /dev/null +++ b/tools/perf/tests/attr/test-stat-no-inherit @@ -0,0 +1,7 @@ +[config] +command = stat +args = -i -e cycles kill >/dev/null 2>&1 +ret = 1 + +[event:base-stat] +inherit=0 diff --git a/tools/perf/tests/builtin-test.c b/tools/perf/tests/builtin-test.c new file mode 100644 index 00000000..acb98e0e --- /dev/null +++ b/tools/perf/tests/builtin-test.c @@ -0,0 +1,203 @@ +/* + * builtin-test.c + * + * Builtin regression testing command: ever growing number of sanity tests + */ +#include "builtin.h" +#include "intlist.h" +#include "tests.h" +#include "debug.h" +#include "color.h" +#include "parse-options.h" +#include "symbol.h" + +static struct test { + const char *desc; + int (*func)(void); +} tests[] = { + { + .desc = "vmlinux symtab matches kallsyms", + .func = test__vmlinux_matches_kallsyms, + }, + { + .desc = "detect open syscall event", + .func = test__open_syscall_event, + }, + { + .desc = "detect open syscall event on all cpus", + .func = test__open_syscall_event_on_all_cpus, + }, + { + .desc = "read samples using the mmap interface", + .func = test__basic_mmap, + }, + { + .desc = "parse events tests", + .func = test__parse_events, + }, +#if defined(__x86_64__) || defined(__i386__) + { + .desc = "x86 rdpmc test", + .func = test__rdpmc, + }, +#endif + { + .desc = "Validate PERF_RECORD_* events & perf_sample fields", + .func = test__PERF_RECORD, + }, + { + .desc = "Test perf pmu format parsing", + .func = test__pmu, + }, + { + .desc = "Test dso data interface", + .func = test__dso_data, + }, + { + .desc = "roundtrip evsel->name check", + .func = test__perf_evsel__roundtrip_name_test, + }, + { + .desc = "Check parsing of sched tracepoints fields", + .func = test__perf_evsel__tp_sched_test, + }, + { + .desc = "Generate and check syscalls:sys_enter_open event fields", + .func = test__syscall_open_tp_fields, + }, + { + .desc = "struct perf_event_attr setup", + .func = test__attr, + }, + { + .desc = "Test matching and linking mutliple hists", + .func = test__hists_link, + }, + { + .desc = "Try 'use perf' in python, checking link problems", + .func = test__python_use, + }, + { + .func = NULL, + }, +}; + +static bool perf_test__matches(int curr, int argc, const char *argv[]) +{ + int i; + + if (argc == 0) + return true; + + for (i = 0; i < argc; ++i) { + char *end; + long nr = strtoul(argv[i], &end, 10); + + if (*end == '\0') { + if (nr == curr + 1) + return true; + continue; + } + + if (strstr(tests[curr].desc, argv[i])) + return true; + } + + return false; +} + +static int __cmd_test(int argc, const char *argv[], struct intlist *skiplist) +{ + int i = 0; + int width = 0; + + while (tests[i].func) { + int len = strlen(tests[i].desc); + + if (width < len) + width = len; + ++i; + } + + i = 0; + while (tests[i].func) { + int curr = i++, err; + + if (!perf_test__matches(curr, argc, argv)) + continue; + + pr_info("%2d: %-*s:", i, width, tests[curr].desc); + + if (intlist__find(skiplist, i)) { + color_fprintf(stderr, PERF_COLOR_YELLOW, " Skip (user override)\n"); + continue; + } + + pr_debug("\n--- start ---\n"); + err = tests[curr].func(); + pr_debug("---- end ----\n%s:", tests[curr].desc); + + switch (err) { + case TEST_OK: + pr_info(" Ok\n"); + break; + case TEST_SKIP: + color_fprintf(stderr, PERF_COLOR_YELLOW, " Skip\n"); + break; + case TEST_FAIL: + default: + color_fprintf(stderr, PERF_COLOR_RED, " FAILED!\n"); + break; + } + } + + return 0; +} + +static int perf_test__list(int argc, const char **argv) +{ + int i = 0; + + while (tests[i].func) { + int curr = i++; + + if (argc > 1 && !strstr(tests[curr].desc, argv[1])) + continue; + + pr_info("%2d: %s\n", i, tests[curr].desc); + } + + return 0; +} + +int cmd_test(int argc, const char **argv, const char *prefix __maybe_unused) +{ + const char * const test_usage[] = { + "perf test [<options>] [{list <test-name-fragment>|[<test-name-fragments>|<test-numbers>]}]", + NULL, + }; + const char *skip = NULL; + const struct option test_options[] = { + OPT_STRING('s', "skip", &skip, "tests", "tests to skip"), + OPT_INCR('v', "verbose", &verbose, + "be more verbose (show symbol address, etc)"), + OPT_END() + }; + struct intlist *skiplist = NULL; + + argc = parse_options(argc, argv, test_options, test_usage, 0); + if (argc >= 1 && !strcmp(argv[0], "list")) + return perf_test__list(argc, argv); + + symbol_conf.priv_size = sizeof(int); + symbol_conf.sort_by_name = true; + symbol_conf.try_vmlinux_path = true; + + if (symbol__init() < 0) + return -1; + + if (skip != NULL) + skiplist = intlist__new(skip); + + return __cmd_test(argc, argv, skiplist); +} diff --git a/tools/perf/tests/dso-data.c b/tools/perf/tests/dso-data.c new file mode 100644 index 00000000..5eaffa2d --- /dev/null +++ b/tools/perf/tests/dso-data.c @@ -0,0 +1,159 @@ +#include "util.h" + +#include <stdlib.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <string.h> + +#include "machine.h" +#include "symbol.h" +#include "tests.h" + +#define TEST_ASSERT_VAL(text, cond) \ +do { \ + if (!(cond)) { \ + pr_debug("FAILED %s:%d %s\n", __FILE__, __LINE__, text); \ + return -1; \ + } \ +} while (0) + +static char *test_file(int size) +{ + static char buf_templ[] = "/tmp/test-XXXXXX"; + char *templ = buf_templ; + int fd, i; + unsigned char *buf; + + fd = mkstemp(templ); + if (fd < 0) { + perror("mkstemp failed"); + return NULL; + } + + buf = malloc(size); + if (!buf) { + close(fd); + return NULL; + } + + for (i = 0; i < size; i++) + buf[i] = (unsigned char) ((int) i % 10); + + if (size != write(fd, buf, size)) + templ = NULL; + + close(fd); + return templ; +} + +#define TEST_FILE_SIZE (DSO__DATA_CACHE_SIZE * 20) + +struct test_data_offset { + off_t offset; + u8 data[10]; + int size; +}; + +struct test_data_offset offsets[] = { + /* Fill first cache page. */ + { + .offset = 10, + .data = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 }, + .size = 10, + }, + /* Read first cache page. */ + { + .offset = 10, + .data = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 }, + .size = 10, + }, + /* Fill cache boundary pages. */ + { + .offset = DSO__DATA_CACHE_SIZE - DSO__DATA_CACHE_SIZE % 10, + .data = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 }, + .size = 10, + }, + /* Read cache boundary pages. */ + { + .offset = DSO__DATA_CACHE_SIZE - DSO__DATA_CACHE_SIZE % 10, + .data = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 }, + .size = 10, + }, + /* Fill final cache page. */ + { + .offset = TEST_FILE_SIZE - 10, + .data = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 }, + .size = 10, + }, + /* Read final cache page. */ + { + .offset = TEST_FILE_SIZE - 10, + .data = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 }, + .size = 10, + }, + /* Read final cache page. */ + { + .offset = TEST_FILE_SIZE - 3, + .data = { 7, 8, 9, 0, 0, 0, 0, 0, 0, 0 }, + .size = 3, + }, +}; + +int test__dso_data(void) +{ + struct machine machine; + struct dso *dso; + char *file = test_file(TEST_FILE_SIZE); + size_t i; + + TEST_ASSERT_VAL("No test file", file); + + memset(&machine, 0, sizeof(machine)); + + dso = dso__new((const char *)file); + + /* Basic 10 bytes tests. */ + for (i = 0; i < ARRAY_SIZE(offsets); i++) { + struct test_data_offset *data = &offsets[i]; + ssize_t size; + u8 buf[10]; + + memset(buf, 0, 10); + size = dso__data_read_offset(dso, &machine, data->offset, + buf, 10); + + TEST_ASSERT_VAL("Wrong size", size == data->size); + TEST_ASSERT_VAL("Wrong data", !memcmp(buf, data->data, 10)); + } + + /* Read cross multiple cache pages. */ + { + ssize_t size; + int c; + u8 *buf; + + buf = malloc(TEST_FILE_SIZE); + TEST_ASSERT_VAL("ENOMEM\n", buf); + + /* First iteration to fill caches, second one to read them. */ + for (c = 0; c < 2; c++) { + memset(buf, 0, TEST_FILE_SIZE); + size = dso__data_read_offset(dso, &machine, 10, + buf, TEST_FILE_SIZE); + + TEST_ASSERT_VAL("Wrong size", + size == (TEST_FILE_SIZE - 10)); + + for (i = 0; i < (size_t)size; i++) + TEST_ASSERT_VAL("Wrong data", + buf[i] == (i % 10)); + } + + free(buf); + } + + dso__delete(dso); + unlink(file); + return 0; +} diff --git a/tools/perf/tests/evsel-roundtrip-name.c b/tools/perf/tests/evsel-roundtrip-name.c new file mode 100644 index 00000000..0fd99a9a --- /dev/null +++ b/tools/perf/tests/evsel-roundtrip-name.c @@ -0,0 +1,114 @@ +#include "evlist.h" +#include "evsel.h" +#include "parse-events.h" +#include "tests.h" + +static int perf_evsel__roundtrip_cache_name_test(void) +{ + char name[128]; + int type, op, err = 0, ret = 0, i, idx; + struct perf_evsel *evsel; + struct perf_evlist *evlist = perf_evlist__new(NULL, NULL); + + if (evlist == NULL) + return -ENOMEM; + + for (type = 0; type < PERF_COUNT_HW_CACHE_MAX; type++) { + for (op = 0; op < PERF_COUNT_HW_CACHE_OP_MAX; op++) { + /* skip invalid cache type */ + if (!perf_evsel__is_cache_op_valid(type, op)) + continue; + + for (i = 0; i < PERF_COUNT_HW_CACHE_RESULT_MAX; i++) { + __perf_evsel__hw_cache_type_op_res_name(type, op, i, + name, sizeof(name)); + err = parse_events(evlist, name); + if (err) + ret = err; + } + } + } + + idx = 0; + evsel = perf_evlist__first(evlist); + + for (type = 0; type < PERF_COUNT_HW_CACHE_MAX; type++) { + for (op = 0; op < PERF_COUNT_HW_CACHE_OP_MAX; op++) { + /* skip invalid cache type */ + if (!perf_evsel__is_cache_op_valid(type, op)) + continue; + + for (i = 0; i < PERF_COUNT_HW_CACHE_RESULT_MAX; i++) { + __perf_evsel__hw_cache_type_op_res_name(type, op, i, + name, sizeof(name)); + if (evsel->idx != idx) + continue; + + ++idx; + + if (strcmp(perf_evsel__name(evsel), name)) { + pr_debug("%s != %s\n", perf_evsel__name(evsel), name); + ret = -1; + } + + evsel = perf_evsel__next(evsel); + } + } + } + + perf_evlist__delete(evlist); + return ret; +} + +static int __perf_evsel__name_array_test(const char *names[], int nr_names) +{ + int i, err; + struct perf_evsel *evsel; + struct perf_evlist *evlist = perf_evlist__new(NULL, NULL); + + if (evlist == NULL) + return -ENOMEM; + + for (i = 0; i < nr_names; ++i) { + err = parse_events(evlist, names[i]); + if (err) { + pr_debug("failed to parse event '%s', err %d\n", + names[i], err); + goto out_delete_evlist; + } + } + + err = 0; + list_for_each_entry(evsel, &evlist->entries, node) { + if (strcmp(perf_evsel__name(evsel), names[evsel->idx])) { + --err; + pr_debug("%s != %s\n", perf_evsel__name(evsel), names[evsel->idx]); + } + } + +out_delete_evlist: + perf_evlist__delete(evlist); + return err; +} + +#define perf_evsel__name_array_test(names) \ + __perf_evsel__name_array_test(names, ARRAY_SIZE(names)) + +int test__perf_evsel__roundtrip_name_test(void) +{ + int err = 0, ret = 0; + + err = perf_evsel__name_array_test(perf_evsel__hw_names); + if (err) + ret = err; + + err = perf_evsel__name_array_test(perf_evsel__sw_names); + if (err) + ret = err; + + err = perf_evsel__roundtrip_cache_name_test(); + if (err) + ret = err; + + return ret; +} diff --git a/tools/perf/tests/evsel-tp-sched.c b/tools/perf/tests/evsel-tp-sched.c new file mode 100644 index 00000000..a5d2fcc5 --- /dev/null +++ b/tools/perf/tests/evsel-tp-sched.c @@ -0,0 +1,84 @@ +#include "evsel.h" +#include "tests.h" +#include "event-parse.h" + +static int perf_evsel__test_field(struct perf_evsel *evsel, const char *name, + int size, bool should_be_signed) +{ + struct format_field *field = perf_evsel__field(evsel, name); + int is_signed; + int ret = 0; + + if (field == NULL) { + pr_debug("%s: \"%s\" field not found!\n", evsel->name, name); + return -1; + } + + is_signed = !!(field->flags | FIELD_IS_SIGNED); + if (should_be_signed && !is_signed) { + pr_debug("%s: \"%s\" signedness(%d) is wrong, should be %d\n", + evsel->name, name, is_signed, should_be_signed); + ret = -1; + } + + if (field->size != size) { + pr_debug("%s: \"%s\" size (%d) should be %d!\n", + evsel->name, name, field->size, size); + ret = -1; + } + + return ret; +} + +int test__perf_evsel__tp_sched_test(void) +{ + struct perf_evsel *evsel = perf_evsel__newtp("sched", "sched_switch", 0); + int ret = 0; + + if (evsel == NULL) { + pr_debug("perf_evsel__new\n"); + return -1; + } + + if (perf_evsel__test_field(evsel, "prev_comm", 16, true)) + ret = -1; + + if (perf_evsel__test_field(evsel, "prev_pid", 4, true)) + ret = -1; + + if (perf_evsel__test_field(evsel, "prev_prio", 4, true)) + ret = -1; + + if (perf_evsel__test_field(evsel, "prev_state", 8, true)) + ret = -1; + + if (perf_evsel__test_field(evsel, "next_comm", 16, true)) + ret = -1; + + if (perf_evsel__test_field(evsel, "next_pid", 4, true)) + ret = -1; + + if (perf_evsel__test_field(evsel, "next_prio", 4, true)) + ret = -1; + + perf_evsel__delete(evsel); + + evsel = perf_evsel__newtp("sched", "sched_wakeup", 0); + + if (perf_evsel__test_field(evsel, "comm", 16, true)) + ret = -1; + + if (perf_evsel__test_field(evsel, "pid", 4, true)) + ret = -1; + + if (perf_evsel__test_field(evsel, "prio", 4, true)) + ret = -1; + + if (perf_evsel__test_field(evsel, "success", 4, true)) + ret = -1; + + if (perf_evsel__test_field(evsel, "target_cpu", 4, true)) + ret = -1; + + return ret; +} diff --git a/tools/perf/tests/hists_link.c b/tools/perf/tests/hists_link.c new file mode 100644 index 00000000..1be64a6c --- /dev/null +++ b/tools/perf/tests/hists_link.c @@ -0,0 +1,500 @@ +#include "perf.h" +#include "tests.h" +#include "debug.h" +#include "symbol.h" +#include "sort.h" +#include "evsel.h" +#include "evlist.h" +#include "machine.h" +#include "thread.h" +#include "parse-events.h" + +static struct { + u32 pid; + const char *comm; +} fake_threads[] = { + { 100, "perf" }, + { 200, "perf" }, + { 300, "bash" }, +}; + +static struct { + u32 pid; + u64 start; + const char *filename; +} fake_mmap_info[] = { + { 100, 0x40000, "perf" }, + { 100, 0x50000, "libc" }, + { 100, 0xf0000, "[kernel]" }, + { 200, 0x40000, "perf" }, + { 200, 0x50000, "libc" }, + { 200, 0xf0000, "[kernel]" }, + { 300, 0x40000, "bash" }, + { 300, 0x50000, "libc" }, + { 300, 0xf0000, "[kernel]" }, +}; + +struct fake_sym { + u64 start; + u64 length; + const char *name; +}; + +static struct fake_sym perf_syms[] = { + { 700, 100, "main" }, + { 800, 100, "run_command" }, + { 900, 100, "cmd_record" }, +}; + +static struct fake_sym bash_syms[] = { + { 700, 100, "main" }, + { 800, 100, "xmalloc" }, + { 900, 100, "xfree" }, +}; + +static struct fake_sym libc_syms[] = { + { 700, 100, "malloc" }, + { 800, 100, "free" }, + { 900, 100, "realloc" }, +}; + +static struct fake_sym kernel_syms[] = { + { 700, 100, "schedule" }, + { 800, 100, "page_fault" }, + { 900, 100, "sys_perf_event_open" }, +}; + +static struct { + const char *dso_name; + struct fake_sym *syms; + size_t nr_syms; +} fake_symbols[] = { + { "perf", perf_syms, ARRAY_SIZE(perf_syms) }, + { "bash", bash_syms, ARRAY_SIZE(bash_syms) }, + { "libc", libc_syms, ARRAY_SIZE(libc_syms) }, + { "[kernel]", kernel_syms, ARRAY_SIZE(kernel_syms) }, +}; + +static struct machine *setup_fake_machine(struct machines *machines) +{ + struct machine *machine = machines__find(machines, HOST_KERNEL_ID); + size_t i; + + if (machine == NULL) { + pr_debug("Not enough memory for machine setup\n"); + return NULL; + } + + for (i = 0; i < ARRAY_SIZE(fake_threads); i++) { + struct thread *thread; + + thread = machine__findnew_thread(machine, fake_threads[i].pid); + if (thread == NULL) + goto out; + + thread__set_comm(thread, fake_threads[i].comm); + } + + for (i = 0; i < ARRAY_SIZE(fake_mmap_info); i++) { + union perf_event fake_mmap_event = { + .mmap = { + .header = { .misc = PERF_RECORD_MISC_USER, }, + .pid = fake_mmap_info[i].pid, + .start = fake_mmap_info[i].start, + .len = 0x1000ULL, + .pgoff = 0ULL, + }, + }; + + strcpy(fake_mmap_event.mmap.filename, + fake_mmap_info[i].filename); + + machine__process_mmap_event(machine, &fake_mmap_event); + } + + for (i = 0; i < ARRAY_SIZE(fake_symbols); i++) { + size_t k; + struct dso *dso; + + dso = __dsos__findnew(&machine->user_dsos, + fake_symbols[i].dso_name); + if (dso == NULL) + goto out; + + /* emulate dso__load() */ + dso__set_loaded(dso, MAP__FUNCTION); + + for (k = 0; k < fake_symbols[i].nr_syms; k++) { + struct symbol *sym; + struct fake_sym *fsym = &fake_symbols[i].syms[k]; + + sym = symbol__new(fsym->start, fsym->length, + STB_GLOBAL, fsym->name); + if (sym == NULL) + goto out; + + symbols__insert(&dso->symbols[MAP__FUNCTION], sym); + } + } + + return machine; + +out: + pr_debug("Not enough memory for machine setup\n"); + machine__delete_threads(machine); + machine__delete(machine); + return NULL; +} + +struct sample { + u32 pid; + u64 ip; + struct thread *thread; + struct map *map; + struct symbol *sym; +}; + +static struct sample fake_common_samples[] = { + /* perf [kernel] schedule() */ + { .pid = 100, .ip = 0xf0000 + 700, }, + /* perf [perf] main() */ + { .pid = 200, .ip = 0x40000 + 700, }, + /* perf [perf] cmd_record() */ + { .pid = 200, .ip = 0x40000 + 900, }, + /* bash [bash] xmalloc() */ + { .pid = 300, .ip = 0x40000 + 800, }, + /* bash [libc] malloc() */ + { .pid = 300, .ip = 0x50000 + 700, }, +}; + +static struct sample fake_samples[][5] = { + { + /* perf [perf] run_command() */ + { .pid = 100, .ip = 0x40000 + 800, }, + /* perf [libc] malloc() */ + { .pid = 100, .ip = 0x50000 + 700, }, + /* perf [kernel] page_fault() */ + { .pid = 100, .ip = 0xf0000 + 800, }, + /* perf [kernel] sys_perf_event_open() */ + { .pid = 200, .ip = 0xf0000 + 900, }, + /* bash [libc] free() */ + { .pid = 300, .ip = 0x50000 + 800, }, + }, + { + /* perf [libc] free() */ + { .pid = 200, .ip = 0x50000 + 800, }, + /* bash [libc] malloc() */ + { .pid = 300, .ip = 0x50000 + 700, }, /* will be merged */ + /* bash [bash] xfee() */ + { .pid = 300, .ip = 0x40000 + 900, }, + /* bash [libc] realloc() */ + { .pid = 300, .ip = 0x50000 + 900, }, + /* bash [kernel] page_fault() */ + { .pid = 300, .ip = 0xf0000 + 800, }, + }, +}; + +static int add_hist_entries(struct perf_evlist *evlist, struct machine *machine) +{ + struct perf_evsel *evsel; + struct addr_location al; + struct hist_entry *he; + struct perf_sample sample = { .cpu = 0, }; + size_t i = 0, k; + + /* + * each evsel will have 10 samples - 5 common and 5 distinct. + * However the second evsel also has a collapsed entry for + * "bash [libc] malloc" so total 9 entries will be in the tree. + */ + list_for_each_entry(evsel, &evlist->entries, node) { + for (k = 0; k < ARRAY_SIZE(fake_common_samples); k++) { + const union perf_event event = { + .ip = { + .header = { + .misc = PERF_RECORD_MISC_USER, + }, + .pid = fake_common_samples[k].pid, + .ip = fake_common_samples[k].ip, + }, + }; + + if (perf_event__preprocess_sample(&event, machine, &al, + &sample, 0) < 0) + goto out; + + he = __hists__add_entry(&evsel->hists, &al, NULL, 1); + if (he == NULL) + goto out; + + fake_common_samples[k].thread = al.thread; + fake_common_samples[k].map = al.map; + fake_common_samples[k].sym = al.sym; + } + + for (k = 0; k < ARRAY_SIZE(fake_samples[i]); k++) { + const union perf_event event = { + .ip = { + .header = { + .misc = PERF_RECORD_MISC_USER, + }, + .pid = fake_samples[i][k].pid, + .ip = fake_samples[i][k].ip, + }, + }; + + if (perf_event__preprocess_sample(&event, machine, &al, + &sample, 0) < 0) + goto out; + + he = __hists__add_entry(&evsel->hists, &al, NULL, 1); + if (he == NULL) + goto out; + + fake_samples[i][k].thread = al.thread; + fake_samples[i][k].map = al.map; + fake_samples[i][k].sym = al.sym; + } + i++; + } + + return 0; + +out: + pr_debug("Not enough memory for adding a hist entry\n"); + return -1; +} + +static int find_sample(struct sample *samples, size_t nr_samples, + struct thread *t, struct map *m, struct symbol *s) +{ + while (nr_samples--) { + if (samples->thread == t && samples->map == m && + samples->sym == s) + return 1; + samples++; + } + return 0; +} + +static int __validate_match(struct hists *hists) +{ + size_t count = 0; + struct rb_root *root; + struct rb_node *node; + + /* + * Only entries from fake_common_samples should have a pair. + */ + if (sort__need_collapse) + root = &hists->entries_collapsed; + else + root = hists->entries_in; + + node = rb_first(root); + while (node) { + struct hist_entry *he; + + he = rb_entry(node, struct hist_entry, rb_node_in); + + if (hist_entry__has_pairs(he)) { + if (find_sample(fake_common_samples, + ARRAY_SIZE(fake_common_samples), + he->thread, he->ms.map, he->ms.sym)) { + count++; + } else { + pr_debug("Can't find the matched entry\n"); + return -1; + } + } + + node = rb_next(node); + } + + if (count != ARRAY_SIZE(fake_common_samples)) { + pr_debug("Invalid count for matched entries: %zd of %zd\n", + count, ARRAY_SIZE(fake_common_samples)); + return -1; + } + + return 0; +} + +static int validate_match(struct hists *leader, struct hists *other) +{ + return __validate_match(leader) || __validate_match(other); +} + +static int __validate_link(struct hists *hists, int idx) +{ + size_t count = 0; + size_t count_pair = 0; + size_t count_dummy = 0; + struct rb_root *root; + struct rb_node *node; + + /* + * Leader hists (idx = 0) will have dummy entries from other, + * and some entries will have no pair. However every entry + * in other hists should have (dummy) pair. + */ + if (sort__need_collapse) + root = &hists->entries_collapsed; + else + root = hists->entries_in; + + node = rb_first(root); + while (node) { + struct hist_entry *he; + + he = rb_entry(node, struct hist_entry, rb_node_in); + + if (hist_entry__has_pairs(he)) { + if (!find_sample(fake_common_samples, + ARRAY_SIZE(fake_common_samples), + he->thread, he->ms.map, he->ms.sym) && + !find_sample(fake_samples[idx], + ARRAY_SIZE(fake_samples[idx]), + he->thread, he->ms.map, he->ms.sym)) { + count_dummy++; + } + count_pair++; + } else if (idx) { + pr_debug("A entry from the other hists should have pair\n"); + return -1; + } + + count++; + node = rb_next(node); + } + + /* + * Note that we have a entry collapsed in the other (idx = 1) hists. + */ + if (idx == 0) { + if (count_dummy != ARRAY_SIZE(fake_samples[1]) - 1) { + pr_debug("Invalid count of dummy entries: %zd of %zd\n", + count_dummy, ARRAY_SIZE(fake_samples[1]) - 1); + return -1; + } + if (count != count_pair + ARRAY_SIZE(fake_samples[0])) { + pr_debug("Invalid count of total leader entries: %zd of %zd\n", + count, count_pair + ARRAY_SIZE(fake_samples[0])); + return -1; + } + } else { + if (count != count_pair) { + pr_debug("Invalid count of total other entries: %zd of %zd\n", + count, count_pair); + return -1; + } + if (count_dummy > 0) { + pr_debug("Other hists should not have dummy entries: %zd\n", + count_dummy); + return -1; + } + } + + return 0; +} + +static int validate_link(struct hists *leader, struct hists *other) +{ + return __validate_link(leader, 0) || __validate_link(other, 1); +} + +static void print_hists(struct hists *hists) +{ + int i = 0; + struct rb_root *root; + struct rb_node *node; + + if (sort__need_collapse) + root = &hists->entries_collapsed; + else + root = hists->entries_in; + + pr_info("----- %s --------\n", __func__); + node = rb_first(root); + while (node) { + struct hist_entry *he; + + he = rb_entry(node, struct hist_entry, rb_node_in); + + pr_info("%2d: entry: %-8s [%-8s] %20s: period = %"PRIu64"\n", + i, he->thread->comm, he->ms.map->dso->short_name, + he->ms.sym->name, he->stat.period); + + i++; + node = rb_next(node); + } +} + +int test__hists_link(void) +{ + int err = -1; + struct machines machines; + struct machine *machine = NULL; + struct perf_evsel *evsel, *first; + struct perf_evlist *evlist = perf_evlist__new(NULL, NULL); + + if (evlist == NULL) + return -ENOMEM; + + err = parse_events(evlist, "cpu-clock"); + if (err) + goto out; + err = parse_events(evlist, "task-clock"); + if (err) + goto out; + + /* default sort order (comm,dso,sym) will be used */ + if (setup_sorting() < 0) + goto out; + + machines__init(&machines); + + /* setup threads/dso/map/symbols also */ + machine = setup_fake_machine(&machines); + if (!machine) + goto out; + + if (verbose > 1) + machine__fprintf(machine, stderr); + + /* process sample events */ + err = add_hist_entries(evlist, machine); + if (err < 0) + goto out; + + list_for_each_entry(evsel, &evlist->entries, node) { + hists__collapse_resort(&evsel->hists); + + if (verbose > 2) + print_hists(&evsel->hists); + } + + first = perf_evlist__first(evlist); + evsel = perf_evlist__last(evlist); + + /* match common entries */ + hists__match(&first->hists, &evsel->hists); + err = validate_match(&first->hists, &evsel->hists); + if (err) + goto out; + + /* link common and/or dummy entries */ + hists__link(&first->hists, &evsel->hists); + err = validate_link(&first->hists, &evsel->hists); + if (err) + goto out; + + err = 0; + +out: + /* tear down everything */ + perf_evlist__delete(evlist); + machines__exit(&machines); + + return err; +} diff --git a/tools/perf/tests/mmap-basic.c b/tools/perf/tests/mmap-basic.c new file mode 100644 index 00000000..cdd50755 --- /dev/null +++ b/tools/perf/tests/mmap-basic.c @@ -0,0 +1,148 @@ +#include "evlist.h" +#include "evsel.h" +#include "thread_map.h" +#include "cpumap.h" +#include "tests.h" + +/* + * This test will generate random numbers of calls to some getpid syscalls, + * then establish an mmap for a group of events that are created to monitor + * the syscalls. + * + * It will receive the events, using mmap, use its PERF_SAMPLE_ID generated + * sample.id field to map back to its respective perf_evsel instance. + * + * Then it checks if the number of syscalls reported as perf events by + * the kernel corresponds to the number of syscalls made. + */ +int test__basic_mmap(void) +{ + int err = -1; + union perf_event *event; + struct thread_map *threads; + struct cpu_map *cpus; + struct perf_evlist *evlist; + cpu_set_t cpu_set; + const char *syscall_names[] = { "getsid", "getppid", "getpgrp", + "getpgid", }; + pid_t (*syscalls[])(void) = { (void *)getsid, getppid, getpgrp, + (void*)getpgid }; +#define nsyscalls ARRAY_SIZE(syscall_names) + unsigned int nr_events[nsyscalls], + expected_nr_events[nsyscalls], i, j; + struct perf_evsel *evsels[nsyscalls], *evsel; + + threads = thread_map__new(-1, getpid(), UINT_MAX); + if (threads == NULL) { + pr_debug("thread_map__new\n"); + return -1; + } + + cpus = cpu_map__new(NULL); + if (cpus == NULL) { + pr_debug("cpu_map__new\n"); + goto out_free_threads; + } + + CPU_ZERO(&cpu_set); + CPU_SET(cpus->map[0], &cpu_set); + sched_setaffinity(0, sizeof(cpu_set), &cpu_set); + if (sched_setaffinity(0, sizeof(cpu_set), &cpu_set) < 0) { + pr_debug("sched_setaffinity() failed on CPU %d: %s ", + cpus->map[0], strerror(errno)); + goto out_free_cpus; + } + + evlist = perf_evlist__new(cpus, threads); + if (evlist == NULL) { + pr_debug("perf_evlist__new\n"); + goto out_free_cpus; + } + + for (i = 0; i < nsyscalls; ++i) { + char name[64]; + + snprintf(name, sizeof(name), "sys_enter_%s", syscall_names[i]); + evsels[i] = perf_evsel__newtp("syscalls", name, i); + if (evsels[i] == NULL) { + pr_debug("perf_evsel__new\n"); + goto out_free_evlist; + } + + evsels[i]->attr.wakeup_events = 1; + perf_evsel__set_sample_id(evsels[i]); + + perf_evlist__add(evlist, evsels[i]); + + if (perf_evsel__open(evsels[i], cpus, threads) < 0) { + pr_debug("failed to open counter: %s, " + "tweak /proc/sys/kernel/perf_event_paranoid?\n", + strerror(errno)); + goto out_close_fd; + } + + nr_events[i] = 0; + expected_nr_events[i] = 1 + rand() % 127; + } + + if (perf_evlist__mmap(evlist, 128, true) < 0) { + pr_debug("failed to mmap events: %d (%s)\n", errno, + strerror(errno)); + goto out_close_fd; + } + + for (i = 0; i < nsyscalls; ++i) + for (j = 0; j < expected_nr_events[i]; ++j) { + int foo = syscalls[i](); + ++foo; + } + + while ((event = perf_evlist__mmap_read(evlist, 0)) != NULL) { + struct perf_sample sample; + + if (event->header.type != PERF_RECORD_SAMPLE) { + pr_debug("unexpected %s event\n", + perf_event__name(event->header.type)); + goto out_munmap; + } + + err = perf_evlist__parse_sample(evlist, event, &sample); + if (err) { + pr_err("Can't parse sample, err = %d\n", err); + goto out_munmap; + } + + err = -1; + evsel = perf_evlist__id2evsel(evlist, sample.id); + if (evsel == NULL) { + pr_debug("event with id %" PRIu64 + " doesn't map to an evsel\n", sample.id); + goto out_munmap; + } + nr_events[evsel->idx]++; + } + + err = 0; + list_for_each_entry(evsel, &evlist->entries, node) { + if (nr_events[evsel->idx] != expected_nr_events[evsel->idx]) { + pr_debug("expected %d %s events, got %d\n", + expected_nr_events[evsel->idx], + perf_evsel__name(evsel), nr_events[evsel->idx]); + err = -1; + goto out_munmap; + } + } + +out_munmap: + perf_evlist__munmap(evlist); +out_close_fd: + for (i = 0; i < nsyscalls; ++i) + perf_evsel__close_fd(evsels[i], 1, threads->nr); +out_free_evlist: + perf_evlist__delete(evlist); +out_free_cpus: + cpu_map__delete(cpus); +out_free_threads: + thread_map__delete(threads); + return err; +} diff --git a/tools/perf/tests/open-syscall-all-cpus.c b/tools/perf/tests/open-syscall-all-cpus.c new file mode 100644 index 00000000..b0657a9c --- /dev/null +++ b/tools/perf/tests/open-syscall-all-cpus.c @@ -0,0 +1,109 @@ +#include "evsel.h" +#include "tests.h" +#include "thread_map.h" +#include "cpumap.h" +#include "debug.h" + +int test__open_syscall_event_on_all_cpus(void) +{ + int err = -1, fd, cpu; + struct cpu_map *cpus; + struct perf_evsel *evsel; + unsigned int nr_open_calls = 111, i; + cpu_set_t cpu_set; + struct thread_map *threads = thread_map__new(-1, getpid(), UINT_MAX); + + if (threads == NULL) { + pr_debug("thread_map__new\n"); + return -1; + } + + cpus = cpu_map__new(NULL); + if (cpus == NULL) { + pr_debug("cpu_map__new\n"); + goto out_thread_map_delete; + } + + CPU_ZERO(&cpu_set); + + evsel = perf_evsel__newtp("syscalls", "sys_enter_open", 0); + if (evsel == NULL) { + pr_debug("is debugfs mounted on /sys/kernel/debug?\n"); + goto out_thread_map_delete; + } + + if (perf_evsel__open(evsel, cpus, threads) < 0) { + pr_debug("failed to open counter: %s, " + "tweak /proc/sys/kernel/perf_event_paranoid?\n", + strerror(errno)); + goto out_evsel_delete; + } + + for (cpu = 0; cpu < cpus->nr; ++cpu) { + unsigned int ncalls = nr_open_calls + cpu; + /* + * XXX eventually lift this restriction in a way that + * keeps perf building on older glibc installations + * without CPU_ALLOC. 1024 cpus in 2010 still seems + * a reasonable upper limit tho :-) + */ + if (cpus->map[cpu] >= CPU_SETSIZE) { + pr_debug("Ignoring CPU %d\n", cpus->map[cpu]); + continue; + } + + CPU_SET(cpus->map[cpu], &cpu_set); + if (sched_setaffinity(0, sizeof(cpu_set), &cpu_set) < 0) { + pr_debug("sched_setaffinity() failed on CPU %d: %s ", + cpus->map[cpu], + strerror(errno)); + goto out_close_fd; + } + for (i = 0; i < ncalls; ++i) { + fd = open("/etc/passwd", O_RDONLY); + close(fd); + } + CPU_CLR(cpus->map[cpu], &cpu_set); + } + + /* + * Here we need to explicitely preallocate the counts, as if + * we use the auto allocation it will allocate just for 1 cpu, + * as we start by cpu 0. + */ + if (perf_evsel__alloc_counts(evsel, cpus->nr) < 0) { + pr_debug("perf_evsel__alloc_counts(ncpus=%d)\n", cpus->nr); + goto out_close_fd; + } + + err = 0; + + for (cpu = 0; cpu < cpus->nr; ++cpu) { + unsigned int expected; + + if (cpus->map[cpu] >= CPU_SETSIZE) + continue; + + if (perf_evsel__read_on_cpu(evsel, cpu, 0) < 0) { + pr_debug("perf_evsel__read_on_cpu\n"); + err = -1; + break; + } + + expected = nr_open_calls + cpu; + if (evsel->counts->cpu[cpu].val != expected) { + pr_debug("perf_evsel__read_on_cpu: expected to intercept %d calls on cpu %d, got %" PRIu64 "\n", + expected, cpus->map[cpu], evsel->counts->cpu[cpu].val); + err = -1; + } + } + + perf_evsel__free_counts(evsel); +out_close_fd: + perf_evsel__close_fd(evsel, 1, threads->nr); +out_evsel_delete: + perf_evsel__delete(evsel); +out_thread_map_delete: + thread_map__delete(threads); + return err; +} diff --git a/tools/perf/tests/open-syscall-tp-fields.c b/tools/perf/tests/open-syscall-tp-fields.c new file mode 100644 index 00000000..1c52fdc1 --- /dev/null +++ b/tools/perf/tests/open-syscall-tp-fields.c @@ -0,0 +1,117 @@ +#include "perf.h" +#include "evlist.h" +#include "evsel.h" +#include "thread_map.h" +#include "tests.h" + +int test__syscall_open_tp_fields(void) +{ + struct perf_record_opts opts = { + .target = { + .uid = UINT_MAX, + .uses_mmap = true, + }, + .no_delay = true, + .freq = 1, + .mmap_pages = 256, + .raw_samples = true, + }; + const char *filename = "/etc/passwd"; + int flags = O_RDONLY | O_DIRECTORY; + struct perf_evlist *evlist = perf_evlist__new(NULL, NULL); + struct perf_evsel *evsel; + int err = -1, i, nr_events = 0, nr_polls = 0; + + if (evlist == NULL) { + pr_debug("%s: perf_evlist__new\n", __func__); + goto out; + } + + evsel = perf_evsel__newtp("syscalls", "sys_enter_open", 0); + if (evsel == NULL) { + pr_debug("%s: perf_evsel__newtp\n", __func__); + goto out_delete_evlist; + } + + perf_evlist__add(evlist, evsel); + + err = perf_evlist__create_maps(evlist, &opts.target); + if (err < 0) { + pr_debug("%s: perf_evlist__create_maps\n", __func__); + goto out_delete_evlist; + } + + perf_evsel__config(evsel, &opts); + + evlist->threads->map[0] = getpid(); + + err = perf_evlist__open(evlist); + if (err < 0) { + pr_debug("perf_evlist__open: %s\n", strerror(errno)); + goto out_delete_evlist; + } + + err = perf_evlist__mmap(evlist, UINT_MAX, false); + if (err < 0) { + pr_debug("perf_evlist__mmap: %s\n", strerror(errno)); + goto out_delete_evlist; + } + + perf_evlist__enable(evlist); + + /* + * Generate the event: + */ + open(filename, flags); + + while (1) { + int before = nr_events; + + for (i = 0; i < evlist->nr_mmaps; i++) { + union perf_event *event; + + while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) { + const u32 type = event->header.type; + int tp_flags; + struct perf_sample sample; + + ++nr_events; + + if (type != PERF_RECORD_SAMPLE) + continue; + + err = perf_evsel__parse_sample(evsel, event, &sample); + if (err) { + pr_err("Can't parse sample, err = %d\n", err); + goto out_munmap; + } + + tp_flags = perf_evsel__intval(evsel, &sample, "flags"); + + if (flags != tp_flags) { + pr_debug("%s: Expected flags=%#x, got %#x\n", + __func__, flags, tp_flags); + goto out_munmap; + } + + goto out_ok; + } + } + + if (nr_events == before) + poll(evlist->pollfd, evlist->nr_fds, 10); + + if (++nr_polls > 5) { + pr_debug("%s: no events!\n", __func__); + goto out_munmap; + } + } +out_ok: + err = 0; +out_munmap: + perf_evlist__munmap(evlist); +out_delete_evlist: + perf_evlist__delete(evlist); +out: + return err; +} diff --git a/tools/perf/tests/open-syscall.c b/tools/perf/tests/open-syscall.c new file mode 100644 index 00000000..befc0671 --- /dev/null +++ b/tools/perf/tests/open-syscall.c @@ -0,0 +1,55 @@ +#include "thread_map.h" +#include "evsel.h" +#include "debug.h" +#include "tests.h" + +int test__open_syscall_event(void) +{ + int err = -1, fd; + struct perf_evsel *evsel; + unsigned int nr_open_calls = 111, i; + struct thread_map *threads = thread_map__new(-1, getpid(), UINT_MAX); + + if (threads == NULL) { + pr_debug("thread_map__new\n"); + return -1; + } + + evsel = perf_evsel__newtp("syscalls", "sys_enter_open", 0); + if (evsel == NULL) { + pr_debug("is debugfs mounted on /sys/kernel/debug?\n"); + goto out_thread_map_delete; + } + + if (perf_evsel__open_per_thread(evsel, threads) < 0) { + pr_debug("failed to open counter: %s, " + "tweak /proc/sys/kernel/perf_event_paranoid?\n", + strerror(errno)); + goto out_evsel_delete; + } + + for (i = 0; i < nr_open_calls; ++i) { + fd = open("/etc/passwd", O_RDONLY); + close(fd); + } + + if (perf_evsel__read_on_cpu(evsel, 0, 0) < 0) { + pr_debug("perf_evsel__read_on_cpu\n"); + goto out_close_fd; + } + + if (evsel->counts->cpu[0].val != nr_open_calls) { + pr_debug("perf_evsel__read_on_cpu: expected to intercept %d calls, got %" PRIu64 "\n", + nr_open_calls, evsel->counts->cpu[0].val); + goto out_close_fd; + } + + err = 0; +out_close_fd: + perf_evsel__close_fd(evsel, 1, threads->nr); +out_evsel_delete: + perf_evsel__delete(evsel); +out_thread_map_delete: + thread_map__delete(threads); + return err; +} diff --git a/tools/perf/tests/parse-events.c b/tools/perf/tests/parse-events.c new file mode 100644 index 00000000..c5636f36 --- /dev/null +++ b/tools/perf/tests/parse-events.c @@ -0,0 +1,1383 @@ + +#include "parse-events.h" +#include "evsel.h" +#include "evlist.h" +#include "sysfs.h" +#include "debugfs.h" +#include "tests.h" +#include <linux/hw_breakpoint.h> + +#define TEST_ASSERT_VAL(text, cond) \ +do { \ + if (!(cond)) { \ + pr_debug("FAILED %s:%d %s\n", __FILE__, __LINE__, text); \ + return -1; \ + } \ +} while (0) + +#define PERF_TP_SAMPLE_TYPE (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME | \ + PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD) + +static int test__checkevent_tracepoint(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong number of groups", 0 == evlist->nr_groups); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_TRACEPOINT == evsel->attr.type); + TEST_ASSERT_VAL("wrong sample_type", + PERF_TP_SAMPLE_TYPE == evsel->attr.sample_type); + TEST_ASSERT_VAL("wrong sample_period", 1 == evsel->attr.sample_period); + return 0; +} + +static int test__checkevent_tracepoint_multi(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel; + + TEST_ASSERT_VAL("wrong number of entries", evlist->nr_entries > 1); + TEST_ASSERT_VAL("wrong number of groups", 0 == evlist->nr_groups); + + list_for_each_entry(evsel, &evlist->entries, node) { + TEST_ASSERT_VAL("wrong type", + PERF_TYPE_TRACEPOINT == evsel->attr.type); + TEST_ASSERT_VAL("wrong sample_type", + PERF_TP_SAMPLE_TYPE == evsel->attr.sample_type); + TEST_ASSERT_VAL("wrong sample_period", + 1 == evsel->attr.sample_period); + } + return 0; +} + +static int test__checkevent_raw(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 0x1a == evsel->attr.config); + return 0; +} + +static int test__checkevent_numeric(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", 1 == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 1 == evsel->attr.config); + return 0; +} + +static int test__checkevent_symbolic_name(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config); + return 0; +} + +static int test__checkevent_symbolic_name_config(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong period", + 100000 == evsel->attr.sample_period); + TEST_ASSERT_VAL("wrong config1", + 0 == evsel->attr.config1); + TEST_ASSERT_VAL("wrong config2", + 1 == evsel->attr.config2); + return 0; +} + +static int test__checkevent_symbolic_alias(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_SOFTWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_SW_PAGE_FAULTS == evsel->attr.config); + return 0; +} + +static int test__checkevent_genhw(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HW_CACHE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", (1 << 16) == evsel->attr.config); + return 0; +} + +static int test__checkevent_breakpoint(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_BREAKPOINT == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config); + TEST_ASSERT_VAL("wrong bp_type", (HW_BREAKPOINT_R | HW_BREAKPOINT_W) == + evsel->attr.bp_type); + TEST_ASSERT_VAL("wrong bp_len", HW_BREAKPOINT_LEN_4 == + evsel->attr.bp_len); + return 0; +} + +static int test__checkevent_breakpoint_x(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_BREAKPOINT == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config); + TEST_ASSERT_VAL("wrong bp_type", + HW_BREAKPOINT_X == evsel->attr.bp_type); + TEST_ASSERT_VAL("wrong bp_len", sizeof(long) == evsel->attr.bp_len); + return 0; +} + +static int test__checkevent_breakpoint_r(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", + PERF_TYPE_BREAKPOINT == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config); + TEST_ASSERT_VAL("wrong bp_type", + HW_BREAKPOINT_R == evsel->attr.bp_type); + TEST_ASSERT_VAL("wrong bp_len", + HW_BREAKPOINT_LEN_4 == evsel->attr.bp_len); + return 0; +} + +static int test__checkevent_breakpoint_w(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", + PERF_TYPE_BREAKPOINT == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config); + TEST_ASSERT_VAL("wrong bp_type", + HW_BREAKPOINT_W == evsel->attr.bp_type); + TEST_ASSERT_VAL("wrong bp_len", + HW_BREAKPOINT_LEN_4 == evsel->attr.bp_len); + return 0; +} + +static int test__checkevent_breakpoint_rw(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", + PERF_TYPE_BREAKPOINT == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config); + TEST_ASSERT_VAL("wrong bp_type", + (HW_BREAKPOINT_R|HW_BREAKPOINT_W) == evsel->attr.bp_type); + TEST_ASSERT_VAL("wrong bp_len", + HW_BREAKPOINT_LEN_4 == evsel->attr.bp_len); + return 0; +} + +static int test__checkevent_tracepoint_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + + return test__checkevent_tracepoint(evlist); +} + +static int +test__checkevent_tracepoint_multi_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel; + + TEST_ASSERT_VAL("wrong number of entries", evlist->nr_entries > 1); + + list_for_each_entry(evsel, &evlist->entries, node) { + TEST_ASSERT_VAL("wrong exclude_user", + !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", + evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + } + + return test__checkevent_tracepoint_multi(evlist); +} + +static int test__checkevent_raw_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip); + + return test__checkevent_raw(evlist); +} + +static int test__checkevent_numeric_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip); + + return test__checkevent_numeric(evlist); +} + +static int test__checkevent_symbolic_name_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + + return test__checkevent_symbolic_name(evlist); +} + +static int test__checkevent_exclude_host_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host); + + return test__checkevent_symbolic_name(evlist); +} + +static int test__checkevent_exclude_guest_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + + return test__checkevent_symbolic_name(evlist); +} + +static int test__checkevent_symbolic_alias_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + + return test__checkevent_symbolic_alias(evlist); +} + +static int test__checkevent_genhw_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip); + + return test__checkevent_genhw(evlist); +} + +static int test__checkevent_breakpoint_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong name", + !strcmp(perf_evsel__name(evsel), "mem:0:u")); + + return test__checkevent_breakpoint(evlist); +} + +static int test__checkevent_breakpoint_x_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong name", + !strcmp(perf_evsel__name(evsel), "mem:0:x:k")); + + return test__checkevent_breakpoint_x(evlist); +} + +static int test__checkevent_breakpoint_r_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong name", + !strcmp(perf_evsel__name(evsel), "mem:0:r:hp")); + + return test__checkevent_breakpoint_r(evlist); +} + +static int test__checkevent_breakpoint_w_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong name", + !strcmp(perf_evsel__name(evsel), "mem:0:w:up")); + + return test__checkevent_breakpoint_w(evlist); +} + +static int test__checkevent_breakpoint_rw_modifier(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong name", + !strcmp(perf_evsel__name(evsel), "mem:0:rw:kp")); + + return test__checkevent_breakpoint_rw(evlist); +} + +static int test__checkevent_pmu(struct perf_evlist *evlist) +{ + + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 10 == evsel->attr.config); + TEST_ASSERT_VAL("wrong config1", 1 == evsel->attr.config1); + TEST_ASSERT_VAL("wrong config2", 3 == evsel->attr.config2); + TEST_ASSERT_VAL("wrong period", 1000 == evsel->attr.sample_period); + + return 0; +} + +static int test__checkevent_list(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + TEST_ASSERT_VAL("wrong number of entries", 3 == evlist->nr_entries); + + /* r1 */ + TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 1 == evsel->attr.config); + TEST_ASSERT_VAL("wrong config1", 0 == evsel->attr.config1); + TEST_ASSERT_VAL("wrong config2", 0 == evsel->attr.config2); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + + /* syscalls:sys_enter_open:k */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_TRACEPOINT == evsel->attr.type); + TEST_ASSERT_VAL("wrong sample_type", + PERF_TP_SAMPLE_TYPE == evsel->attr.sample_type); + TEST_ASSERT_VAL("wrong sample_period", 1 == evsel->attr.sample_period); + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + + /* 1:1:hp */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", 1 == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 1 == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip); + + return 0; +} + +static int test__checkevent_pmu_name(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + + /* cpu/config=1,name=krava/u */ + TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 1 == evsel->attr.config); + TEST_ASSERT_VAL("wrong name", !strcmp(perf_evsel__name(evsel), "krava")); + + /* cpu/config=2/u" */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 2 == evsel->attr.config); + TEST_ASSERT_VAL("wrong name", + !strcmp(perf_evsel__name(evsel), "cpu/config=2/u")); + + return 0; +} + +static int test__checkevent_pmu_events(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel; + + evsel = list_entry(evlist->entries.next, struct perf_evsel, node); + TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type); + TEST_ASSERT_VAL("wrong exclude_user", + !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", + evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + + return 0; +} + +static int test__checkterms_simple(struct list_head *terms) +{ + struct parse_events_term *term; + + /* config=10 */ + term = list_entry(terms->next, struct parse_events_term, list); + TEST_ASSERT_VAL("wrong type term", + term->type_term == PARSE_EVENTS__TERM_TYPE_CONFIG); + TEST_ASSERT_VAL("wrong type val", + term->type_val == PARSE_EVENTS__TERM_TYPE_NUM); + TEST_ASSERT_VAL("wrong val", term->val.num == 10); + TEST_ASSERT_VAL("wrong config", !term->config); + + /* config1 */ + term = list_entry(term->list.next, struct parse_events_term, list); + TEST_ASSERT_VAL("wrong type term", + term->type_term == PARSE_EVENTS__TERM_TYPE_CONFIG1); + TEST_ASSERT_VAL("wrong type val", + term->type_val == PARSE_EVENTS__TERM_TYPE_NUM); + TEST_ASSERT_VAL("wrong val", term->val.num == 1); + TEST_ASSERT_VAL("wrong config", !term->config); + + /* config2=3 */ + term = list_entry(term->list.next, struct parse_events_term, list); + TEST_ASSERT_VAL("wrong type term", + term->type_term == PARSE_EVENTS__TERM_TYPE_CONFIG2); + TEST_ASSERT_VAL("wrong type val", + term->type_val == PARSE_EVENTS__TERM_TYPE_NUM); + TEST_ASSERT_VAL("wrong val", term->val.num == 3); + TEST_ASSERT_VAL("wrong config", !term->config); + + /* umask=1*/ + term = list_entry(term->list.next, struct parse_events_term, list); + TEST_ASSERT_VAL("wrong type term", + term->type_term == PARSE_EVENTS__TERM_TYPE_USER); + TEST_ASSERT_VAL("wrong type val", + term->type_val == PARSE_EVENTS__TERM_TYPE_NUM); + TEST_ASSERT_VAL("wrong val", term->val.num == 1); + TEST_ASSERT_VAL("wrong config", !strcmp(term->config, "umask")); + + return 0; +} + +static int test__group1(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel, *leader; + + TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups); + + /* instructions:k */ + evsel = leader = perf_evlist__first(evlist); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* cycles:upp */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + /* use of precise requires exclude_guest */ + TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip == 2); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + return 0; +} + +static int test__group2(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel, *leader; + + TEST_ASSERT_VAL("wrong number of entries", 3 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups); + + /* faults + :ku modifier */ + evsel = leader = perf_evlist__first(evlist); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_SOFTWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_SW_PAGE_FAULTS == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* cache-references + :u modifier */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CACHE_REFERENCES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + /* cycles:k */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + + return 0; +} + +static int test__group3(struct perf_evlist *evlist __maybe_unused) +{ + struct perf_evsel *evsel, *leader; + + TEST_ASSERT_VAL("wrong number of entries", 5 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong number of groups", 2 == evlist->nr_groups); + + /* group1 syscalls:sys_enter_open:H */ + evsel = leader = perf_evlist__first(evlist); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_TRACEPOINT == evsel->attr.type); + TEST_ASSERT_VAL("wrong sample_type", + PERF_TP_SAMPLE_TYPE == evsel->attr.sample_type); + TEST_ASSERT_VAL("wrong sample_period", 1 == evsel->attr.sample_period); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong group name", + !strcmp(leader->group_name, "group1")); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* group1 cycles:kppp */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + /* use of precise requires exclude_guest */ + TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip == 3); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group name", !evsel->group_name); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + /* group2 cycles + G modifier */ + evsel = leader = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong group name", + !strcmp(leader->group_name, "group2")); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* group2 1:3 + G modifier */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", 1 == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", 3 == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + /* instructions:u */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + + return 0; +} + +static int test__group4(struct perf_evlist *evlist __maybe_unused) +{ + struct perf_evsel *evsel, *leader; + + TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups); + + /* cycles:u + p */ + evsel = leader = perf_evlist__first(evlist); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + /* use of precise requires exclude_guest */ + TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip == 1); + TEST_ASSERT_VAL("wrong group name", !evsel->group_name); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* instructions:kp + p */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + /* use of precise requires exclude_guest */ + TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip == 2); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + return 0; +} + +static int test__group5(struct perf_evlist *evlist __maybe_unused) +{ + struct perf_evsel *evsel, *leader; + + TEST_ASSERT_VAL("wrong number of entries", 5 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong number of groups", 2 == evlist->nr_groups); + + /* cycles + G */ + evsel = leader = perf_evlist__first(evlist); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong group name", !evsel->group_name); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* instructions + G */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + /* cycles:G */ + evsel = leader = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong group name", !evsel->group_name); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* instructions:G */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + /* cycles */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + + return 0; +} + +static int test__group_gh1(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel, *leader; + + TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups); + + /* cycles + :H group modifier */ + evsel = leader = perf_evlist__first(evlist); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong group name", !evsel->group_name); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* cache-misses:G + :H group modifier */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CACHE_MISSES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + return 0; +} + +static int test__group_gh2(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel, *leader; + + TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups); + + /* cycles + :G group modifier */ + evsel = leader = perf_evlist__first(evlist); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong group name", !evsel->group_name); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* cache-misses:H + :G group modifier */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CACHE_MISSES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + return 0; +} + +static int test__group_gh3(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel, *leader; + + TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups); + + /* cycles:G + :u group modifier */ + evsel = leader = perf_evlist__first(evlist); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong group name", !evsel->group_name); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* cache-misses:H + :u group modifier */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CACHE_MISSES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + return 0; +} + +static int test__group_gh4(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel, *leader; + + TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries); + TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups); + + /* cycles:G + :uG group modifier */ + evsel = leader = perf_evlist__first(evlist); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong group name", !evsel->group_name); + TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel)); + TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0); + + /* cache-misses:H + :uG group modifier */ + evsel = perf_evsel__next(evsel); + TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type); + TEST_ASSERT_VAL("wrong config", + PERF_COUNT_HW_CACHE_MISSES == evsel->attr.config); + TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user); + TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel); + TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv); + TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host); + TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip); + TEST_ASSERT_VAL("wrong leader", evsel->leader == leader); + TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1); + + return 0; +} + +static int count_tracepoints(void) +{ + char events_path[PATH_MAX]; + struct dirent *events_ent; + DIR *events_dir; + int cnt = 0; + + scnprintf(events_path, PATH_MAX, "%s/tracing/events", + debugfs_find_mountpoint()); + + events_dir = opendir(events_path); + + TEST_ASSERT_VAL("Can't open events dir", events_dir); + + while ((events_ent = readdir(events_dir))) { + char sys_path[PATH_MAX]; + struct dirent *sys_ent; + DIR *sys_dir; + + if (!strcmp(events_ent->d_name, ".") + || !strcmp(events_ent->d_name, "..") + || !strcmp(events_ent->d_name, "enable") + || !strcmp(events_ent->d_name, "header_event") + || !strcmp(events_ent->d_name, "header_page")) + continue; + + scnprintf(sys_path, PATH_MAX, "%s/%s", + events_path, events_ent->d_name); + + sys_dir = opendir(sys_path); + TEST_ASSERT_VAL("Can't open sys dir", sys_dir); + + while ((sys_ent = readdir(sys_dir))) { + if (!strcmp(sys_ent->d_name, ".") + || !strcmp(sys_ent->d_name, "..") + || !strcmp(sys_ent->d_name, "enable") + || !strcmp(sys_ent->d_name, "filter")) + continue; + + cnt++; + } + + closedir(sys_dir); + } + + closedir(events_dir); + return cnt; +} + +static int test__all_tracepoints(struct perf_evlist *evlist) +{ + TEST_ASSERT_VAL("wrong events count", + count_tracepoints() == evlist->nr_entries); + + return test__checkevent_tracepoint_multi(evlist); +} + +struct evlist_test { + const char *name; + __u32 type; + int (*check)(struct perf_evlist *evlist); +}; + +static struct evlist_test test__events[] = { + [0] = { + .name = "syscalls:sys_enter_open", + .check = test__checkevent_tracepoint, + }, + [1] = { + .name = "syscalls:*", + .check = test__checkevent_tracepoint_multi, + }, + [2] = { + .name = "r1a", + .check = test__checkevent_raw, + }, + [3] = { + .name = "1:1", + .check = test__checkevent_numeric, + }, + [4] = { + .name = "instructions", + .check = test__checkevent_symbolic_name, + }, + [5] = { + .name = "cycles/period=100000,config2/", + .check = test__checkevent_symbolic_name_config, + }, + [6] = { + .name = "faults", + .check = test__checkevent_symbolic_alias, + }, + [7] = { + .name = "L1-dcache-load-miss", + .check = test__checkevent_genhw, + }, + [8] = { + .name = "mem:0", + .check = test__checkevent_breakpoint, + }, + [9] = { + .name = "mem:0:x", + .check = test__checkevent_breakpoint_x, + }, + [10] = { + .name = "mem:0:r", + .check = test__checkevent_breakpoint_r, + }, + [11] = { + .name = "mem:0:w", + .check = test__checkevent_breakpoint_w, + }, + [12] = { + .name = "syscalls:sys_enter_open:k", + .check = test__checkevent_tracepoint_modifier, + }, + [13] = { + .name = "syscalls:*:u", + .check = test__checkevent_tracepoint_multi_modifier, + }, + [14] = { + .name = "r1a:kp", + .check = test__checkevent_raw_modifier, + }, + [15] = { + .name = "1:1:hp", + .check = test__checkevent_numeric_modifier, + }, + [16] = { + .name = "instructions:h", + .check = test__checkevent_symbolic_name_modifier, + }, + [17] = { + .name = "faults:u", + .check = test__checkevent_symbolic_alias_modifier, + }, + [18] = { + .name = "L1-dcache-load-miss:kp", + .check = test__checkevent_genhw_modifier, + }, + [19] = { + .name = "mem:0:u", + .check = test__checkevent_breakpoint_modifier, + }, + [20] = { + .name = "mem:0:x:k", + .check = test__checkevent_breakpoint_x_modifier, + }, + [21] = { + .name = "mem:0:r:hp", + .check = test__checkevent_breakpoint_r_modifier, + }, + [22] = { + .name = "mem:0:w:up", + .check = test__checkevent_breakpoint_w_modifier, + }, + [23] = { + .name = "r1,syscalls:sys_enter_open:k,1:1:hp", + .check = test__checkevent_list, + }, + [24] = { + .name = "instructions:G", + .check = test__checkevent_exclude_host_modifier, + }, + [25] = { + .name = "instructions:H", + .check = test__checkevent_exclude_guest_modifier, + }, + [26] = { + .name = "mem:0:rw", + .check = test__checkevent_breakpoint_rw, + }, + [27] = { + .name = "mem:0:rw:kp", + .check = test__checkevent_breakpoint_rw_modifier, + }, + [28] = { + .name = "{instructions:k,cycles:upp}", + .check = test__group1, + }, + [29] = { + .name = "{faults:k,cache-references}:u,cycles:k", + .check = test__group2, + }, + [30] = { + .name = "group1{syscalls:sys_enter_open:H,cycles:kppp},group2{cycles,1:3}:G,instructions:u", + .check = test__group3, + }, + [31] = { + .name = "{cycles:u,instructions:kp}:p", + .check = test__group4, + }, + [32] = { + .name = "{cycles,instructions}:G,{cycles:G,instructions:G},cycles", + .check = test__group5, + }, + [33] = { + .name = "*:*", + .check = test__all_tracepoints, + }, + [34] = { + .name = "{cycles,cache-misses:G}:H", + .check = test__group_gh1, + }, + [35] = { + .name = "{cycles,cache-misses:H}:G", + .check = test__group_gh2, + }, + [36] = { + .name = "{cycles:G,cache-misses:H}:u", + .check = test__group_gh3, + }, + [37] = { + .name = "{cycles:G,cache-misses:H}:uG", + .check = test__group_gh4, + }, +}; + +static struct evlist_test test__events_pmu[] = { + [0] = { + .name = "cpu/config=10,config1,config2=3,period=1000/u", + .check = test__checkevent_pmu, + }, + [1] = { + .name = "cpu/config=1,name=krava/u,cpu/config=2/u", + .check = test__checkevent_pmu_name, + }, +}; + +struct terms_test { + const char *str; + __u32 type; + int (*check)(struct list_head *terms); +}; + +static struct terms_test test__terms[] = { + [0] = { + .str = "config=10,config1,config2=3,umask=1", + .check = test__checkterms_simple, + }, +}; + +static int test_event(struct evlist_test *e) +{ + struct perf_evlist *evlist; + int ret; + + evlist = perf_evlist__new(NULL, NULL); + if (evlist == NULL) + return -ENOMEM; + + ret = parse_events(evlist, e->name); + if (ret) { + pr_debug("failed to parse event '%s', err %d\n", + e->name, ret); + return ret; + } + + ret = e->check(evlist); + perf_evlist__delete(evlist); + + return ret; +} + +static int test_events(struct evlist_test *events, unsigned cnt) +{ + int ret1, ret2 = 0; + unsigned i; + + for (i = 0; i < cnt; i++) { + struct evlist_test *e = &events[i]; + + pr_debug("running test %d '%s'\n", i, e->name); + ret1 = test_event(e); + if (ret1) + ret2 = ret1; + } + + return ret2; +} + +static int test_term(struct terms_test *t) +{ + struct list_head *terms; + int ret; + + terms = malloc(sizeof(*terms)); + if (!terms) + return -ENOMEM; + + INIT_LIST_HEAD(terms); + + ret = parse_events_terms(terms, t->str); + if (ret) { + pr_debug("failed to parse terms '%s', err %d\n", + t->str , ret); + return ret; + } + + ret = t->check(terms); + parse_events__free_terms(terms); + + return ret; +} + +static int test_terms(struct terms_test *terms, unsigned cnt) +{ + int ret = 0; + unsigned i; + + for (i = 0; i < cnt; i++) { + struct terms_test *t = &terms[i]; + + pr_debug("running test %d '%s'\n", i, t->str); + ret = test_term(t); + if (ret) + break; + } + + return ret; +} + +static int test_pmu(void) +{ + struct stat st; + char path[PATH_MAX]; + int ret; + + snprintf(path, PATH_MAX, "%s/bus/event_source/devices/cpu/format/", + sysfs_find_mountpoint()); + + ret = stat(path, &st); + if (ret) + pr_debug("omitting PMU cpu tests\n"); + return !ret; +} + +static int test_pmu_events(void) +{ + struct stat st; + char path[PATH_MAX]; + struct dirent *ent; + DIR *dir; + int ret; + + snprintf(path, PATH_MAX, "%s/bus/event_source/devices/cpu/events/", + sysfs_find_mountpoint()); + + ret = stat(path, &st); + if (ret) { + pr_debug("ommiting PMU cpu events tests\n"); + return 0; + } + + dir = opendir(path); + if (!dir) { + pr_debug("can't open pmu event dir"); + return -1; + } + + while (!ret && (ent = readdir(dir))) { +#define MAX_NAME 100 + struct evlist_test e; + char name[MAX_NAME]; + + if (!strcmp(ent->d_name, ".") || + !strcmp(ent->d_name, "..")) + continue; + + snprintf(name, MAX_NAME, "cpu/event=%s/u", ent->d_name); + + e.name = name; + e.check = test__checkevent_pmu_events; + + ret = test_event(&e); +#undef MAX_NAME + } + + closedir(dir); + return ret; +} + +int test__parse_events(void) +{ + int ret1, ret2 = 0; + +#define TEST_EVENTS(tests) \ +do { \ + ret1 = test_events(tests, ARRAY_SIZE(tests)); \ + if (!ret2) \ + ret2 = ret1; \ +} while (0) + + TEST_EVENTS(test__events); + + if (test_pmu()) + TEST_EVENTS(test__events_pmu); + + if (test_pmu()) { + int ret = test_pmu_events(); + if (ret) + return ret; + } + + ret1 = test_terms(test__terms, ARRAY_SIZE(test__terms)); + if (!ret2) + ret2 = ret1; + + return ret2; +} diff --git a/tools/perf/tests/perf-record.c b/tools/perf/tests/perf-record.c new file mode 100644 index 00000000..1e8e5128 --- /dev/null +++ b/tools/perf/tests/perf-record.c @@ -0,0 +1,314 @@ +#include <sched.h> +#include "evlist.h" +#include "evsel.h" +#include "perf.h" +#include "debug.h" +#include "tests.h" + +static int sched__get_first_possible_cpu(pid_t pid, cpu_set_t *maskp) +{ + int i, cpu = -1, nrcpus = 1024; +realloc: + CPU_ZERO(maskp); + + if (sched_getaffinity(pid, sizeof(*maskp), maskp) == -1) { + if (errno == EINVAL && nrcpus < (1024 << 8)) { + nrcpus = nrcpus << 2; + goto realloc; + } + perror("sched_getaffinity"); + return -1; + } + + for (i = 0; i < nrcpus; i++) { + if (CPU_ISSET(i, maskp)) { + if (cpu == -1) + cpu = i; + else + CPU_CLR(i, maskp); + } + } + + return cpu; +} + +int test__PERF_RECORD(void) +{ + struct perf_record_opts opts = { + .target = { + .uid = UINT_MAX, + .uses_mmap = true, + }, + .no_delay = true, + .freq = 10, + .mmap_pages = 256, + }; + cpu_set_t cpu_mask; + size_t cpu_mask_size = sizeof(cpu_mask); + struct perf_evlist *evlist = perf_evlist__new(NULL, NULL); + struct perf_evsel *evsel; + struct perf_sample sample; + const char *cmd = "sleep"; + const char *argv[] = { cmd, "1", NULL, }; + char *bname; + u64 prev_time = 0; + bool found_cmd_mmap = false, + found_libc_mmap = false, + found_vdso_mmap = false, + found_ld_mmap = false; + int err = -1, errs = 0, i, wakeups = 0; + u32 cpu; + int total_events = 0, nr_events[PERF_RECORD_MAX] = { 0, }; + + if (evlist == NULL || argv == NULL) { + pr_debug("Not enough memory to create evlist\n"); + goto out; + } + + /* + * We need at least one evsel in the evlist, use the default + * one: "cycles". + */ + err = perf_evlist__add_default(evlist); + if (err < 0) { + pr_debug("Not enough memory to create evsel\n"); + goto out_delete_evlist; + } + + /* + * Create maps of threads and cpus to monitor. In this case + * we start with all threads and cpus (-1, -1) but then in + * perf_evlist__prepare_workload we'll fill in the only thread + * we're monitoring, the one forked there. + */ + err = perf_evlist__create_maps(evlist, &opts.target); + if (err < 0) { + pr_debug("Not enough memory to create thread/cpu maps\n"); + goto out_delete_evlist; + } + + /* + * Prepare the workload in argv[] to run, it'll fork it, and then wait + * for perf_evlist__start_workload() to exec it. This is done this way + * so that we have time to open the evlist (calling sys_perf_event_open + * on all the fds) and then mmap them. + */ + err = perf_evlist__prepare_workload(evlist, &opts, argv); + if (err < 0) { + pr_debug("Couldn't run the workload!\n"); + goto out_delete_maps; + } + + /* + * Config the evsels, setting attr->comm on the first one, etc. + */ + evsel = perf_evlist__first(evlist); + perf_evsel__set_sample_bit(evsel, CPU); + perf_evsel__set_sample_bit(evsel, TID); + perf_evsel__set_sample_bit(evsel, TIME); + perf_evlist__config(evlist, &opts); + + err = sched__get_first_possible_cpu(evlist->workload.pid, &cpu_mask); + if (err < 0) { + pr_debug("sched__get_first_possible_cpu: %s\n", strerror(errno)); + goto out_delete_maps; + } + + cpu = err; + + /* + * So that we can check perf_sample.cpu on all the samples. + */ + if (sched_setaffinity(evlist->workload.pid, cpu_mask_size, &cpu_mask) < 0) { + pr_debug("sched_setaffinity: %s\n", strerror(errno)); + goto out_delete_maps; + } + + /* + * Call sys_perf_event_open on all the fds on all the evsels, + * grouping them if asked to. + */ + err = perf_evlist__open(evlist); + if (err < 0) { + pr_debug("perf_evlist__open: %s\n", strerror(errno)); + goto out_delete_maps; + } + + /* + * mmap the first fd on a given CPU and ask for events for the other + * fds in the same CPU to be injected in the same mmap ring buffer + * (using ioctl(PERF_EVENT_IOC_SET_OUTPUT)). + */ + err = perf_evlist__mmap(evlist, opts.mmap_pages, false); + if (err < 0) { + pr_debug("perf_evlist__mmap: %s\n", strerror(errno)); + goto out_delete_maps; + } + + /* + * Now that all is properly set up, enable the events, they will + * count just on workload.pid, which will start... + */ + perf_evlist__enable(evlist); + + /* + * Now! + */ + perf_evlist__start_workload(evlist); + + while (1) { + int before = total_events; + + for (i = 0; i < evlist->nr_mmaps; i++) { + union perf_event *event; + + while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) { + const u32 type = event->header.type; + const char *name = perf_event__name(type); + + ++total_events; + if (type < PERF_RECORD_MAX) + nr_events[type]++; + + err = perf_evlist__parse_sample(evlist, event, &sample); + if (err < 0) { + if (verbose) + perf_event__fprintf(event, stderr); + pr_debug("Couldn't parse sample\n"); + goto out_err; + } + + if (verbose) { + pr_info("%" PRIu64" %d ", sample.time, sample.cpu); + perf_event__fprintf(event, stderr); + } + + if (prev_time > sample.time) { + pr_debug("%s going backwards in time, prev=%" PRIu64 ", curr=%" PRIu64 "\n", + name, prev_time, sample.time); + ++errs; + } + + prev_time = sample.time; + + if (sample.cpu != cpu) { + pr_debug("%s with unexpected cpu, expected %d, got %d\n", + name, cpu, sample.cpu); + ++errs; + } + + if ((pid_t)sample.pid != evlist->workload.pid) { + pr_debug("%s with unexpected pid, expected %d, got %d\n", + name, evlist->workload.pid, sample.pid); + ++errs; + } + + if ((pid_t)sample.tid != evlist->workload.pid) { + pr_debug("%s with unexpected tid, expected %d, got %d\n", + name, evlist->workload.pid, sample.tid); + ++errs; + } + + if ((type == PERF_RECORD_COMM || + type == PERF_RECORD_MMAP || + type == PERF_RECORD_FORK || + type == PERF_RECORD_EXIT) && + (pid_t)event->comm.pid != evlist->workload.pid) { + pr_debug("%s with unexpected pid/tid\n", name); + ++errs; + } + + if ((type == PERF_RECORD_COMM || + type == PERF_RECORD_MMAP) && + event->comm.pid != event->comm.tid) { + pr_debug("%s with different pid/tid!\n", name); + ++errs; + } + + switch (type) { + case PERF_RECORD_COMM: + if (strcmp(event->comm.comm, cmd)) { + pr_debug("%s with unexpected comm!\n", name); + ++errs; + } + break; + case PERF_RECORD_EXIT: + goto found_exit; + case PERF_RECORD_MMAP: + bname = strrchr(event->mmap.filename, '/'); + if (bname != NULL) { + if (!found_cmd_mmap) + found_cmd_mmap = !strcmp(bname + 1, cmd); + if (!found_libc_mmap) + found_libc_mmap = !strncmp(bname + 1, "libc", 4); + if (!found_ld_mmap) + found_ld_mmap = !strncmp(bname + 1, "ld", 2); + } else if (!found_vdso_mmap) + found_vdso_mmap = !strcmp(event->mmap.filename, "[vdso]"); + break; + + case PERF_RECORD_SAMPLE: + /* Just ignore samples for now */ + break; + default: + pr_debug("Unexpected perf_event->header.type %d!\n", + type); + ++errs; + } + } + } + + /* + * We don't use poll here because at least at 3.1 times the + * PERF_RECORD_{!SAMPLE} events don't honour + * perf_event_attr.wakeup_events, just PERF_EVENT_SAMPLE does. + */ + if (total_events == before && false) + poll(evlist->pollfd, evlist->nr_fds, -1); + + sleep(1); + if (++wakeups > 5) { + pr_debug("No PERF_RECORD_EXIT event!\n"); + break; + } + } + +found_exit: + if (nr_events[PERF_RECORD_COMM] > 1) { + pr_debug("Excessive number of PERF_RECORD_COMM events!\n"); + ++errs; + } + + if (nr_events[PERF_RECORD_COMM] == 0) { + pr_debug("Missing PERF_RECORD_COMM for %s!\n", cmd); + ++errs; + } + + if (!found_cmd_mmap) { + pr_debug("PERF_RECORD_MMAP for %s missing!\n", cmd); + ++errs; + } + + if (!found_libc_mmap) { + pr_debug("PERF_RECORD_MMAP for %s missing!\n", "libc"); + ++errs; + } + + if (!found_ld_mmap) { + pr_debug("PERF_RECORD_MMAP for %s missing!\n", "ld"); + ++errs; + } + + if (!found_vdso_mmap) { + pr_debug("PERF_RECORD_MMAP for %s missing!\n", "[vdso]"); + ++errs; + } +out_err: + perf_evlist__munmap(evlist); +out_delete_maps: + perf_evlist__delete_maps(evlist); +out_delete_evlist: + perf_evlist__delete(evlist); +out: + return (err < 0 || errs > 0) ? -1 : 0; +} diff --git a/tools/perf/tests/pmu.c b/tools/perf/tests/pmu.c new file mode 100644 index 00000000..12b322fa --- /dev/null +++ b/tools/perf/tests/pmu.c @@ -0,0 +1,173 @@ +#include "parse-events.h" +#include "pmu.h" +#include "util.h" +#include "tests.h" + +/* Simulated format definitions. */ +static struct test_format { + const char *name; + const char *value; +} test_formats[] = { + { "krava01", "config:0-1,62-63\n", }, + { "krava02", "config:10-17\n", }, + { "krava03", "config:5\n", }, + { "krava11", "config1:0,2,4,6,8,20-28\n", }, + { "krava12", "config1:63\n", }, + { "krava13", "config1:45-47\n", }, + { "krava21", "config2:0-3,10-13,20-23,30-33,40-43,50-53,60-63\n", }, + { "krava22", "config2:8,18,48,58\n", }, + { "krava23", "config2:28-29,38\n", }, +}; + +/* Simulated users input. */ +static struct parse_events_term test_terms[] = { + { + .config = (char *) "krava01", + .val.num = 15, + .type_val = PARSE_EVENTS__TERM_TYPE_NUM, + .type_term = PARSE_EVENTS__TERM_TYPE_USER, + }, + { + .config = (char *) "krava02", + .val.num = 170, + .type_val = PARSE_EVENTS__TERM_TYPE_NUM, + .type_term = PARSE_EVENTS__TERM_TYPE_USER, + }, + { + .config = (char *) "krava03", + .val.num = 1, + .type_val = PARSE_EVENTS__TERM_TYPE_NUM, + .type_term = PARSE_EVENTS__TERM_TYPE_USER, + }, + { + .config = (char *) "krava11", + .val.num = 27, + .type_val = PARSE_EVENTS__TERM_TYPE_NUM, + .type_term = PARSE_EVENTS__TERM_TYPE_USER, + }, + { + .config = (char *) "krava12", + .val.num = 1, + .type_val = PARSE_EVENTS__TERM_TYPE_NUM, + .type_term = PARSE_EVENTS__TERM_TYPE_USER, + }, + { + .config = (char *) "krava13", + .val.num = 2, + .type_val = PARSE_EVENTS__TERM_TYPE_NUM, + .type_term = PARSE_EVENTS__TERM_TYPE_USER, + }, + { + .config = (char *) "krava21", + .val.num = 119, + .type_val = PARSE_EVENTS__TERM_TYPE_NUM, + .type_term = PARSE_EVENTS__TERM_TYPE_USER, + }, + { + .config = (char *) "krava22", + .val.num = 11, + .type_val = PARSE_EVENTS__TERM_TYPE_NUM, + .type_term = PARSE_EVENTS__TERM_TYPE_USER, + }, + { + .config = (char *) "krava23", + .val.num = 2, + .type_val = PARSE_EVENTS__TERM_TYPE_NUM, + .type_term = PARSE_EVENTS__TERM_TYPE_USER, + }, +}; + +/* + * Prepare format directory data, exported by kernel + * at /sys/bus/event_source/devices/<dev>/format. + */ +static char *test_format_dir_get(void) +{ + static char dir[PATH_MAX]; + unsigned int i; + + snprintf(dir, PATH_MAX, "/tmp/perf-pmu-test-format-XXXXXX"); + if (!mkdtemp(dir)) + return NULL; + + for (i = 0; i < ARRAY_SIZE(test_formats); i++) { + static char name[PATH_MAX]; + struct test_format *format = &test_formats[i]; + FILE *file; + + snprintf(name, PATH_MAX, "%s/%s", dir, format->name); + + file = fopen(name, "w"); + if (!file) + return NULL; + + if (1 != fwrite(format->value, strlen(format->value), 1, file)) + break; + + fclose(file); + } + + return dir; +} + +/* Cleanup format directory. */ +static int test_format_dir_put(char *dir) +{ + char buf[PATH_MAX]; + snprintf(buf, PATH_MAX, "rm -f %s/*\n", dir); + if (system(buf)) + return -1; + + snprintf(buf, PATH_MAX, "rmdir %s\n", dir); + return system(buf); +} + +static struct list_head *test_terms_list(void) +{ + static LIST_HEAD(terms); + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(test_terms); i++) + list_add_tail(&test_terms[i].list, &terms); + + return &terms; +} + +int test__pmu(void) +{ + char *format = test_format_dir_get(); + LIST_HEAD(formats); + struct list_head *terms = test_terms_list(); + int ret; + + if (!format) + return -EINVAL; + + do { + struct perf_event_attr attr; + + memset(&attr, 0, sizeof(attr)); + + ret = perf_pmu__format_parse(format, &formats); + if (ret) + break; + + ret = perf_pmu__config_terms(&formats, &attr, terms); + if (ret) + break; + + ret = -EINVAL; + + if (attr.config != 0xc00000000002a823) + break; + if (attr.config1 != 0x8000400000000145) + break; + if (attr.config2 != 0x0400000020041d07) + break; + + ret = 0; + } while (0); + + test_format_dir_put(format); + return ret; +} diff --git a/tools/perf/tests/python-use.c b/tools/perf/tests/python-use.c new file mode 100644 index 00000000..7760277c --- /dev/null +++ b/tools/perf/tests/python-use.c @@ -0,0 +1,23 @@ +/* + * Just test if we can load the python binding. + */ + +#include <stdio.h> +#include <stdlib.h> +#include "tests.h" + +extern int verbose; + +int test__python_use(void) +{ + char *cmd; + int ret; + + if (asprintf(&cmd, "echo \"import sys ; sys.path.append('%s'); import perf\" | %s %s", + PYTHONPATH, PYTHON, verbose ? "" : "2> /dev/null") < 0) + return -1; + + ret = system(cmd) ? -1 : 0; + free(cmd); + return ret; +} diff --git a/tools/perf/tests/rdpmc.c b/tools/perf/tests/rdpmc.c new file mode 100644 index 00000000..ff94886a --- /dev/null +++ b/tools/perf/tests/rdpmc.c @@ -0,0 +1,175 @@ +#include <unistd.h> +#include <stdlib.h> +#include <signal.h> +#include <sys/mman.h> +#include "types.h" +#include "perf.h" +#include "debug.h" +#include "tests.h" + +#if defined(__x86_64__) || defined(__i386__) + +#define barrier() asm volatile("" ::: "memory") + +static u64 rdpmc(unsigned int counter) +{ + unsigned int low, high; + + asm volatile("rdpmc" : "=a" (low), "=d" (high) : "c" (counter)); + + return low | ((u64)high) << 32; +} + +static u64 rdtsc(void) +{ + unsigned int low, high; + + asm volatile("rdtsc" : "=a" (low), "=d" (high)); + + return low | ((u64)high) << 32; +} + +static u64 mmap_read_self(void *addr) +{ + struct perf_event_mmap_page *pc = addr; + u32 seq, idx, time_mult = 0, time_shift = 0; + u64 count, cyc = 0, time_offset = 0, enabled, running, delta; + + do { + seq = pc->lock; + barrier(); + + enabled = pc->time_enabled; + running = pc->time_running; + + if (enabled != running) { + cyc = rdtsc(); + time_mult = pc->time_mult; + time_shift = pc->time_shift; + time_offset = pc->time_offset; + } + + idx = pc->index; + count = pc->offset; + if (idx) + count += rdpmc(idx - 1); + + barrier(); + } while (pc->lock != seq); + + if (enabled != running) { + u64 quot, rem; + + quot = (cyc >> time_shift); + rem = cyc & ((1 << time_shift) - 1); + delta = time_offset + quot * time_mult + + ((rem * time_mult) >> time_shift); + + enabled += delta; + if (idx) + running += delta; + + quot = count / running; + rem = count % running; + count = quot * enabled + (rem * enabled) / running; + } + + return count; +} + +/* + * If the RDPMC instruction faults then signal this back to the test parent task: + */ +static void segfault_handler(int sig __maybe_unused, + siginfo_t *info __maybe_unused, + void *uc __maybe_unused) +{ + exit(-1); +} + +static int __test__rdpmc(void) +{ + volatile int tmp = 0; + u64 i, loops = 1000; + int n; + int fd; + void *addr; + struct perf_event_attr attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_INSTRUCTIONS, + .exclude_kernel = 1, + }; + u64 delta_sum = 0; + struct sigaction sa; + + sigfillset(&sa.sa_mask); + sa.sa_sigaction = segfault_handler; + sigaction(SIGSEGV, &sa, NULL); + + fd = sys_perf_event_open(&attr, 0, -1, -1, 0); + if (fd < 0) { + pr_err("Error: sys_perf_event_open() syscall returned " + "with %d (%s)\n", fd, strerror(errno)); + return -1; + } + + addr = mmap(NULL, page_size, PROT_READ, MAP_SHARED, fd, 0); + if (addr == (void *)(-1)) { + pr_err("Error: mmap() syscall returned with (%s)\n", + strerror(errno)); + goto out_close; + } + + for (n = 0; n < 6; n++) { + u64 stamp, now, delta; + + stamp = mmap_read_self(addr); + + for (i = 0; i < loops; i++) + tmp++; + + now = mmap_read_self(addr); + loops *= 10; + + delta = now - stamp; + pr_debug("%14d: %14Lu\n", n, (long long)delta); + + delta_sum += delta; + } + + munmap(addr, page_size); + pr_debug(" "); +out_close: + close(fd); + + if (!delta_sum) + return -1; + + return 0; +} + +int test__rdpmc(void) +{ + int status = 0; + int wret = 0; + int ret; + int pid; + + pid = fork(); + if (pid < 0) + return -1; + + if (!pid) { + ret = __test__rdpmc(); + + exit(ret); + } + + wret = waitpid(pid, &status, 0); + if (wret < 0 || status) + return -1; + + return 0; +} + +#endif diff --git a/tools/perf/tests/tests.h b/tools/perf/tests/tests.h new file mode 100644 index 00000000..5de0be1f --- /dev/null +++ b/tools/perf/tests/tests.h @@ -0,0 +1,27 @@ +#ifndef TESTS_H +#define TESTS_H + +enum { + TEST_OK = 0, + TEST_FAIL = -1, + TEST_SKIP = -2, +}; + +/* Tests */ +int test__vmlinux_matches_kallsyms(void); +int test__open_syscall_event(void); +int test__open_syscall_event_on_all_cpus(void); +int test__basic_mmap(void); +int test__PERF_RECORD(void); +int test__rdpmc(void); +int test__perf_evsel__roundtrip_name_test(void); +int test__perf_evsel__tp_sched_test(void); +int test__syscall_open_tp_fields(void); +int test__pmu(void); +int test__attr(void); +int test__dso_data(void); +int test__parse_events(void); +int test__hists_link(void); +int test__python_use(void); + +#endif /* TESTS_H */ diff --git a/tools/perf/tests/vmlinux-kallsyms.c b/tools/perf/tests/vmlinux-kallsyms.c new file mode 100644 index 00000000..7b4c4d26 --- /dev/null +++ b/tools/perf/tests/vmlinux-kallsyms.c @@ -0,0 +1,233 @@ +#include <linux/compiler.h> +#include <linux/rbtree.h> +#include <string.h> +#include "map.h" +#include "symbol.h" +#include "util.h" +#include "tests.h" +#include "debug.h" +#include "machine.h" + +static int vmlinux_matches_kallsyms_filter(struct map *map __maybe_unused, + struct symbol *sym) +{ + bool *visited = symbol__priv(sym); + *visited = true; + return 0; +} + +int test__vmlinux_matches_kallsyms(void) +{ + int err = -1; + struct rb_node *nd; + struct symbol *sym; + struct map *kallsyms_map, *vmlinux_map; + struct machine kallsyms, vmlinux; + enum map_type type = MAP__FUNCTION; + struct ref_reloc_sym ref_reloc_sym = { .name = "_stext", }; + + /* + * Step 1: + * + * Init the machines that will hold kernel, modules obtained from + * both vmlinux + .ko files and from /proc/kallsyms split by modules. + */ + machine__init(&kallsyms, "", HOST_KERNEL_ID); + machine__init(&vmlinux, "", HOST_KERNEL_ID); + + /* + * Step 2: + * + * Create the kernel maps for kallsyms and the DSO where we will then + * load /proc/kallsyms. Also create the modules maps from /proc/modules + * and find the .ko files that match them in /lib/modules/`uname -r`/. + */ + if (machine__create_kernel_maps(&kallsyms) < 0) { + pr_debug("machine__create_kernel_maps "); + goto out; + } + + /* + * Step 3: + * + * Load and split /proc/kallsyms into multiple maps, one per module. + */ + if (machine__load_kallsyms(&kallsyms, "/proc/kallsyms", type, NULL) <= 0) { + pr_debug("dso__load_kallsyms "); + goto out; + } + + /* + * Step 4: + * + * kallsyms will be internally on demand sorted by name so that we can + * find the reference relocation * symbol, i.e. the symbol we will use + * to see if the running kernel was relocated by checking if it has the + * same value in the vmlinux file we load. + */ + kallsyms_map = machine__kernel_map(&kallsyms, type); + + sym = map__find_symbol_by_name(kallsyms_map, ref_reloc_sym.name, NULL); + if (sym == NULL) { + pr_debug("dso__find_symbol_by_name "); + goto out; + } + + ref_reloc_sym.addr = sym->start; + + /* + * Step 5: + * + * Now repeat step 2, this time for the vmlinux file we'll auto-locate. + */ + if (machine__create_kernel_maps(&vmlinux) < 0) { + pr_debug("machine__create_kernel_maps "); + goto out; + } + + vmlinux_map = machine__kernel_map(&vmlinux, type); + map__kmap(vmlinux_map)->ref_reloc_sym = &ref_reloc_sym; + + /* + * Step 6: + * + * Locate a vmlinux file in the vmlinux path that has a buildid that + * matches the one of the running kernel. + * + * While doing that look if we find the ref reloc symbol, if we find it + * we'll have its ref_reloc_symbol.unrelocated_addr and then + * maps__reloc_vmlinux will notice and set proper ->[un]map_ip routines + * to fixup the symbols. + */ + if (machine__load_vmlinux_path(&vmlinux, type, + vmlinux_matches_kallsyms_filter) <= 0) { + pr_debug("Couldn't find a vmlinux that matches the kernel running on this machine, skipping test\n"); + err = TEST_SKIP; + goto out; + } + + err = 0; + /* + * Step 7: + * + * Now look at the symbols in the vmlinux DSO and check if we find all of them + * in the kallsyms dso. For the ones that are in both, check its names and + * end addresses too. + */ + for (nd = rb_first(&vmlinux_map->dso->symbols[type]); nd; nd = rb_next(nd)) { + struct symbol *pair, *first_pair; + bool backwards = true; + + sym = rb_entry(nd, struct symbol, rb_node); + + if (sym->start == sym->end) + continue; + + first_pair = machine__find_kernel_symbol(&kallsyms, type, sym->start, NULL, NULL); + pair = first_pair; + + if (pair && pair->start == sym->start) { +next_pair: + if (strcmp(sym->name, pair->name) == 0) { + /* + * kallsyms don't have the symbol end, so we + * set that by using the next symbol start - 1, + * in some cases we get this up to a page + * wrong, trace_kmalloc when I was developing + * this code was one such example, 2106 bytes + * off the real size. More than that and we + * _really_ have a problem. + */ + s64 skew = sym->end - pair->end; + if (llabs(skew) < page_size) + continue; + + pr_debug("%#" PRIx64 ": diff end addr for %s v: %#" PRIx64 " k: %#" PRIx64 "\n", + sym->start, sym->name, sym->end, pair->end); + } else { + struct rb_node *nnd; +detour: + nnd = backwards ? rb_prev(&pair->rb_node) : + rb_next(&pair->rb_node); + if (nnd) { + struct symbol *next = rb_entry(nnd, struct symbol, rb_node); + + if (next->start == sym->start) { + pair = next; + goto next_pair; + } + } + + if (backwards) { + backwards = false; + pair = first_pair; + goto detour; + } + + pr_debug("%#" PRIx64 ": diff name v: %s k: %s\n", + sym->start, sym->name, pair->name); + } + } else + pr_debug("%#" PRIx64 ": %s not on kallsyms\n", sym->start, sym->name); + + err = -1; + } + + if (!verbose) + goto out; + + pr_info("Maps only in vmlinux:\n"); + + for (nd = rb_first(&vmlinux.kmaps.maps[type]); nd; nd = rb_next(nd)) { + struct map *pos = rb_entry(nd, struct map, rb_node), *pair; + /* + * If it is the kernel, kallsyms is always "[kernel.kallsyms]", while + * the kernel will have the path for the vmlinux file being used, + * so use the short name, less descriptive but the same ("[kernel]" in + * both cases. + */ + pair = map_groups__find_by_name(&kallsyms.kmaps, type, + (pos->dso->kernel ? + pos->dso->short_name : + pos->dso->name)); + if (pair) + pair->priv = 1; + else + map__fprintf(pos, stderr); + } + + pr_info("Maps in vmlinux with a different name in kallsyms:\n"); + + for (nd = rb_first(&vmlinux.kmaps.maps[type]); nd; nd = rb_next(nd)) { + struct map *pos = rb_entry(nd, struct map, rb_node), *pair; + + pair = map_groups__find(&kallsyms.kmaps, type, pos->start); + if (pair == NULL || pair->priv) + continue; + + if (pair->start == pos->start) { + pair->priv = 1; + pr_info(" %" PRIx64 "-%" PRIx64 " %" PRIx64 " %s in kallsyms as", + pos->start, pos->end, pos->pgoff, pos->dso->name); + if (pos->pgoff != pair->pgoff || pos->end != pair->end) + pr_info(": \n*%" PRIx64 "-%" PRIx64 " %" PRIx64 "", + pair->start, pair->end, pair->pgoff); + pr_info(" %s\n", pair->dso->name); + pair->priv = 1; + } + } + + pr_info("Maps only in kallsyms:\n"); + + for (nd = rb_first(&kallsyms.kmaps.maps[type]); + nd; nd = rb_next(nd)) { + struct map *pos = rb_entry(nd, struct map, rb_node); + + if (!pos->priv) + map__fprintf(pos, stderr); + } +out: + machine__exit(&kallsyms); + machine__exit(&vmlinux); + return err; +} diff --git a/tools/perf/ui/browser.c b/tools/perf/ui/browser.c new file mode 100644 index 00000000..809ea463 --- /dev/null +++ b/tools/perf/ui/browser.c @@ -0,0 +1,716 @@ +#include "../util.h" +#include "../cache.h" +#include "../../perf.h" +#include "libslang.h" +#include <newt.h> +#include "ui.h" +#include "util.h" +#include <linux/compiler.h> +#include <linux/list.h> +#include <linux/rbtree.h> +#include <stdlib.h> +#include <sys/ttydefaults.h> +#include "browser.h" +#include "helpline.h" +#include "keysyms.h" +#include "../color.h" + +static int ui_browser__percent_color(struct ui_browser *browser, + double percent, bool current) +{ + if (current && (!browser->use_navkeypressed || browser->navkeypressed)) + return HE_COLORSET_SELECTED; + if (percent >= MIN_RED) + return HE_COLORSET_TOP; + if (percent >= MIN_GREEN) + return HE_COLORSET_MEDIUM; + return HE_COLORSET_NORMAL; +} + +int ui_browser__set_color(struct ui_browser *browser, int color) +{ + int ret = browser->current_color; + browser->current_color = color; + SLsmg_set_color(color); + return ret; +} + +void ui_browser__set_percent_color(struct ui_browser *browser, + double percent, bool current) +{ + int color = ui_browser__percent_color(browser, percent, current); + ui_browser__set_color(browser, color); +} + +void ui_browser__gotorc(struct ui_browser *browser, int y, int x) +{ + SLsmg_gotorc(browser->y + y, browser->x + x); +} + +static struct list_head * +ui_browser__list_head_filter_entries(struct ui_browser *browser, + struct list_head *pos) +{ + do { + if (!browser->filter || !browser->filter(browser, pos)) + return pos; + pos = pos->next; + } while (pos != browser->entries); + + return NULL; +} + +static struct list_head * +ui_browser__list_head_filter_prev_entries(struct ui_browser *browser, + struct list_head *pos) +{ + do { + if (!browser->filter || !browser->filter(browser, pos)) + return pos; + pos = pos->prev; + } while (pos != browser->entries); + + return NULL; +} + +void ui_browser__list_head_seek(struct ui_browser *browser, off_t offset, int whence) +{ + struct list_head *head = browser->entries; + struct list_head *pos; + + if (browser->nr_entries == 0) + return; + + switch (whence) { + case SEEK_SET: + pos = ui_browser__list_head_filter_entries(browser, head->next); + break; + case SEEK_CUR: + pos = browser->top; + break; + case SEEK_END: + pos = ui_browser__list_head_filter_prev_entries(browser, head->prev); + break; + default: + return; + } + + assert(pos != NULL); + + if (offset > 0) { + while (offset-- != 0) + pos = ui_browser__list_head_filter_entries(browser, pos->next); + } else { + while (offset++ != 0) + pos = ui_browser__list_head_filter_prev_entries(browser, pos->prev); + } + + browser->top = pos; +} + +void ui_browser__rb_tree_seek(struct ui_browser *browser, off_t offset, int whence) +{ + struct rb_root *root = browser->entries; + struct rb_node *nd; + + switch (whence) { + case SEEK_SET: + nd = rb_first(root); + break; + case SEEK_CUR: + nd = browser->top; + break; + case SEEK_END: + nd = rb_last(root); + break; + default: + return; + } + + if (offset > 0) { + while (offset-- != 0) + nd = rb_next(nd); + } else { + while (offset++ != 0) + nd = rb_prev(nd); + } + + browser->top = nd; +} + +unsigned int ui_browser__rb_tree_refresh(struct ui_browser *browser) +{ + struct rb_node *nd; + int row = 0; + + if (browser->top == NULL) + browser->top = rb_first(browser->entries); + + nd = browser->top; + + while (nd != NULL) { + ui_browser__gotorc(browser, row, 0); + browser->write(browser, nd, row); + if (++row == browser->height) + break; + nd = rb_next(nd); + } + + return row; +} + +bool ui_browser__is_current_entry(struct ui_browser *browser, unsigned row) +{ + return browser->top_idx + row == browser->index; +} + +void ui_browser__refresh_dimensions(struct ui_browser *browser) +{ + browser->width = SLtt_Screen_Cols - 1; + browser->height = SLtt_Screen_Rows - 2; + browser->y = 1; + browser->x = 0; +} + +void ui_browser__handle_resize(struct ui_browser *browser) +{ + ui__refresh_dimensions(false); + ui_browser__show(browser, browser->title, ui_helpline__current); + ui_browser__refresh(browser); +} + +int ui_browser__warning(struct ui_browser *browser, int timeout, + const char *format, ...) +{ + va_list args; + char *text; + int key = 0, err; + + va_start(args, format); + err = vasprintf(&text, format, args); + va_end(args); + + if (err < 0) { + va_start(args, format); + ui_helpline__vpush(format, args); + va_end(args); + } else { + while ((key == ui__question_window("Warning!", text, + "Press any key...", + timeout)) == K_RESIZE) + ui_browser__handle_resize(browser); + free(text); + } + + return key; +} + +int ui_browser__help_window(struct ui_browser *browser, const char *text) +{ + int key; + + while ((key = ui__help_window(text)) == K_RESIZE) + ui_browser__handle_resize(browser); + + return key; +} + +bool ui_browser__dialog_yesno(struct ui_browser *browser, const char *text) +{ + int key; + + while ((key = ui__dialog_yesno(text)) == K_RESIZE) + ui_browser__handle_resize(browser); + + return key == K_ENTER || toupper(key) == 'Y'; +} + +void ui_browser__reset_index(struct ui_browser *browser) +{ + browser->index = browser->top_idx = 0; + browser->seek(browser, 0, SEEK_SET); +} + +void __ui_browser__show_title(struct ui_browser *browser, const char *title) +{ + SLsmg_gotorc(0, 0); + ui_browser__set_color(browser, NEWT_COLORSET_ROOT); + slsmg_write_nstring(title, browser->width + 1); +} + +void ui_browser__show_title(struct ui_browser *browser, const char *title) +{ + pthread_mutex_lock(&ui__lock); + __ui_browser__show_title(browser, title); + pthread_mutex_unlock(&ui__lock); +} + +int ui_browser__show(struct ui_browser *browser, const char *title, + const char *helpline, ...) +{ + int err; + va_list ap; + + ui_browser__refresh_dimensions(browser); + + pthread_mutex_lock(&ui__lock); + __ui_browser__show_title(browser, title); + + browser->title = title; + free(browser->helpline); + browser->helpline = NULL; + + va_start(ap, helpline); + err = vasprintf(&browser->helpline, helpline, ap); + va_end(ap); + if (err > 0) + ui_helpline__push(browser->helpline); + pthread_mutex_unlock(&ui__lock); + return err ? 0 : -1; +} + +void ui_browser__hide(struct ui_browser *browser __maybe_unused) +{ + pthread_mutex_lock(&ui__lock); + ui_helpline__pop(); + free(browser->helpline); + browser->helpline = NULL; + pthread_mutex_unlock(&ui__lock); +} + +static void ui_browser__scrollbar_set(struct ui_browser *browser) +{ + int height = browser->height, h = 0, pct = 0, + col = browser->width, + row = browser->y - 1; + + if (browser->nr_entries > 1) { + pct = ((browser->index * (browser->height - 1)) / + (browser->nr_entries - 1)); + } + + SLsmg_set_char_set(1); + + while (h < height) { + ui_browser__gotorc(browser, row++, col); + SLsmg_write_char(h == pct ? SLSMG_DIAMOND_CHAR : SLSMG_CKBRD_CHAR); + ++h; + } + + SLsmg_set_char_set(0); +} + +static int __ui_browser__refresh(struct ui_browser *browser) +{ + int row; + int width = browser->width; + + row = browser->refresh(browser); + ui_browser__set_color(browser, HE_COLORSET_NORMAL); + + if (!browser->use_navkeypressed || browser->navkeypressed) + ui_browser__scrollbar_set(browser); + else + width += 1; + + SLsmg_fill_region(browser->y + row, browser->x, + browser->height - row, width, ' '); + + return 0; +} + +int ui_browser__refresh(struct ui_browser *browser) +{ + pthread_mutex_lock(&ui__lock); + __ui_browser__refresh(browser); + pthread_mutex_unlock(&ui__lock); + + return 0; +} + +/* + * Here we're updating nr_entries _after_ we started browsing, i.e. we have to + * forget about any reference to any entry in the underlying data structure, + * that is why we do a SEEK_SET. Think about 'perf top' in the hists browser + * after an output_resort and hist decay. + */ +void ui_browser__update_nr_entries(struct ui_browser *browser, u32 nr_entries) +{ + off_t offset = nr_entries - browser->nr_entries; + + browser->nr_entries = nr_entries; + + if (offset < 0) { + if (browser->top_idx < (u64)-offset) + offset = -browser->top_idx; + + browser->index += offset; + browser->top_idx += offset; + } + + browser->top = NULL; + browser->seek(browser, browser->top_idx, SEEK_SET); +} + +int ui_browser__run(struct ui_browser *browser, int delay_secs) +{ + int err, key; + + while (1) { + off_t offset; + + pthread_mutex_lock(&ui__lock); + err = __ui_browser__refresh(browser); + SLsmg_refresh(); + pthread_mutex_unlock(&ui__lock); + if (err < 0) + break; + + key = ui__getch(delay_secs); + + if (key == K_RESIZE) { + ui__refresh_dimensions(false); + ui_browser__refresh_dimensions(browser); + __ui_browser__show_title(browser, browser->title); + ui_helpline__puts(browser->helpline); + continue; + } + + if (browser->use_navkeypressed && !browser->navkeypressed) { + if (key == K_DOWN || key == K_UP || + key == K_PGDN || key == K_PGUP || + key == K_HOME || key == K_END || + key == ' ') { + browser->navkeypressed = true; + continue; + } else + return key; + } + + switch (key) { + case K_DOWN: + if (browser->index == browser->nr_entries - 1) + break; + ++browser->index; + if (browser->index == browser->top_idx + browser->height) { + ++browser->top_idx; + browser->seek(browser, +1, SEEK_CUR); + } + break; + case K_UP: + if (browser->index == 0) + break; + --browser->index; + if (browser->index < browser->top_idx) { + --browser->top_idx; + browser->seek(browser, -1, SEEK_CUR); + } + break; + case K_PGDN: + case ' ': + if (browser->top_idx + browser->height > browser->nr_entries - 1) + break; + + offset = browser->height; + if (browser->index + offset > browser->nr_entries - 1) + offset = browser->nr_entries - 1 - browser->index; + browser->index += offset; + browser->top_idx += offset; + browser->seek(browser, +offset, SEEK_CUR); + break; + case K_PGUP: + if (browser->top_idx == 0) + break; + + if (browser->top_idx < browser->height) + offset = browser->top_idx; + else + offset = browser->height; + + browser->index -= offset; + browser->top_idx -= offset; + browser->seek(browser, -offset, SEEK_CUR); + break; + case K_HOME: + ui_browser__reset_index(browser); + break; + case K_END: + offset = browser->height - 1; + if (offset >= browser->nr_entries) + offset = browser->nr_entries - 1; + + browser->index = browser->nr_entries - 1; + browser->top_idx = browser->index - offset; + browser->seek(browser, -offset, SEEK_END); + break; + default: + return key; + } + } + return -1; +} + +unsigned int ui_browser__list_head_refresh(struct ui_browser *browser) +{ + struct list_head *pos; + struct list_head *head = browser->entries; + int row = 0; + + if (browser->top == NULL || browser->top == browser->entries) + browser->top = ui_browser__list_head_filter_entries(browser, head->next); + + pos = browser->top; + + list_for_each_from(pos, head) { + if (!browser->filter || !browser->filter(browser, pos)) { + ui_browser__gotorc(browser, row, 0); + browser->write(browser, pos, row); + if (++row == browser->height) + break; + } + } + + return row; +} + +static struct ui_browser_colorset { + const char *name, *fg, *bg; + int colorset; +} ui_browser__colorsets[] = { + { + .colorset = HE_COLORSET_TOP, + .name = "top", + .fg = "red", + .bg = "default", + }, + { + .colorset = HE_COLORSET_MEDIUM, + .name = "medium", + .fg = "green", + .bg = "default", + }, + { + .colorset = HE_COLORSET_NORMAL, + .name = "normal", + .fg = "default", + .bg = "default", + }, + { + .colorset = HE_COLORSET_SELECTED, + .name = "selected", + .fg = "black", + .bg = "lightgray", + }, + { + .colorset = HE_COLORSET_CODE, + .name = "code", + .fg = "blue", + .bg = "default", + }, + { + .colorset = HE_COLORSET_ADDR, + .name = "addr", + .fg = "magenta", + .bg = "default", + }, + { + .name = NULL, + } +}; + + +static int ui_browser__color_config(const char *var, const char *value, + void *data __maybe_unused) +{ + char *fg = NULL, *bg; + int i; + + /* same dir for all commands */ + if (prefixcmp(var, "colors.") != 0) + return 0; + + for (i = 0; ui_browser__colorsets[i].name != NULL; ++i) { + const char *name = var + 7; + + if (strcmp(ui_browser__colorsets[i].name, name) != 0) + continue; + + fg = strdup(value); + if (fg == NULL) + break; + + bg = strchr(fg, ','); + if (bg == NULL) + break; + + *bg = '\0'; + while (isspace(*++bg)); + ui_browser__colorsets[i].bg = bg; + ui_browser__colorsets[i].fg = fg; + return 0; + } + + free(fg); + return -1; +} + +void ui_browser__argv_seek(struct ui_browser *browser, off_t offset, int whence) +{ + switch (whence) { + case SEEK_SET: + browser->top = browser->entries; + break; + case SEEK_CUR: + browser->top = browser->top + browser->top_idx + offset; + break; + case SEEK_END: + browser->top = browser->top + browser->nr_entries + offset; + break; + default: + return; + } +} + +unsigned int ui_browser__argv_refresh(struct ui_browser *browser) +{ + unsigned int row = 0, idx = browser->top_idx; + char **pos; + + if (browser->top == NULL) + browser->top = browser->entries; + + pos = (char **)browser->top; + while (idx < browser->nr_entries) { + if (!browser->filter || !browser->filter(browser, *pos)) { + ui_browser__gotorc(browser, row, 0); + browser->write(browser, pos, row); + if (++row == browser->height) + break; + } + + ++idx; + ++pos; + } + + return row; +} + +void __ui_browser__vline(struct ui_browser *browser, unsigned int column, + u16 start, u16 end) +{ + SLsmg_set_char_set(1); + ui_browser__gotorc(browser, start, column); + SLsmg_draw_vline(end - start + 1); + SLsmg_set_char_set(0); +} + +void ui_browser__write_graph(struct ui_browser *browser __maybe_unused, + int graph) +{ + SLsmg_set_char_set(1); + SLsmg_write_char(graph); + SLsmg_set_char_set(0); +} + +static void __ui_browser__line_arrow_up(struct ui_browser *browser, + unsigned int column, + u64 start, u64 end) +{ + unsigned int row, end_row; + + SLsmg_set_char_set(1); + + if (start < browser->top_idx + browser->height) { + row = start - browser->top_idx; + ui_browser__gotorc(browser, row, column); + SLsmg_write_char(SLSMG_LLCORN_CHAR); + ui_browser__gotorc(browser, row, column + 1); + SLsmg_draw_hline(2); + + if (row-- == 0) + goto out; + } else + row = browser->height - 1; + + if (end > browser->top_idx) + end_row = end - browser->top_idx; + else + end_row = 0; + + ui_browser__gotorc(browser, end_row, column); + SLsmg_draw_vline(row - end_row + 1); + + ui_browser__gotorc(browser, end_row, column); + if (end >= browser->top_idx) { + SLsmg_write_char(SLSMG_ULCORN_CHAR); + ui_browser__gotorc(browser, end_row, column + 1); + SLsmg_write_char(SLSMG_HLINE_CHAR); + ui_browser__gotorc(browser, end_row, column + 2); + SLsmg_write_char(SLSMG_RARROW_CHAR); + } +out: + SLsmg_set_char_set(0); +} + +static void __ui_browser__line_arrow_down(struct ui_browser *browser, + unsigned int column, + u64 start, u64 end) +{ + unsigned int row, end_row; + + SLsmg_set_char_set(1); + + if (start >= browser->top_idx) { + row = start - browser->top_idx; + ui_browser__gotorc(browser, row, column); + SLsmg_write_char(SLSMG_ULCORN_CHAR); + ui_browser__gotorc(browser, row, column + 1); + SLsmg_draw_hline(2); + + if (row++ == 0) + goto out; + } else + row = 0; + + if (end >= browser->top_idx + browser->height) + end_row = browser->height - 1; + else + end_row = end - browser->top_idx;; + + ui_browser__gotorc(browser, row, column); + SLsmg_draw_vline(end_row - row + 1); + + ui_browser__gotorc(browser, end_row, column); + if (end < browser->top_idx + browser->height) { + SLsmg_write_char(SLSMG_LLCORN_CHAR); + ui_browser__gotorc(browser, end_row, column + 1); + SLsmg_write_char(SLSMG_HLINE_CHAR); + ui_browser__gotorc(browser, end_row, column + 2); + SLsmg_write_char(SLSMG_RARROW_CHAR); + } +out: + SLsmg_set_char_set(0); +} + +void __ui_browser__line_arrow(struct ui_browser *browser, unsigned int column, + u64 start, u64 end) +{ + if (start > end) + __ui_browser__line_arrow_up(browser, column, start, end); + else + __ui_browser__line_arrow_down(browser, column, start, end); +} + +void ui_browser__init(void) +{ + int i = 0; + + perf_config(ui_browser__color_config, NULL); + + while (ui_browser__colorsets[i].name) { + struct ui_browser_colorset *c = &ui_browser__colorsets[i++]; + sltt_set_color(c->colorset, c->name, c->fg, c->bg); + } + + annotate_browser__init(); +} diff --git a/tools/perf/ui/browser.h b/tools/perf/ui/browser.h new file mode 100644 index 00000000..af703146 --- /dev/null +++ b/tools/perf/ui/browser.h @@ -0,0 +1,73 @@ +#ifndef _PERF_UI_BROWSER_H_ +#define _PERF_UI_BROWSER_H_ 1 + +#include <stdbool.h> +#include <sys/types.h> +#include "../types.h" + +#define HE_COLORSET_TOP 50 +#define HE_COLORSET_MEDIUM 51 +#define HE_COLORSET_NORMAL 52 +#define HE_COLORSET_SELECTED 53 +#define HE_COLORSET_CODE 54 +#define HE_COLORSET_ADDR 55 + +struct ui_browser { + u64 index, top_idx; + void *top, *entries; + u16 y, x, width, height; + int current_color; + void *priv; + const char *title; + char *helpline; + unsigned int (*refresh)(struct ui_browser *self); + void (*write)(struct ui_browser *self, void *entry, int row); + void (*seek)(struct ui_browser *self, off_t offset, int whence); + bool (*filter)(struct ui_browser *self, void *entry); + u32 nr_entries; + bool navkeypressed; + bool use_navkeypressed; +}; + +int ui_browser__set_color(struct ui_browser *browser, int color); +void ui_browser__set_percent_color(struct ui_browser *self, + double percent, bool current); +bool ui_browser__is_current_entry(struct ui_browser *self, unsigned row); +void ui_browser__refresh_dimensions(struct ui_browser *self); +void ui_browser__reset_index(struct ui_browser *self); + +void ui_browser__gotorc(struct ui_browser *self, int y, int x); +void ui_browser__write_graph(struct ui_browser *browser, int graph); +void __ui_browser__line_arrow(struct ui_browser *browser, unsigned int column, + u64 start, u64 end); +void __ui_browser__show_title(struct ui_browser *browser, const char *title); +void ui_browser__show_title(struct ui_browser *browser, const char *title); +int ui_browser__show(struct ui_browser *self, const char *title, + const char *helpline, ...); +void ui_browser__hide(struct ui_browser *self); +int ui_browser__refresh(struct ui_browser *self); +int ui_browser__run(struct ui_browser *browser, int delay_secs); +void ui_browser__update_nr_entries(struct ui_browser *browser, u32 nr_entries); +void ui_browser__handle_resize(struct ui_browser *browser); +void __ui_browser__vline(struct ui_browser *browser, unsigned int column, + u16 start, u16 end); + +int ui_browser__warning(struct ui_browser *browser, int timeout, + const char *format, ...); +int ui_browser__help_window(struct ui_browser *browser, const char *text); +bool ui_browser__dialog_yesno(struct ui_browser *browser, const char *text); +int ui_browser__input_window(const char *title, const char *text, char *input, + const char *exit_msg, int delay_sec); + +void ui_browser__argv_seek(struct ui_browser *browser, off_t offset, int whence); +unsigned int ui_browser__argv_refresh(struct ui_browser *browser); + +void ui_browser__rb_tree_seek(struct ui_browser *self, off_t offset, int whence); +unsigned int ui_browser__rb_tree_refresh(struct ui_browser *self); + +void ui_browser__list_head_seek(struct ui_browser *self, off_t offset, int whence); +unsigned int ui_browser__list_head_refresh(struct ui_browser *self); + +void ui_browser__init(void); +void annotate_browser__init(void); +#endif /* _PERF_UI_BROWSER_H_ */ diff --git a/tools/perf/ui/browsers/annotate.c b/tools/perf/ui/browsers/annotate.c new file mode 100644 index 00000000..7dca1555 --- /dev/null +++ b/tools/perf/ui/browsers/annotate.c @@ -0,0 +1,971 @@ +#include "../../util/util.h" +#include "../browser.h" +#include "../helpline.h" +#include "../libslang.h" +#include "../ui.h" +#include "../util.h" +#include "../../util/annotate.h" +#include "../../util/hist.h" +#include "../../util/sort.h" +#include "../../util/symbol.h" +#include <pthread.h> +#include <newt.h> + +struct browser_disasm_line { + struct rb_node rb_node; + double percent; + u32 idx; + int idx_asm; + int jump_sources; +}; + +static struct annotate_browser_opt { + bool hide_src_code, + use_offset, + jump_arrows, + show_nr_jumps; +} annotate_browser__opts = { + .use_offset = true, + .jump_arrows = true, +}; + +struct annotate_browser { + struct ui_browser b; + struct rb_root entries; + struct rb_node *curr_hot; + struct disasm_line *selection; + struct disasm_line **offsets; + u64 start; + int nr_asm_entries; + int nr_entries; + int max_jump_sources; + int nr_jumps; + bool searching_backwards; + u8 addr_width; + u8 jumps_width; + u8 target_width; + u8 min_addr_width; + u8 max_addr_width; + char search_bf[128]; +}; + +static inline struct browser_disasm_line *disasm_line__browser(struct disasm_line *dl) +{ + return (struct browser_disasm_line *)(dl + 1); +} + +static bool disasm_line__filter(struct ui_browser *browser __maybe_unused, + void *entry) +{ + if (annotate_browser__opts.hide_src_code) { + struct disasm_line *dl = list_entry(entry, struct disasm_line, node); + return dl->offset == -1; + } + + return false; +} + +static int annotate_browser__jumps_percent_color(struct annotate_browser *browser, + int nr, bool current) +{ + if (current && (!browser->b.use_navkeypressed || browser->b.navkeypressed)) + return HE_COLORSET_SELECTED; + if (nr == browser->max_jump_sources) + return HE_COLORSET_TOP; + if (nr > 1) + return HE_COLORSET_MEDIUM; + return HE_COLORSET_NORMAL; +} + +static int annotate_browser__set_jumps_percent_color(struct annotate_browser *browser, + int nr, bool current) +{ + int color = annotate_browser__jumps_percent_color(browser, nr, current); + return ui_browser__set_color(&browser->b, color); +} + +static void annotate_browser__write(struct ui_browser *browser, void *entry, int row) +{ + struct annotate_browser *ab = container_of(browser, struct annotate_browser, b); + struct disasm_line *dl = list_entry(entry, struct disasm_line, node); + struct browser_disasm_line *bdl = disasm_line__browser(dl); + bool current_entry = ui_browser__is_current_entry(browser, row); + bool change_color = (!annotate_browser__opts.hide_src_code && + (!current_entry || (browser->use_navkeypressed && + !browser->navkeypressed))); + int width = browser->width, printed; + char bf[256]; + + if (dl->offset != -1 && bdl->percent != 0.0) { + ui_browser__set_percent_color(browser, bdl->percent, current_entry); + slsmg_printf("%6.2f ", bdl->percent); + } else { + ui_browser__set_percent_color(browser, 0, current_entry); + slsmg_write_nstring(" ", 7); + } + + SLsmg_write_char(' '); + + /* The scroll bar isn't being used */ + if (!browser->navkeypressed) + width += 1; + + if (!*dl->line) + slsmg_write_nstring(" ", width - 7); + else if (dl->offset == -1) { + printed = scnprintf(bf, sizeof(bf), "%*s ", + ab->addr_width, " "); + slsmg_write_nstring(bf, printed); + slsmg_write_nstring(dl->line, width - printed - 6); + } else { + u64 addr = dl->offset; + int color = -1; + + if (!annotate_browser__opts.use_offset) + addr += ab->start; + + if (!annotate_browser__opts.use_offset) { + printed = scnprintf(bf, sizeof(bf), "%" PRIx64 ": ", addr); + } else { + if (bdl->jump_sources) { + if (annotate_browser__opts.show_nr_jumps) { + int prev; + printed = scnprintf(bf, sizeof(bf), "%*d ", + ab->jumps_width, + bdl->jump_sources); + prev = annotate_browser__set_jumps_percent_color(ab, bdl->jump_sources, + current_entry); + slsmg_write_nstring(bf, printed); + ui_browser__set_color(browser, prev); + } + + printed = scnprintf(bf, sizeof(bf), "%*" PRIx64 ": ", + ab->target_width, addr); + } else { + printed = scnprintf(bf, sizeof(bf), "%*s ", + ab->addr_width, " "); + } + } + + if (change_color) + color = ui_browser__set_color(browser, HE_COLORSET_ADDR); + slsmg_write_nstring(bf, printed); + if (change_color) + ui_browser__set_color(browser, color); + if (dl->ins && dl->ins->ops->scnprintf) { + if (ins__is_jump(dl->ins)) { + bool fwd = dl->ops.target.offset > (u64)dl->offset; + + ui_browser__write_graph(browser, fwd ? SLSMG_DARROW_CHAR : + SLSMG_UARROW_CHAR); + SLsmg_write_char(' '); + } else if (ins__is_call(dl->ins)) { + ui_browser__write_graph(browser, SLSMG_RARROW_CHAR); + SLsmg_write_char(' '); + } else { + slsmg_write_nstring(" ", 2); + } + } else { + if (strcmp(dl->name, "retq")) { + slsmg_write_nstring(" ", 2); + } else { + ui_browser__write_graph(browser, SLSMG_LARROW_CHAR); + SLsmg_write_char(' '); + } + } + + disasm_line__scnprintf(dl, bf, sizeof(bf), !annotate_browser__opts.use_offset); + slsmg_write_nstring(bf, width - 10 - printed); + } + + if (current_entry) + ab->selection = dl; +} + +static bool disasm_line__is_valid_jump(struct disasm_line *dl, struct symbol *sym) +{ + if (!dl || !dl->ins || !ins__is_jump(dl->ins) + || !disasm_line__has_offset(dl) + || dl->ops.target.offset >= symbol__size(sym)) + return false; + + return true; +} + +static void annotate_browser__draw_current_jump(struct ui_browser *browser) +{ + struct annotate_browser *ab = container_of(browser, struct annotate_browser, b); + struct disasm_line *cursor = ab->selection, *target; + struct browser_disasm_line *btarget, *bcursor; + unsigned int from, to; + struct map_symbol *ms = ab->b.priv; + struct symbol *sym = ms->sym; + + /* PLT symbols contain external offsets */ + if (strstr(sym->name, "@plt")) + return; + + if (!disasm_line__is_valid_jump(cursor, sym)) + return; + + target = ab->offsets[cursor->ops.target.offset]; + if (!target) + return; + + bcursor = disasm_line__browser(cursor); + btarget = disasm_line__browser(target); + + if (annotate_browser__opts.hide_src_code) { + from = bcursor->idx_asm; + to = btarget->idx_asm; + } else { + from = (u64)bcursor->idx; + to = (u64)btarget->idx; + } + + ui_browser__set_color(browser, HE_COLORSET_CODE); + __ui_browser__line_arrow(browser, 9 + ab->addr_width, from, to); +} + +static unsigned int annotate_browser__refresh(struct ui_browser *browser) +{ + int ret = ui_browser__list_head_refresh(browser); + + if (annotate_browser__opts.jump_arrows) + annotate_browser__draw_current_jump(browser); + + ui_browser__set_color(browser, HE_COLORSET_NORMAL); + __ui_browser__vline(browser, 7, 0, browser->height - 1); + return ret; +} + +static double disasm_line__calc_percent(struct disasm_line *dl, struct symbol *sym, int evidx) +{ + double percent = 0.0; + + if (dl->offset != -1) { + int len = sym->end - sym->start; + unsigned int hits = 0; + struct annotation *notes = symbol__annotation(sym); + struct source_line *src_line = notes->src->lines; + struct sym_hist *h = annotation__histogram(notes, evidx); + s64 offset = dl->offset; + struct disasm_line *next; + + next = disasm__get_next_ip_line(¬es->src->source, dl); + while (offset < (s64)len && + (next == NULL || offset < next->offset)) { + if (src_line) { + percent += src_line[offset].percent; + } else + hits += h->addr[offset]; + + ++offset; + } + /* + * If the percentage wasn't already calculated in + * symbol__get_source_line, do it now: + */ + if (src_line == NULL && h->sum) + percent = 100.0 * hits / h->sum; + } + + return percent; +} + +static void disasm_rb_tree__insert(struct rb_root *root, struct browser_disasm_line *bdl) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent = NULL; + struct browser_disasm_line *l; + + while (*p != NULL) { + parent = *p; + l = rb_entry(parent, struct browser_disasm_line, rb_node); + if (bdl->percent < l->percent) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + rb_link_node(&bdl->rb_node, parent, p); + rb_insert_color(&bdl->rb_node, root); +} + +static void annotate_browser__set_top(struct annotate_browser *browser, + struct disasm_line *pos, u32 idx) +{ + unsigned back; + + ui_browser__refresh_dimensions(&browser->b); + back = browser->b.height / 2; + browser->b.top_idx = browser->b.index = idx; + + while (browser->b.top_idx != 0 && back != 0) { + pos = list_entry(pos->node.prev, struct disasm_line, node); + + if (disasm_line__filter(&browser->b, &pos->node)) + continue; + + --browser->b.top_idx; + --back; + } + + browser->b.top = pos; + browser->b.navkeypressed = true; +} + +static void annotate_browser__set_rb_top(struct annotate_browser *browser, + struct rb_node *nd) +{ + struct browser_disasm_line *bpos; + struct disasm_line *pos; + u32 idx; + + bpos = rb_entry(nd, struct browser_disasm_line, rb_node); + pos = ((struct disasm_line *)bpos) - 1; + idx = bpos->idx; + if (annotate_browser__opts.hide_src_code) + idx = bpos->idx_asm; + annotate_browser__set_top(browser, pos, idx); + browser->curr_hot = nd; +} + +static void annotate_browser__calc_percent(struct annotate_browser *browser, + int evidx) +{ + struct map_symbol *ms = browser->b.priv; + struct symbol *sym = ms->sym; + struct annotation *notes = symbol__annotation(sym); + struct disasm_line *pos; + + browser->entries = RB_ROOT; + + pthread_mutex_lock(¬es->lock); + + list_for_each_entry(pos, ¬es->src->source, node) { + struct browser_disasm_line *bpos = disasm_line__browser(pos); + bpos->percent = disasm_line__calc_percent(pos, sym, evidx); + if (bpos->percent < 0.01) { + RB_CLEAR_NODE(&bpos->rb_node); + continue; + } + disasm_rb_tree__insert(&browser->entries, bpos); + } + pthread_mutex_unlock(¬es->lock); + + browser->curr_hot = rb_last(&browser->entries); +} + +static bool annotate_browser__toggle_source(struct annotate_browser *browser) +{ + struct disasm_line *dl; + struct browser_disasm_line *bdl; + off_t offset = browser->b.index - browser->b.top_idx; + + browser->b.seek(&browser->b, offset, SEEK_CUR); + dl = list_entry(browser->b.top, struct disasm_line, node); + bdl = disasm_line__browser(dl); + + if (annotate_browser__opts.hide_src_code) { + if (bdl->idx_asm < offset) + offset = bdl->idx; + + browser->b.nr_entries = browser->nr_entries; + annotate_browser__opts.hide_src_code = false; + browser->b.seek(&browser->b, -offset, SEEK_CUR); + browser->b.top_idx = bdl->idx - offset; + browser->b.index = bdl->idx; + } else { + if (bdl->idx_asm < 0) { + ui_helpline__puts("Only available for assembly lines."); + browser->b.seek(&browser->b, -offset, SEEK_CUR); + return false; + } + + if (bdl->idx_asm < offset) + offset = bdl->idx_asm; + + browser->b.nr_entries = browser->nr_asm_entries; + annotate_browser__opts.hide_src_code = true; + browser->b.seek(&browser->b, -offset, SEEK_CUR); + browser->b.top_idx = bdl->idx_asm - offset; + browser->b.index = bdl->idx_asm; + } + + return true; +} + +static void annotate_browser__init_asm_mode(struct annotate_browser *browser) +{ + ui_browser__reset_index(&browser->b); + browser->b.nr_entries = browser->nr_asm_entries; +} + +static bool annotate_browser__callq(struct annotate_browser *browser, int evidx, + struct hist_browser_timer *hbt) +{ + struct map_symbol *ms = browser->b.priv; + struct disasm_line *dl = browser->selection; + struct symbol *sym = ms->sym; + struct annotation *notes; + struct symbol *target; + u64 ip; + + if (!ins__is_call(dl->ins)) + return false; + + ip = ms->map->map_ip(ms->map, dl->ops.target.addr); + target = map__find_symbol(ms->map, ip, NULL); + if (target == NULL) { + ui_helpline__puts("The called function was not found."); + return true; + } + + notes = symbol__annotation(target); + pthread_mutex_lock(¬es->lock); + + if (notes->src == NULL && symbol__alloc_hist(target) < 0) { + pthread_mutex_unlock(¬es->lock); + ui__warning("Not enough memory for annotating '%s' symbol!\n", + target->name); + return true; + } + + pthread_mutex_unlock(¬es->lock); + symbol__tui_annotate(target, ms->map, evidx, hbt); + ui_browser__show_title(&browser->b, sym->name); + return true; +} + +static +struct disasm_line *annotate_browser__find_offset(struct annotate_browser *browser, + s64 offset, s64 *idx) +{ + struct map_symbol *ms = browser->b.priv; + struct symbol *sym = ms->sym; + struct annotation *notes = symbol__annotation(sym); + struct disasm_line *pos; + + *idx = 0; + list_for_each_entry(pos, ¬es->src->source, node) { + if (pos->offset == offset) + return pos; + if (!disasm_line__filter(&browser->b, &pos->node)) + ++*idx; + } + + return NULL; +} + +static bool annotate_browser__jump(struct annotate_browser *browser) +{ + struct disasm_line *dl = browser->selection; + s64 idx; + + if (!ins__is_jump(dl->ins)) + return false; + + dl = annotate_browser__find_offset(browser, dl->ops.target.offset, &idx); + if (dl == NULL) { + ui_helpline__puts("Invallid jump offset"); + return true; + } + + annotate_browser__set_top(browser, dl, idx); + + return true; +} + +static +struct disasm_line *annotate_browser__find_string(struct annotate_browser *browser, + char *s, s64 *idx) +{ + struct map_symbol *ms = browser->b.priv; + struct symbol *sym = ms->sym; + struct annotation *notes = symbol__annotation(sym); + struct disasm_line *pos = browser->selection; + + *idx = browser->b.index; + list_for_each_entry_continue(pos, ¬es->src->source, node) { + if (disasm_line__filter(&browser->b, &pos->node)) + continue; + + ++*idx; + + if (pos->line && strstr(pos->line, s) != NULL) + return pos; + } + + return NULL; +} + +static bool __annotate_browser__search(struct annotate_browser *browser) +{ + struct disasm_line *dl; + s64 idx; + + dl = annotate_browser__find_string(browser, browser->search_bf, &idx); + if (dl == NULL) { + ui_helpline__puts("String not found!"); + return false; + } + + annotate_browser__set_top(browser, dl, idx); + browser->searching_backwards = false; + return true; +} + +static +struct disasm_line *annotate_browser__find_string_reverse(struct annotate_browser *browser, + char *s, s64 *idx) +{ + struct map_symbol *ms = browser->b.priv; + struct symbol *sym = ms->sym; + struct annotation *notes = symbol__annotation(sym); + struct disasm_line *pos = browser->selection; + + *idx = browser->b.index; + list_for_each_entry_continue_reverse(pos, ¬es->src->source, node) { + if (disasm_line__filter(&browser->b, &pos->node)) + continue; + + --*idx; + + if (pos->line && strstr(pos->line, s) != NULL) + return pos; + } + + return NULL; +} + +static bool __annotate_browser__search_reverse(struct annotate_browser *browser) +{ + struct disasm_line *dl; + s64 idx; + + dl = annotate_browser__find_string_reverse(browser, browser->search_bf, &idx); + if (dl == NULL) { + ui_helpline__puts("String not found!"); + return false; + } + + annotate_browser__set_top(browser, dl, idx); + browser->searching_backwards = true; + return true; +} + +static bool annotate_browser__search_window(struct annotate_browser *browser, + int delay_secs) +{ + if (ui_browser__input_window("Search", "String: ", browser->search_bf, + "ENTER: OK, ESC: Cancel", + delay_secs * 2) != K_ENTER || + !*browser->search_bf) + return false; + + return true; +} + +static bool annotate_browser__search(struct annotate_browser *browser, int delay_secs) +{ + if (annotate_browser__search_window(browser, delay_secs)) + return __annotate_browser__search(browser); + + return false; +} + +static bool annotate_browser__continue_search(struct annotate_browser *browser, + int delay_secs) +{ + if (!*browser->search_bf) + return annotate_browser__search(browser, delay_secs); + + return __annotate_browser__search(browser); +} + +static bool annotate_browser__search_reverse(struct annotate_browser *browser, + int delay_secs) +{ + if (annotate_browser__search_window(browser, delay_secs)) + return __annotate_browser__search_reverse(browser); + + return false; +} + +static +bool annotate_browser__continue_search_reverse(struct annotate_browser *browser, + int delay_secs) +{ + if (!*browser->search_bf) + return annotate_browser__search_reverse(browser, delay_secs); + + return __annotate_browser__search_reverse(browser); +} + +static void annotate_browser__update_addr_width(struct annotate_browser *browser) +{ + if (annotate_browser__opts.use_offset) + browser->target_width = browser->min_addr_width; + else + browser->target_width = browser->max_addr_width; + + browser->addr_width = browser->target_width; + + if (annotate_browser__opts.show_nr_jumps) + browser->addr_width += browser->jumps_width + 1; +} + +static int annotate_browser__run(struct annotate_browser *browser, int evidx, + struct hist_browser_timer *hbt) +{ + struct rb_node *nd = NULL; + struct map_symbol *ms = browser->b.priv; + struct symbol *sym = ms->sym; + const char *help = "Press 'h' for help on key bindings"; + int delay_secs = hbt ? hbt->refresh : 0; + int key; + + if (ui_browser__show(&browser->b, sym->name, help) < 0) + return -1; + + annotate_browser__calc_percent(browser, evidx); + + if (browser->curr_hot) { + annotate_browser__set_rb_top(browser, browser->curr_hot); + browser->b.navkeypressed = false; + } + + nd = browser->curr_hot; + + while (1) { + key = ui_browser__run(&browser->b, delay_secs); + + if (delay_secs != 0) { + annotate_browser__calc_percent(browser, evidx); + /* + * Current line focus got out of the list of most active + * lines, NULL it so that if TAB|UNTAB is pressed, we + * move to curr_hot (current hottest line). + */ + if (nd != NULL && RB_EMPTY_NODE(nd)) + nd = NULL; + } + + switch (key) { + case K_TIMER: + if (hbt) + hbt->timer(hbt->arg); + + if (delay_secs != 0) + symbol__annotate_decay_histogram(sym, evidx); + continue; + case K_TAB: + if (nd != NULL) { + nd = rb_prev(nd); + if (nd == NULL) + nd = rb_last(&browser->entries); + } else + nd = browser->curr_hot; + break; + case K_UNTAB: + if (nd != NULL) + nd = rb_next(nd); + if (nd == NULL) + nd = rb_first(&browser->entries); + else + nd = browser->curr_hot; + break; + case K_F1: + case 'h': + ui_browser__help_window(&browser->b, + "UP/DOWN/PGUP\n" + "PGDN/SPACE Navigate\n" + "q/ESC/CTRL+C Exit\n\n" + "-> Go to target\n" + "<- Exit\n" + "H Cycle thru hottest instructions\n" + "j Toggle showing jump to target arrows\n" + "J Toggle showing number of jump sources on targets\n" + "n Search next string\n" + "o Toggle disassembler output/simplified view\n" + "s Toggle source code view\n" + "/ Search string\n" + "r Run available scripts\n" + "? Search previous string\n"); + continue; + case 'r': + { + script_browse(NULL); + continue; + } + case 'H': + nd = browser->curr_hot; + break; + case 's': + if (annotate_browser__toggle_source(browser)) + ui_helpline__puts(help); + continue; + case 'o': + annotate_browser__opts.use_offset = !annotate_browser__opts.use_offset; + annotate_browser__update_addr_width(browser); + continue; + case 'j': + annotate_browser__opts.jump_arrows = !annotate_browser__opts.jump_arrows; + continue; + case 'J': + annotate_browser__opts.show_nr_jumps = !annotate_browser__opts.show_nr_jumps; + annotate_browser__update_addr_width(browser); + continue; + case '/': + if (annotate_browser__search(browser, delay_secs)) { +show_help: + ui_helpline__puts(help); + } + continue; + case 'n': + if (browser->searching_backwards ? + annotate_browser__continue_search_reverse(browser, delay_secs) : + annotate_browser__continue_search(browser, delay_secs)) + goto show_help; + continue; + case '?': + if (annotate_browser__search_reverse(browser, delay_secs)) + goto show_help; + continue; + case 'D': { + static int seq; + ui_helpline__pop(); + ui_helpline__fpush("%d: nr_ent=%d, height=%d, idx=%d, top_idx=%d, nr_asm_entries=%d", + seq++, browser->b.nr_entries, + browser->b.height, + browser->b.index, + browser->b.top_idx, + browser->nr_asm_entries); + } + continue; + case K_ENTER: + case K_RIGHT: + if (browser->selection == NULL) + ui_helpline__puts("Huh? No selection. Report to linux-kernel@vger.kernel.org"); + else if (browser->selection->offset == -1) + ui_helpline__puts("Actions are only available for assembly lines."); + else if (!browser->selection->ins) { + if (strcmp(browser->selection->name, "retq")) + goto show_sup_ins; + goto out; + } else if (!(annotate_browser__jump(browser) || + annotate_browser__callq(browser, evidx, hbt))) { +show_sup_ins: + ui_helpline__puts("Actions are only available for 'callq', 'retq' & jump instructions."); + } + continue; + case K_LEFT: + case K_ESC: + case 'q': + case CTRL('c'): + goto out; + default: + continue; + } + + if (nd != NULL) + annotate_browser__set_rb_top(browser, nd); + } +out: + ui_browser__hide(&browser->b); + return key; +} + +int hist_entry__tui_annotate(struct hist_entry *he, int evidx, + struct hist_browser_timer *hbt) +{ + return symbol__tui_annotate(he->ms.sym, he->ms.map, evidx, hbt); +} + +static void annotate_browser__mark_jump_targets(struct annotate_browser *browser, + size_t size) +{ + u64 offset; + struct map_symbol *ms = browser->b.priv; + struct symbol *sym = ms->sym; + + /* PLT symbols contain external offsets */ + if (strstr(sym->name, "@plt")) + return; + + for (offset = 0; offset < size; ++offset) { + struct disasm_line *dl = browser->offsets[offset], *dlt; + struct browser_disasm_line *bdlt; + + if (!disasm_line__is_valid_jump(dl, sym)) + continue; + + dlt = browser->offsets[dl->ops.target.offset]; + /* + * FIXME: Oops, no jump target? Buggy disassembler? Or do we + * have to adjust to the previous offset? + */ + if (dlt == NULL) + continue; + + bdlt = disasm_line__browser(dlt); + if (++bdlt->jump_sources > browser->max_jump_sources) + browser->max_jump_sources = bdlt->jump_sources; + + ++browser->nr_jumps; + } + +} + +static inline int width_jumps(int n) +{ + if (n >= 100) + return 5; + if (n / 10) + return 2; + return 1; +} + +int symbol__tui_annotate(struct symbol *sym, struct map *map, int evidx, + struct hist_browser_timer *hbt) +{ + struct disasm_line *pos, *n; + struct annotation *notes; + size_t size; + struct map_symbol ms = { + .map = map, + .sym = sym, + }; + struct annotate_browser browser = { + .b = { + .refresh = annotate_browser__refresh, + .seek = ui_browser__list_head_seek, + .write = annotate_browser__write, + .filter = disasm_line__filter, + .priv = &ms, + .use_navkeypressed = true, + }, + }; + int ret = -1; + + if (sym == NULL) + return -1; + + size = symbol__size(sym); + + if (map->dso->annotate_warned) + return -1; + + browser.offsets = zalloc(size * sizeof(struct disasm_line *)); + if (browser.offsets == NULL) { + ui__error("Not enough memory!"); + return -1; + } + + if (symbol__annotate(sym, map, sizeof(struct browser_disasm_line)) < 0) { + ui__error("%s", ui_helpline__last_msg); + goto out_free_offsets; + } + + ui_helpline__push("Press <- or ESC to exit"); + + notes = symbol__annotation(sym); + browser.start = map__rip_2objdump(map, sym->start); + + list_for_each_entry(pos, ¬es->src->source, node) { + struct browser_disasm_line *bpos; + size_t line_len = strlen(pos->line); + + if (browser.b.width < line_len) + browser.b.width = line_len; + bpos = disasm_line__browser(pos); + bpos->idx = browser.nr_entries++; + if (pos->offset != -1) { + bpos->idx_asm = browser.nr_asm_entries++; + /* + * FIXME: short term bandaid to cope with assembly + * routines that comes with labels in the same column + * as the address in objdump, sigh. + * + * E.g. copy_user_generic_unrolled + */ + if (pos->offset < (s64)size) + browser.offsets[pos->offset] = pos; + } else + bpos->idx_asm = -1; + } + + annotate_browser__mark_jump_targets(&browser, size); + + browser.addr_width = browser.target_width = browser.min_addr_width = hex_width(size); + browser.max_addr_width = hex_width(sym->end); + browser.jumps_width = width_jumps(browser.max_jump_sources); + browser.b.nr_entries = browser.nr_entries; + browser.b.entries = ¬es->src->source, + browser.b.width += 18; /* Percentage */ + + if (annotate_browser__opts.hide_src_code) + annotate_browser__init_asm_mode(&browser); + + annotate_browser__update_addr_width(&browser); + + ret = annotate_browser__run(&browser, evidx, hbt); + list_for_each_entry_safe(pos, n, ¬es->src->source, node) { + list_del(&pos->node); + disasm_line__free(pos); + } + +out_free_offsets: + free(browser.offsets); + return ret; +} + +#define ANNOTATE_CFG(n) \ + { .name = #n, .value = &annotate_browser__opts.n, } + +/* + * Keep the entries sorted, they are bsearch'ed + */ +static struct annotate_config { + const char *name; + bool *value; +} annotate__configs[] = { + ANNOTATE_CFG(hide_src_code), + ANNOTATE_CFG(jump_arrows), + ANNOTATE_CFG(show_nr_jumps), + ANNOTATE_CFG(use_offset), +}; + +#undef ANNOTATE_CFG + +static int annotate_config__cmp(const void *name, const void *cfgp) +{ + const struct annotate_config *cfg = cfgp; + + return strcmp(name, cfg->name); +} + +static int annotate__config(const char *var, const char *value, + void *data __maybe_unused) +{ + struct annotate_config *cfg; + const char *name; + + if (prefixcmp(var, "annotate.") != 0) + return 0; + + name = var + 9; + cfg = bsearch(name, annotate__configs, ARRAY_SIZE(annotate__configs), + sizeof(struct annotate_config), annotate_config__cmp); + + if (cfg == NULL) + return -1; + + *cfg->value = perf_config_bool(name, value); + return 0; +} + +void annotate_browser__init(void) +{ + perf_config(annotate__config, NULL); +} diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c new file mode 100644 index 00000000..aa227040 --- /dev/null +++ b/tools/perf/ui/browsers/hists.c @@ -0,0 +1,1906 @@ +#include <stdio.h> +#include "../libslang.h" +#include <stdlib.h> +#include <string.h> +#include <newt.h> +#include <linux/rbtree.h> + +#include "../../util/evsel.h" +#include "../../util/evlist.h" +#include "../../util/hist.h" +#include "../../util/pstack.h" +#include "../../util/sort.h" +#include "../../util/util.h" +#include "../../arch/common.h" + +#include "../browser.h" +#include "../helpline.h" +#include "../util.h" +#include "../ui.h" +#include "map.h" + +struct hist_browser { + struct ui_browser b; + struct hists *hists; + struct hist_entry *he_selection; + struct map_symbol *selection; + int print_seq; + bool show_dso; + bool has_symbols; +}; + +extern void hist_browser__init_hpp(void); + +static int hists__browser_title(struct hists *hists, char *bf, size_t size, + const char *ev_name); + +static void hist_browser__refresh_dimensions(struct hist_browser *browser) +{ + /* 3 == +/- toggle symbol before actual hist_entry rendering */ + browser->b.width = 3 + (hists__sort_list_width(browser->hists) + + sizeof("[k]")); +} + +static void hist_browser__reset(struct hist_browser *browser) +{ + browser->b.nr_entries = browser->hists->nr_entries; + hist_browser__refresh_dimensions(browser); + ui_browser__reset_index(&browser->b); +} + +static char tree__folded_sign(bool unfolded) +{ + return unfolded ? '-' : '+'; +} + +static char map_symbol__folded(const struct map_symbol *ms) +{ + return ms->has_children ? tree__folded_sign(ms->unfolded) : ' '; +} + +static char hist_entry__folded(const struct hist_entry *he) +{ + return map_symbol__folded(&he->ms); +} + +static char callchain_list__folded(const struct callchain_list *cl) +{ + return map_symbol__folded(&cl->ms); +} + +static void map_symbol__set_folding(struct map_symbol *ms, bool unfold) +{ + ms->unfolded = unfold ? ms->has_children : false; +} + +static int callchain_node__count_rows_rb_tree(struct callchain_node *node) +{ + int n = 0; + struct rb_node *nd; + + for (nd = rb_first(&node->rb_root); nd; nd = rb_next(nd)) { + struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node); + struct callchain_list *chain; + char folded_sign = ' '; /* No children */ + + list_for_each_entry(chain, &child->val, list) { + ++n; + /* We need this because we may not have children */ + folded_sign = callchain_list__folded(chain); + if (folded_sign == '+') + break; + } + + if (folded_sign == '-') /* Have children and they're unfolded */ + n += callchain_node__count_rows_rb_tree(child); + } + + return n; +} + +static int callchain_node__count_rows(struct callchain_node *node) +{ + struct callchain_list *chain; + bool unfolded = false; + int n = 0; + + list_for_each_entry(chain, &node->val, list) { + ++n; + unfolded = chain->ms.unfolded; + } + + if (unfolded) + n += callchain_node__count_rows_rb_tree(node); + + return n; +} + +static int callchain__count_rows(struct rb_root *chain) +{ + struct rb_node *nd; + int n = 0; + + for (nd = rb_first(chain); nd; nd = rb_next(nd)) { + struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node); + n += callchain_node__count_rows(node); + } + + return n; +} + +static bool map_symbol__toggle_fold(struct map_symbol *ms) +{ + if (!ms) + return false; + + if (!ms->has_children) + return false; + + ms->unfolded = !ms->unfolded; + return true; +} + +static void callchain_node__init_have_children_rb_tree(struct callchain_node *node) +{ + struct rb_node *nd = rb_first(&node->rb_root); + + for (nd = rb_first(&node->rb_root); nd; nd = rb_next(nd)) { + struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node); + struct callchain_list *chain; + bool first = true; + + list_for_each_entry(chain, &child->val, list) { + if (first) { + first = false; + chain->ms.has_children = chain->list.next != &child->val || + !RB_EMPTY_ROOT(&child->rb_root); + } else + chain->ms.has_children = chain->list.next == &child->val && + !RB_EMPTY_ROOT(&child->rb_root); + } + + callchain_node__init_have_children_rb_tree(child); + } +} + +static void callchain_node__init_have_children(struct callchain_node *node) +{ + struct callchain_list *chain; + + list_for_each_entry(chain, &node->val, list) + chain->ms.has_children = !RB_EMPTY_ROOT(&node->rb_root); + + callchain_node__init_have_children_rb_tree(node); +} + +static void callchain__init_have_children(struct rb_root *root) +{ + struct rb_node *nd; + + for (nd = rb_first(root); nd; nd = rb_next(nd)) { + struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node); + callchain_node__init_have_children(node); + } +} + +static void hist_entry__init_have_children(struct hist_entry *he) +{ + if (!he->init_have_children) { + he->ms.has_children = !RB_EMPTY_ROOT(&he->sorted_chain); + callchain__init_have_children(&he->sorted_chain); + he->init_have_children = true; + } +} + +static bool hist_browser__toggle_fold(struct hist_browser *browser) +{ + if (map_symbol__toggle_fold(browser->selection)) { + struct hist_entry *he = browser->he_selection; + + hist_entry__init_have_children(he); + browser->hists->nr_entries -= he->nr_rows; + + if (he->ms.unfolded) + he->nr_rows = callchain__count_rows(&he->sorted_chain); + else + he->nr_rows = 0; + browser->hists->nr_entries += he->nr_rows; + browser->b.nr_entries = browser->hists->nr_entries; + + return true; + } + + /* If it doesn't have children, no toggling performed */ + return false; +} + +static int callchain_node__set_folding_rb_tree(struct callchain_node *node, bool unfold) +{ + int n = 0; + struct rb_node *nd; + + for (nd = rb_first(&node->rb_root); nd; nd = rb_next(nd)) { + struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node); + struct callchain_list *chain; + bool has_children = false; + + list_for_each_entry(chain, &child->val, list) { + ++n; + map_symbol__set_folding(&chain->ms, unfold); + has_children = chain->ms.has_children; + } + + if (has_children) + n += callchain_node__set_folding_rb_tree(child, unfold); + } + + return n; +} + +static int callchain_node__set_folding(struct callchain_node *node, bool unfold) +{ + struct callchain_list *chain; + bool has_children = false; + int n = 0; + + list_for_each_entry(chain, &node->val, list) { + ++n; + map_symbol__set_folding(&chain->ms, unfold); + has_children = chain->ms.has_children; + } + + if (has_children) + n += callchain_node__set_folding_rb_tree(node, unfold); + + return n; +} + +static int callchain__set_folding(struct rb_root *chain, bool unfold) +{ + struct rb_node *nd; + int n = 0; + + for (nd = rb_first(chain); nd; nd = rb_next(nd)) { + struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node); + n += callchain_node__set_folding(node, unfold); + } + + return n; +} + +static void hist_entry__set_folding(struct hist_entry *he, bool unfold) +{ + hist_entry__init_have_children(he); + map_symbol__set_folding(&he->ms, unfold); + + if (he->ms.has_children) { + int n = callchain__set_folding(&he->sorted_chain, unfold); + he->nr_rows = unfold ? n : 0; + } else + he->nr_rows = 0; +} + +static void hists__set_folding(struct hists *hists, bool unfold) +{ + struct rb_node *nd; + + hists->nr_entries = 0; + + for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) { + struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node); + hist_entry__set_folding(he, unfold); + hists->nr_entries += 1 + he->nr_rows; + } +} + +static void hist_browser__set_folding(struct hist_browser *browser, bool unfold) +{ + hists__set_folding(browser->hists, unfold); + browser->b.nr_entries = browser->hists->nr_entries; + /* Go to the start, we may be way after valid entries after a collapse */ + ui_browser__reset_index(&browser->b); +} + +static void ui_browser__warn_lost_events(struct ui_browser *browser) +{ + ui_browser__warning(browser, 4, + "Events are being lost, check IO/CPU overload!\n\n" + "You may want to run 'perf' using a RT scheduler policy:\n\n" + " perf top -r 80\n\n" + "Or reduce the sampling frequency."); +} + +static int hist_browser__run(struct hist_browser *browser, const char *ev_name, + struct hist_browser_timer *hbt) +{ + int key; + char title[160]; + int delay_secs = hbt ? hbt->refresh : 0; + + browser->b.entries = &browser->hists->entries; + browser->b.nr_entries = browser->hists->nr_entries; + + hist_browser__refresh_dimensions(browser); + hists__browser_title(browser->hists, title, sizeof(title), ev_name); + + if (ui_browser__show(&browser->b, title, + "Press '?' for help on key bindings") < 0) + return -1; + + while (1) { + key = ui_browser__run(&browser->b, delay_secs); + + switch (key) { + case K_TIMER: + hbt->timer(hbt->arg); + ui_browser__update_nr_entries(&browser->b, browser->hists->nr_entries); + + if (browser->hists->stats.nr_lost_warned != + browser->hists->stats.nr_events[PERF_RECORD_LOST]) { + browser->hists->stats.nr_lost_warned = + browser->hists->stats.nr_events[PERF_RECORD_LOST]; + ui_browser__warn_lost_events(&browser->b); + } + + hists__browser_title(browser->hists, title, sizeof(title), ev_name); + ui_browser__show_title(&browser->b, title); + continue; + case 'D': { /* Debug */ + static int seq; + struct hist_entry *h = rb_entry(browser->b.top, + struct hist_entry, rb_node); + ui_helpline__pop(); + ui_helpline__fpush("%d: nr_ent=(%d,%d), height=%d, idx=%d, fve: idx=%d, row_off=%d, nrows=%d", + seq++, browser->b.nr_entries, + browser->hists->nr_entries, + browser->b.height, + browser->b.index, + browser->b.top_idx, + h->row_offset, h->nr_rows); + } + break; + case 'C': + /* Collapse the whole world. */ + hist_browser__set_folding(browser, false); + break; + case 'E': + /* Expand the whole world. */ + hist_browser__set_folding(browser, true); + break; + case K_ENTER: + if (hist_browser__toggle_fold(browser)) + break; + /* fall thru */ + default: + goto out; + } + } +out: + ui_browser__hide(&browser->b); + return key; +} + +static char *callchain_list__sym_name(struct callchain_list *cl, + char *bf, size_t bfsize, bool show_dso) +{ + int printed; + + if (cl->ms.sym) + printed = scnprintf(bf, bfsize, "%s", cl->ms.sym->name); + else + printed = scnprintf(bf, bfsize, "%#" PRIx64, cl->ip); + + if (show_dso) + scnprintf(bf + printed, bfsize - printed, " %s", + cl->ms.map ? cl->ms.map->dso->short_name : "unknown"); + + return bf; +} + +#define LEVEL_OFFSET_STEP 3 + +static int hist_browser__show_callchain_node_rb_tree(struct hist_browser *browser, + struct callchain_node *chain_node, + u64 total, int level, + unsigned short row, + off_t *row_offset, + bool *is_current_entry) +{ + struct rb_node *node; + int first_row = row, width, offset = level * LEVEL_OFFSET_STEP; + u64 new_total, remaining; + + if (callchain_param.mode == CHAIN_GRAPH_REL) + new_total = chain_node->children_hit; + else + new_total = total; + + remaining = new_total; + node = rb_first(&chain_node->rb_root); + while (node) { + struct callchain_node *child = rb_entry(node, struct callchain_node, rb_node); + struct rb_node *next = rb_next(node); + u64 cumul = callchain_cumul_hits(child); + struct callchain_list *chain; + char folded_sign = ' '; + int first = true; + int extra_offset = 0; + + remaining -= cumul; + + list_for_each_entry(chain, &child->val, list) { + char bf[1024], *alloc_str; + const char *str; + int color; + bool was_first = first; + + if (first) + first = false; + else + extra_offset = LEVEL_OFFSET_STEP; + + folded_sign = callchain_list__folded(chain); + if (*row_offset != 0) { + --*row_offset; + goto do_next; + } + + alloc_str = NULL; + str = callchain_list__sym_name(chain, bf, sizeof(bf), + browser->show_dso); + if (was_first) { + double percent = cumul * 100.0 / new_total; + + if (asprintf(&alloc_str, "%2.2f%% %s", percent, str) < 0) + str = "Not enough memory!"; + else + str = alloc_str; + } + + color = HE_COLORSET_NORMAL; + width = browser->b.width - (offset + extra_offset + 2); + if (ui_browser__is_current_entry(&browser->b, row)) { + browser->selection = &chain->ms; + color = HE_COLORSET_SELECTED; + *is_current_entry = true; + } + + ui_browser__set_color(&browser->b, color); + ui_browser__gotorc(&browser->b, row, 0); + slsmg_write_nstring(" ", offset + extra_offset); + slsmg_printf("%c ", folded_sign); + slsmg_write_nstring(str, width); + free(alloc_str); + + if (++row == browser->b.height) + goto out; +do_next: + if (folded_sign == '+') + break; + } + + if (folded_sign == '-') { + const int new_level = level + (extra_offset ? 2 : 1); + row += hist_browser__show_callchain_node_rb_tree(browser, child, new_total, + new_level, row, row_offset, + is_current_entry); + } + if (row == browser->b.height) + goto out; + node = next; + } +out: + return row - first_row; +} + +static int hist_browser__show_callchain_node(struct hist_browser *browser, + struct callchain_node *node, + int level, unsigned short row, + off_t *row_offset, + bool *is_current_entry) +{ + struct callchain_list *chain; + int first_row = row, + offset = level * LEVEL_OFFSET_STEP, + width = browser->b.width - offset; + char folded_sign = ' '; + + list_for_each_entry(chain, &node->val, list) { + char bf[1024], *s; + int color; + + folded_sign = callchain_list__folded(chain); + + if (*row_offset != 0) { + --*row_offset; + continue; + } + + color = HE_COLORSET_NORMAL; + if (ui_browser__is_current_entry(&browser->b, row)) { + browser->selection = &chain->ms; + color = HE_COLORSET_SELECTED; + *is_current_entry = true; + } + + s = callchain_list__sym_name(chain, bf, sizeof(bf), + browser->show_dso); + ui_browser__gotorc(&browser->b, row, 0); + ui_browser__set_color(&browser->b, color); + slsmg_write_nstring(" ", offset); + slsmg_printf("%c ", folded_sign); + slsmg_write_nstring(s, width - 2); + + if (++row == browser->b.height) + goto out; + } + + if (folded_sign == '-') + row += hist_browser__show_callchain_node_rb_tree(browser, node, + browser->hists->stats.total_period, + level + 1, row, + row_offset, + is_current_entry); +out: + return row - first_row; +} + +static int hist_browser__show_callchain(struct hist_browser *browser, + struct rb_root *chain, + int level, unsigned short row, + off_t *row_offset, + bool *is_current_entry) +{ + struct rb_node *nd; + int first_row = row; + + for (nd = rb_first(chain); nd; nd = rb_next(nd)) { + struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node); + + row += hist_browser__show_callchain_node(browser, node, level, + row, row_offset, + is_current_entry); + if (row == browser->b.height) + break; + } + + return row - first_row; +} + +struct hpp_arg { + struct ui_browser *b; + char folded_sign; + bool current_entry; +}; + +static int __hpp__color_callchain(struct hpp_arg *arg) +{ + if (!symbol_conf.use_callchain) + return 0; + + slsmg_printf("%c ", arg->folded_sign); + return 2; +} + +static int __hpp__color_fmt(struct perf_hpp *hpp, struct hist_entry *he, + u64 (*get_field)(struct hist_entry *), + int (*callchain_cb)(struct hpp_arg *)) +{ + int ret = 0; + double percent = 0.0; + struct hists *hists = he->hists; + struct hpp_arg *arg = hpp->ptr; + + if (hists->stats.total_period) + percent = 100.0 * get_field(he) / hists->stats.total_period; + + ui_browser__set_percent_color(arg->b, percent, arg->current_entry); + + if (callchain_cb) + ret += callchain_cb(arg); + + ret += scnprintf(hpp->buf, hpp->size, "%6.2f%%", percent); + slsmg_printf("%s", hpp->buf); + + if (symbol_conf.event_group) { + int prev_idx, idx_delta; + struct perf_evsel *evsel = hists_to_evsel(hists); + struct hist_entry *pair; + int nr_members = evsel->nr_members; + + if (nr_members <= 1) + goto out; + + prev_idx = perf_evsel__group_idx(evsel); + + list_for_each_entry(pair, &he->pairs.head, pairs.node) { + u64 period = get_field(pair); + u64 total = pair->hists->stats.total_period; + + if (!total) + continue; + + evsel = hists_to_evsel(pair->hists); + idx_delta = perf_evsel__group_idx(evsel) - prev_idx - 1; + + while (idx_delta--) { + /* + * zero-fill group members in the middle which + * have no sample + */ + ui_browser__set_percent_color(arg->b, 0.0, + arg->current_entry); + ret += scnprintf(hpp->buf, hpp->size, + " %6.2f%%", 0.0); + slsmg_printf("%s", hpp->buf); + } + + percent = 100.0 * period / total; + ui_browser__set_percent_color(arg->b, percent, + arg->current_entry); + ret += scnprintf(hpp->buf, hpp->size, + " %6.2f%%", percent); + slsmg_printf("%s", hpp->buf); + + prev_idx = perf_evsel__group_idx(evsel); + } + + idx_delta = nr_members - prev_idx - 1; + + while (idx_delta--) { + /* + * zero-fill group members at last which have no sample + */ + ui_browser__set_percent_color(arg->b, 0.0, + arg->current_entry); + ret += scnprintf(hpp->buf, hpp->size, + " %6.2f%%", 0.0); + slsmg_printf("%s", hpp->buf); + } + } +out: + if (!arg->current_entry || !arg->b->navkeypressed) + ui_browser__set_color(arg->b, HE_COLORSET_NORMAL); + + return ret; +} + +#define __HPP_COLOR_PERCENT_FN(_type, _field, _cb) \ +static u64 __hpp_get_##_field(struct hist_entry *he) \ +{ \ + return he->stat._field; \ +} \ + \ +static int hist_browser__hpp_color_##_type(struct perf_hpp *hpp, \ + struct hist_entry *he) \ +{ \ + return __hpp__color_fmt(hpp, he, __hpp_get_##_field, _cb); \ +} + +__HPP_COLOR_PERCENT_FN(overhead, period, __hpp__color_callchain) +__HPP_COLOR_PERCENT_FN(overhead_sys, period_sys, NULL) +__HPP_COLOR_PERCENT_FN(overhead_us, period_us, NULL) +__HPP_COLOR_PERCENT_FN(overhead_guest_sys, period_guest_sys, NULL) +__HPP_COLOR_PERCENT_FN(overhead_guest_us, period_guest_us, NULL) + +#undef __HPP_COLOR_PERCENT_FN + +void hist_browser__init_hpp(void) +{ + perf_hpp__column_enable(PERF_HPP__OVERHEAD); + + perf_hpp__init(); + + perf_hpp__format[PERF_HPP__OVERHEAD].color = + hist_browser__hpp_color_overhead; + perf_hpp__format[PERF_HPP__OVERHEAD_SYS].color = + hist_browser__hpp_color_overhead_sys; + perf_hpp__format[PERF_HPP__OVERHEAD_US].color = + hist_browser__hpp_color_overhead_us; + perf_hpp__format[PERF_HPP__OVERHEAD_GUEST_SYS].color = + hist_browser__hpp_color_overhead_guest_sys; + perf_hpp__format[PERF_HPP__OVERHEAD_GUEST_US].color = + hist_browser__hpp_color_overhead_guest_us; +} + +static int hist_browser__show_entry(struct hist_browser *browser, + struct hist_entry *entry, + unsigned short row) +{ + char s[256]; + int printed = 0; + int width = browser->b.width; + char folded_sign = ' '; + bool current_entry = ui_browser__is_current_entry(&browser->b, row); + off_t row_offset = entry->row_offset; + bool first = true; + struct perf_hpp_fmt *fmt; + + if (current_entry) { + browser->he_selection = entry; + browser->selection = &entry->ms; + } + + if (symbol_conf.use_callchain) { + hist_entry__init_have_children(entry); + folded_sign = hist_entry__folded(entry); + } + + if (row_offset == 0) { + struct hpp_arg arg = { + .b = &browser->b, + .folded_sign = folded_sign, + .current_entry = current_entry, + }; + struct perf_hpp hpp = { + .buf = s, + .size = sizeof(s), + .ptr = &arg, + }; + + ui_browser__gotorc(&browser->b, row, 0); + + perf_hpp__for_each_format(fmt) { + if (!first) { + slsmg_printf(" "); + width -= 2; + } + first = false; + + if (fmt->color) { + width -= fmt->color(&hpp, entry); + } else { + width -= fmt->entry(&hpp, entry); + slsmg_printf("%s", s); + } + } + + /* The scroll bar isn't being used */ + if (!browser->b.navkeypressed) + width += 1; + + hist_entry__sort_snprintf(entry, s, sizeof(s), browser->hists); + slsmg_write_nstring(s, width); + ++row; + ++printed; + } else + --row_offset; + + if (folded_sign == '-' && row != browser->b.height) { + printed += hist_browser__show_callchain(browser, &entry->sorted_chain, + 1, row, &row_offset, + ¤t_entry); + if (current_entry) + browser->he_selection = entry; + } + + return printed; +} + +static void ui_browser__hists_init_top(struct ui_browser *browser) +{ + if (browser->top == NULL) { + struct hist_browser *hb; + + hb = container_of(browser, struct hist_browser, b); + browser->top = rb_first(&hb->hists->entries); + } +} + +static unsigned int hist_browser__refresh(struct ui_browser *browser) +{ + unsigned row = 0; + struct rb_node *nd; + struct hist_browser *hb = container_of(browser, struct hist_browser, b); + + ui_browser__hists_init_top(browser); + + for (nd = browser->top; nd; nd = rb_next(nd)) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + + if (h->filtered) + continue; + + row += hist_browser__show_entry(hb, h, row); + if (row == browser->height) + break; + } + + return row; +} + +static struct rb_node *hists__filter_entries(struct rb_node *nd) +{ + while (nd != NULL) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + if (!h->filtered) + return nd; + + nd = rb_next(nd); + } + + return NULL; +} + +static struct rb_node *hists__filter_prev_entries(struct rb_node *nd) +{ + while (nd != NULL) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + if (!h->filtered) + return nd; + + nd = rb_prev(nd); + } + + return NULL; +} + +static void ui_browser__hists_seek(struct ui_browser *browser, + off_t offset, int whence) +{ + struct hist_entry *h; + struct rb_node *nd; + bool first = true; + + if (browser->nr_entries == 0) + return; + + ui_browser__hists_init_top(browser); + + switch (whence) { + case SEEK_SET: + nd = hists__filter_entries(rb_first(browser->entries)); + break; + case SEEK_CUR: + nd = browser->top; + goto do_offset; + case SEEK_END: + nd = hists__filter_prev_entries(rb_last(browser->entries)); + first = false; + break; + default: + return; + } + + /* + * Moves not relative to the first visible entry invalidates its + * row_offset: + */ + h = rb_entry(browser->top, struct hist_entry, rb_node); + h->row_offset = 0; + + /* + * Here we have to check if nd is expanded (+), if it is we can't go + * the next top level hist_entry, instead we must compute an offset of + * what _not_ to show and not change the first visible entry. + * + * This offset increments when we are going from top to bottom and + * decreases when we're going from bottom to top. + * + * As we don't have backpointers to the top level in the callchains + * structure, we need to always print the whole hist_entry callchain, + * skipping the first ones that are before the first visible entry + * and stop when we printed enough lines to fill the screen. + */ +do_offset: + if (offset > 0) { + do { + h = rb_entry(nd, struct hist_entry, rb_node); + if (h->ms.unfolded) { + u16 remaining = h->nr_rows - h->row_offset; + if (offset > remaining) { + offset -= remaining; + h->row_offset = 0; + } else { + h->row_offset += offset; + offset = 0; + browser->top = nd; + break; + } + } + nd = hists__filter_entries(rb_next(nd)); + if (nd == NULL) + break; + --offset; + browser->top = nd; + } while (offset != 0); + } else if (offset < 0) { + while (1) { + h = rb_entry(nd, struct hist_entry, rb_node); + if (h->ms.unfolded) { + if (first) { + if (-offset > h->row_offset) { + offset += h->row_offset; + h->row_offset = 0; + } else { + h->row_offset += offset; + offset = 0; + browser->top = nd; + break; + } + } else { + if (-offset > h->nr_rows) { + offset += h->nr_rows; + h->row_offset = 0; + } else { + h->row_offset = h->nr_rows + offset; + offset = 0; + browser->top = nd; + break; + } + } + } + + nd = hists__filter_prev_entries(rb_prev(nd)); + if (nd == NULL) + break; + ++offset; + browser->top = nd; + if (offset == 0) { + /* + * Last unfiltered hist_entry, check if it is + * unfolded, if it is then we should have + * row_offset at its last entry. + */ + h = rb_entry(nd, struct hist_entry, rb_node); + if (h->ms.unfolded) + h->row_offset = h->nr_rows; + break; + } + first = false; + } + } else { + browser->top = nd; + h = rb_entry(nd, struct hist_entry, rb_node); + h->row_offset = 0; + } +} + +static int hist_browser__fprintf_callchain_node_rb_tree(struct hist_browser *browser, + struct callchain_node *chain_node, + u64 total, int level, + FILE *fp) +{ + struct rb_node *node; + int offset = level * LEVEL_OFFSET_STEP; + u64 new_total, remaining; + int printed = 0; + + if (callchain_param.mode == CHAIN_GRAPH_REL) + new_total = chain_node->children_hit; + else + new_total = total; + + remaining = new_total; + node = rb_first(&chain_node->rb_root); + while (node) { + struct callchain_node *child = rb_entry(node, struct callchain_node, rb_node); + struct rb_node *next = rb_next(node); + u64 cumul = callchain_cumul_hits(child); + struct callchain_list *chain; + char folded_sign = ' '; + int first = true; + int extra_offset = 0; + + remaining -= cumul; + + list_for_each_entry(chain, &child->val, list) { + char bf[1024], *alloc_str; + const char *str; + bool was_first = first; + + if (first) + first = false; + else + extra_offset = LEVEL_OFFSET_STEP; + + folded_sign = callchain_list__folded(chain); + + alloc_str = NULL; + str = callchain_list__sym_name(chain, bf, sizeof(bf), + browser->show_dso); + if (was_first) { + double percent = cumul * 100.0 / new_total; + + if (asprintf(&alloc_str, "%2.2f%% %s", percent, str) < 0) + str = "Not enough memory!"; + else + str = alloc_str; + } + + printed += fprintf(fp, "%*s%c %s\n", offset + extra_offset, " ", folded_sign, str); + free(alloc_str); + if (folded_sign == '+') + break; + } + + if (folded_sign == '-') { + const int new_level = level + (extra_offset ? 2 : 1); + printed += hist_browser__fprintf_callchain_node_rb_tree(browser, child, new_total, + new_level, fp); + } + + node = next; + } + + return printed; +} + +static int hist_browser__fprintf_callchain_node(struct hist_browser *browser, + struct callchain_node *node, + int level, FILE *fp) +{ + struct callchain_list *chain; + int offset = level * LEVEL_OFFSET_STEP; + char folded_sign = ' '; + int printed = 0; + + list_for_each_entry(chain, &node->val, list) { + char bf[1024], *s; + + folded_sign = callchain_list__folded(chain); + s = callchain_list__sym_name(chain, bf, sizeof(bf), browser->show_dso); + printed += fprintf(fp, "%*s%c %s\n", offset, " ", folded_sign, s); + } + + if (folded_sign == '-') + printed += hist_browser__fprintf_callchain_node_rb_tree(browser, node, + browser->hists->stats.total_period, + level + 1, fp); + return printed; +} + +static int hist_browser__fprintf_callchain(struct hist_browser *browser, + struct rb_root *chain, int level, FILE *fp) +{ + struct rb_node *nd; + int printed = 0; + + for (nd = rb_first(chain); nd; nd = rb_next(nd)) { + struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node); + + printed += hist_browser__fprintf_callchain_node(browser, node, level, fp); + } + + return printed; +} + +static int hist_browser__fprintf_entry(struct hist_browser *browser, + struct hist_entry *he, FILE *fp) +{ + char s[8192]; + double percent; + int printed = 0; + char folded_sign = ' '; + + if (symbol_conf.use_callchain) + folded_sign = hist_entry__folded(he); + + hist_entry__sort_snprintf(he, s, sizeof(s), browser->hists); + percent = (he->stat.period * 100.0) / browser->hists->stats.total_period; + + if (symbol_conf.use_callchain) + printed += fprintf(fp, "%c ", folded_sign); + + printed += fprintf(fp, " %5.2f%%", percent); + + if (symbol_conf.show_nr_samples) + printed += fprintf(fp, " %11u", he->stat.nr_events); + + if (symbol_conf.show_total_period) + printed += fprintf(fp, " %12" PRIu64, he->stat.period); + + printed += fprintf(fp, "%s\n", rtrim(s)); + + if (folded_sign == '-') + printed += hist_browser__fprintf_callchain(browser, &he->sorted_chain, 1, fp); + + return printed; +} + +static int hist_browser__fprintf(struct hist_browser *browser, FILE *fp) +{ + struct rb_node *nd = hists__filter_entries(rb_first(browser->b.entries)); + int printed = 0; + + while (nd) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + + printed += hist_browser__fprintf_entry(browser, h, fp); + nd = hists__filter_entries(rb_next(nd)); + } + + return printed; +} + +static int hist_browser__dump(struct hist_browser *browser) +{ + char filename[64]; + FILE *fp; + + while (1) { + scnprintf(filename, sizeof(filename), "perf.hist.%d", browser->print_seq); + if (access(filename, F_OK)) + break; + /* + * XXX: Just an arbitrary lazy upper limit + */ + if (++browser->print_seq == 8192) { + ui_helpline__fpush("Too many perf.hist.N files, nothing written!"); + return -1; + } + } + + fp = fopen(filename, "w"); + if (fp == NULL) { + char bf[64]; + const char *err = strerror_r(errno, bf, sizeof(bf)); + ui_helpline__fpush("Couldn't write to %s: %s", filename, err); + return -1; + } + + ++browser->print_seq; + hist_browser__fprintf(browser, fp); + fclose(fp); + ui_helpline__fpush("%s written!", filename); + + return 0; +} + +static struct hist_browser *hist_browser__new(struct hists *hists) +{ + struct hist_browser *browser = zalloc(sizeof(*browser)); + + if (browser) { + browser->hists = hists; + browser->b.refresh = hist_browser__refresh; + browser->b.seek = ui_browser__hists_seek; + browser->b.use_navkeypressed = true; + if (sort__branch_mode == 1) + browser->has_symbols = sort_sym_from.list.next != NULL; + else + browser->has_symbols = sort_sym.list.next != NULL; + } + + return browser; +} + +static void hist_browser__delete(struct hist_browser *browser) +{ + free(browser); +} + +static struct hist_entry *hist_browser__selected_entry(struct hist_browser *browser) +{ + return browser->he_selection; +} + +static struct thread *hist_browser__selected_thread(struct hist_browser *browser) +{ + return browser->he_selection->thread; +} + +static int hists__browser_title(struct hists *hists, char *bf, size_t size, + const char *ev_name) +{ + char unit; + int printed; + const struct dso *dso = hists->dso_filter; + const struct thread *thread = hists->thread_filter; + unsigned long nr_samples = hists->stats.nr_events[PERF_RECORD_SAMPLE]; + u64 nr_events = hists->stats.total_period; + struct perf_evsel *evsel = hists_to_evsel(hists); + char buf[512]; + size_t buflen = sizeof(buf); + + if (symbol_conf.event_group && evsel->nr_members > 1) { + struct perf_evsel *pos; + + perf_evsel__group_desc(evsel, buf, buflen); + ev_name = buf; + + for_each_group_member(pos, evsel) { + nr_samples += pos->hists.stats.nr_events[PERF_RECORD_SAMPLE]; + nr_events += pos->hists.stats.total_period; + } + } + + nr_samples = convert_unit(nr_samples, &unit); + printed = scnprintf(bf, size, + "Samples: %lu%c of event '%s', Event count (approx.): %lu", + nr_samples, unit, ev_name, nr_events); + + + if (hists->uid_filter_str) + printed += snprintf(bf + printed, size - printed, + ", UID: %s", hists->uid_filter_str); + if (thread) + printed += scnprintf(bf + printed, size - printed, + ", Thread: %s(%d)", + (thread->comm_set ? thread->comm : ""), + thread->pid); + if (dso) + printed += scnprintf(bf + printed, size - printed, + ", DSO: %s", dso->short_name); + return printed; +} + +static inline void free_popup_options(char **options, int n) +{ + int i; + + for (i = 0; i < n; ++i) { + free(options[i]); + options[i] = NULL; + } +} + +/* Check whether the browser is for 'top' or 'report' */ +static inline bool is_report_browser(void *timer) +{ + return timer == NULL; +} + +/* + * Only runtime switching of perf data file will make "input_name" point + * to a malloced buffer. So add "is_input_name_malloced" flag to decide + * whether we need to call free() for current "input_name" during the switch. + */ +static bool is_input_name_malloced = false; + +static int switch_data_file(void) +{ + char *pwd, *options[32], *abs_path[32], *tmp; + DIR *pwd_dir; + int nr_options = 0, choice = -1, ret = -1; + struct dirent *dent; + + pwd = getenv("PWD"); + if (!pwd) + return ret; + + pwd_dir = opendir(pwd); + if (!pwd_dir) + return ret; + + memset(options, 0, sizeof(options)); + memset(options, 0, sizeof(abs_path)); + + while ((dent = readdir(pwd_dir))) { + char path[PATH_MAX]; + u64 magic; + char *name = dent->d_name; + FILE *file; + + if (!(dent->d_type == DT_REG)) + continue; + + snprintf(path, sizeof(path), "%s/%s", pwd, name); + + file = fopen(path, "r"); + if (!file) + continue; + + if (fread(&magic, 1, 8, file) < 8) + goto close_file_and_continue; + + if (is_perf_magic(magic)) { + options[nr_options] = strdup(name); + if (!options[nr_options]) + goto close_file_and_continue; + + abs_path[nr_options] = strdup(path); + if (!abs_path[nr_options]) { + free(options[nr_options]); + ui__warning("Can't search all data files due to memory shortage.\n"); + fclose(file); + break; + } + + nr_options++; + } + +close_file_and_continue: + fclose(file); + if (nr_options >= 32) { + ui__warning("Too many perf data files in PWD!\n" + "Only the first 32 files will be listed.\n"); + break; + } + } + closedir(pwd_dir); + + if (nr_options) { + choice = ui__popup_menu(nr_options, options); + if (choice < nr_options && choice >= 0) { + tmp = strdup(abs_path[choice]); + if (tmp) { + if (is_input_name_malloced) + free((void *)input_name); + input_name = tmp; + is_input_name_malloced = true; + ret = 0; + } else + ui__warning("Data switch failed due to memory shortage!\n"); + } + } + + free_popup_options(options, nr_options); + free_popup_options(abs_path, nr_options); + return ret; +} + + +static int perf_evsel__hists_browse(struct perf_evsel *evsel, int nr_events, + const char *helpline, const char *ev_name, + bool left_exits, + struct hist_browser_timer *hbt, + struct perf_session_env *env) +{ + struct hists *hists = &evsel->hists; + struct hist_browser *browser = hist_browser__new(hists); + struct branch_info *bi; + struct pstack *fstack; + char *options[16]; + int nr_options = 0; + int key = -1; + char buf[64]; + char script_opt[64]; + int delay_secs = hbt ? hbt->refresh : 0; + + if (browser == NULL) + return -1; + + fstack = pstack__new(2); + if (fstack == NULL) + goto out; + + ui_helpline__push(helpline); + + memset(options, 0, sizeof(options)); + + while (1) { + const struct thread *thread = NULL; + const struct dso *dso = NULL; + int choice = 0, + annotate = -2, zoom_dso = -2, zoom_thread = -2, + annotate_f = -2, annotate_t = -2, browse_map = -2; + int scripts_comm = -2, scripts_symbol = -2, + scripts_all = -2, switch_data = -2; + + nr_options = 0; + + key = hist_browser__run(browser, ev_name, hbt); + + if (browser->he_selection != NULL) { + thread = hist_browser__selected_thread(browser); + dso = browser->selection->map ? browser->selection->map->dso : NULL; + } + switch (key) { + case K_TAB: + case K_UNTAB: + if (nr_events == 1) + continue; + /* + * Exit the browser, let hists__browser_tree + * go to the next or previous + */ + goto out_free_stack; + case 'a': + if (!browser->has_symbols) { + ui_browser__warning(&browser->b, delay_secs * 2, + "Annotation is only available for symbolic views, " + "include \"sym*\" in --sort to use it."); + continue; + } + + if (browser->selection == NULL || + browser->selection->sym == NULL || + browser->selection->map->dso->annotate_warned) + continue; + goto do_annotate; + case 'P': + hist_browser__dump(browser); + continue; + case 'd': + goto zoom_dso; + case 'V': + browser->show_dso = !browser->show_dso; + continue; + case 't': + goto zoom_thread; + case '/': + if (ui_browser__input_window("Symbol to show", + "Please enter the name of symbol you want to see", + buf, "ENTER: OK, ESC: Cancel", + delay_secs * 2) == K_ENTER) { + hists->symbol_filter_str = *buf ? buf : NULL; + hists__filter_by_symbol(hists); + hist_browser__reset(browser); + } + continue; + case 'r': + if (is_report_browser(hbt)) + goto do_scripts; + continue; + case 's': + if (is_report_browser(hbt)) + goto do_data_switch; + continue; + case K_F1: + case 'h': + case '?': + ui_browser__help_window(&browser->b, + "h/?/F1 Show this window\n" + "UP/DOWN/PGUP\n" + "PGDN/SPACE Navigate\n" + "q/ESC/CTRL+C Exit browser\n\n" + "For multiple event sessions:\n\n" + "TAB/UNTAB Switch events\n\n" + "For symbolic views (--sort has sym):\n\n" + "-> Zoom into DSO/Threads & Annotate current symbol\n" + "<- Zoom out\n" + "a Annotate current symbol\n" + "C Collapse all callchains\n" + "E Expand all callchains\n" + "d Zoom into current DSO\n" + "t Zoom into current Thread\n" + "r Run available scripts('perf report' only)\n" + "s Switch to another data file in PWD ('perf report' only)\n" + "P Print histograms to perf.hist.N\n" + "V Verbose (DSO names in callchains, etc)\n" + "/ Filter symbol by name"); + continue; + case K_ENTER: + case K_RIGHT: + /* menu */ + break; + case K_LEFT: { + const void *top; + + if (pstack__empty(fstack)) { + /* + * Go back to the perf_evsel_menu__run or other user + */ + if (left_exits) + goto out_free_stack; + continue; + } + top = pstack__pop(fstack); + if (top == &browser->hists->dso_filter) + goto zoom_out_dso; + if (top == &browser->hists->thread_filter) + goto zoom_out_thread; + continue; + } + case K_ESC: + if (!left_exits && + !ui_browser__dialog_yesno(&browser->b, + "Do you really want to exit?")) + continue; + /* Fall thru */ + case 'q': + case CTRL('c'): + goto out_free_stack; + default: + continue; + } + + if (!browser->has_symbols) + goto add_exit_option; + + if (sort__branch_mode == 1) { + bi = browser->he_selection->branch_info; + if (browser->selection != NULL && + bi && + bi->from.sym != NULL && + !bi->from.map->dso->annotate_warned && + asprintf(&options[nr_options], "Annotate %s", + bi->from.sym->name) > 0) + annotate_f = nr_options++; + + if (browser->selection != NULL && + bi && + bi->to.sym != NULL && + !bi->to.map->dso->annotate_warned && + (bi->to.sym != bi->from.sym || + bi->to.map->dso != bi->from.map->dso) && + asprintf(&options[nr_options], "Annotate %s", + bi->to.sym->name) > 0) + annotate_t = nr_options++; + } else { + + if (browser->selection != NULL && + browser->selection->sym != NULL && + !browser->selection->map->dso->annotate_warned && + asprintf(&options[nr_options], "Annotate %s", + browser->selection->sym->name) > 0) + annotate = nr_options++; + } + + if (thread != NULL && + asprintf(&options[nr_options], "Zoom %s %s(%d) thread", + (browser->hists->thread_filter ? "out of" : "into"), + (thread->comm_set ? thread->comm : ""), + thread->pid) > 0) + zoom_thread = nr_options++; + + if (dso != NULL && + asprintf(&options[nr_options], "Zoom %s %s DSO", + (browser->hists->dso_filter ? "out of" : "into"), + (dso->kernel ? "the Kernel" : dso->short_name)) > 0) + zoom_dso = nr_options++; + + if (browser->selection != NULL && + browser->selection->map != NULL && + asprintf(&options[nr_options], "Browse map details") > 0) + browse_map = nr_options++; + + /* perf script support */ + if (browser->he_selection) { + struct symbol *sym; + + if (asprintf(&options[nr_options], "Run scripts for samples of thread [%s]", + browser->he_selection->thread->comm) > 0) + scripts_comm = nr_options++; + + sym = browser->he_selection->ms.sym; + if (sym && sym->namelen && + asprintf(&options[nr_options], "Run scripts for samples of symbol [%s]", + sym->name) > 0) + scripts_symbol = nr_options++; + } + + if (asprintf(&options[nr_options], "Run scripts for all samples") > 0) + scripts_all = nr_options++; + + if (is_report_browser(hbt) && asprintf(&options[nr_options], + "Switch to another data file in PWD") > 0) + switch_data = nr_options++; +add_exit_option: + options[nr_options++] = (char *)"Exit"; +retry_popup_menu: + choice = ui__popup_menu(nr_options, options); + + if (choice == nr_options - 1) + break; + + if (choice == -1) { + free_popup_options(options, nr_options - 1); + continue; + } + + if (choice == annotate || choice == annotate_t || choice == annotate_f) { + struct hist_entry *he; + int err; +do_annotate: + if (!objdump_path && perf_session_env__lookup_objdump(env)) + continue; + + he = hist_browser__selected_entry(browser); + if (he == NULL) + continue; + + /* + * we stash the branch_info symbol + map into the + * the ms so we don't have to rewrite all the annotation + * code to use branch_info. + * in branch mode, the ms struct is not used + */ + if (choice == annotate_f) { + he->ms.sym = he->branch_info->from.sym; + he->ms.map = he->branch_info->from.map; + } else if (choice == annotate_t) { + he->ms.sym = he->branch_info->to.sym; + he->ms.map = he->branch_info->to.map; + } + + /* + * Don't let this be freed, say, by hists__decay_entry. + */ + he->used = true; + err = hist_entry__tui_annotate(he, evsel->idx, hbt); + he->used = false; + /* + * offer option to annotate the other branch source or target + * (if they exists) when returning from annotate + */ + if ((err == 'q' || err == CTRL('c')) + && annotate_t != -2 && annotate_f != -2) + goto retry_popup_menu; + + ui_browser__update_nr_entries(&browser->b, browser->hists->nr_entries); + if (err) + ui_browser__handle_resize(&browser->b); + + } else if (choice == browse_map) + map__browse(browser->selection->map); + else if (choice == zoom_dso) { +zoom_dso: + if (browser->hists->dso_filter) { + pstack__remove(fstack, &browser->hists->dso_filter); +zoom_out_dso: + ui_helpline__pop(); + browser->hists->dso_filter = NULL; + sort_dso.elide = false; + } else { + if (dso == NULL) + continue; + ui_helpline__fpush("To zoom out press <- or -> + \"Zoom out of %s DSO\"", + dso->kernel ? "the Kernel" : dso->short_name); + browser->hists->dso_filter = dso; + sort_dso.elide = true; + pstack__push(fstack, &browser->hists->dso_filter); + } + hists__filter_by_dso(hists); + hist_browser__reset(browser); + } else if (choice == zoom_thread) { +zoom_thread: + if (browser->hists->thread_filter) { + pstack__remove(fstack, &browser->hists->thread_filter); +zoom_out_thread: + ui_helpline__pop(); + browser->hists->thread_filter = NULL; + sort_thread.elide = false; + } else { + ui_helpline__fpush("To zoom out press <- or -> + \"Zoom out of %s(%d) thread\"", + thread->comm_set ? thread->comm : "", + thread->pid); + browser->hists->thread_filter = thread; + sort_thread.elide = true; + pstack__push(fstack, &browser->hists->thread_filter); + } + hists__filter_by_thread(hists); + hist_browser__reset(browser); + } + /* perf scripts support */ + else if (choice == scripts_all || choice == scripts_comm || + choice == scripts_symbol) { +do_scripts: + memset(script_opt, 0, 64); + + if (choice == scripts_comm) + sprintf(script_opt, " -c %s ", browser->he_selection->thread->comm); + + if (choice == scripts_symbol) + sprintf(script_opt, " -S %s ", browser->he_selection->ms.sym->name); + + script_browse(script_opt); + } + /* Switch to another data file */ + else if (choice == switch_data) { +do_data_switch: + if (!switch_data_file()) { + key = K_SWITCH_INPUT_DATA; + break; + } else + ui__warning("Won't switch the data files due to\n" + "no valid data file get selected!\n"); + } + } +out_free_stack: + pstack__delete(fstack); +out: + hist_browser__delete(browser); + free_popup_options(options, nr_options - 1); + return key; +} + +struct perf_evsel_menu { + struct ui_browser b; + struct perf_evsel *selection; + bool lost_events, lost_events_warned; + struct perf_session_env *env; +}; + +static void perf_evsel_menu__write(struct ui_browser *browser, + void *entry, int row) +{ + struct perf_evsel_menu *menu = container_of(browser, + struct perf_evsel_menu, b); + struct perf_evsel *evsel = list_entry(entry, struct perf_evsel, node); + bool current_entry = ui_browser__is_current_entry(browser, row); + unsigned long nr_events = evsel->hists.stats.nr_events[PERF_RECORD_SAMPLE]; + const char *ev_name = perf_evsel__name(evsel); + char bf[256], unit; + const char *warn = " "; + size_t printed; + + ui_browser__set_color(browser, current_entry ? HE_COLORSET_SELECTED : + HE_COLORSET_NORMAL); + + if (symbol_conf.event_group && evsel->nr_members > 1) { + struct perf_evsel *pos; + + ev_name = perf_evsel__group_name(evsel); + + for_each_group_member(pos, evsel) { + nr_events += pos->hists.stats.nr_events[PERF_RECORD_SAMPLE]; + } + } + + nr_events = convert_unit(nr_events, &unit); + printed = scnprintf(bf, sizeof(bf), "%lu%c%s%s", nr_events, + unit, unit == ' ' ? "" : " ", ev_name); + slsmg_printf("%s", bf); + + nr_events = evsel->hists.stats.nr_events[PERF_RECORD_LOST]; + if (nr_events != 0) { + menu->lost_events = true; + if (!current_entry) + ui_browser__set_color(browser, HE_COLORSET_TOP); + nr_events = convert_unit(nr_events, &unit); + printed += scnprintf(bf, sizeof(bf), ": %ld%c%schunks LOST!", + nr_events, unit, unit == ' ' ? "" : " "); + warn = bf; + } + + slsmg_write_nstring(warn, browser->width - printed); + + if (current_entry) + menu->selection = evsel; +} + +static int perf_evsel_menu__run(struct perf_evsel_menu *menu, + int nr_events, const char *help, + struct hist_browser_timer *hbt) +{ + struct perf_evlist *evlist = menu->b.priv; + struct perf_evsel *pos; + const char *ev_name, *title = "Available samples"; + int delay_secs = hbt ? hbt->refresh : 0; + int key; + + if (ui_browser__show(&menu->b, title, + "ESC: exit, ENTER|->: Browse histograms") < 0) + return -1; + + while (1) { + key = ui_browser__run(&menu->b, delay_secs); + + switch (key) { + case K_TIMER: + hbt->timer(hbt->arg); + + if (!menu->lost_events_warned && menu->lost_events) { + ui_browser__warn_lost_events(&menu->b); + menu->lost_events_warned = true; + } + continue; + case K_RIGHT: + case K_ENTER: + if (!menu->selection) + continue; + pos = menu->selection; +browse_hists: + perf_evlist__set_selected(evlist, pos); + /* + * Give the calling tool a chance to populate the non + * default evsel resorted hists tree. + */ + if (hbt) + hbt->timer(hbt->arg); + ev_name = perf_evsel__name(pos); + key = perf_evsel__hists_browse(pos, nr_events, help, + ev_name, true, hbt, + menu->env); + ui_browser__show_title(&menu->b, title); + switch (key) { + case K_TAB: + if (pos->node.next == &evlist->entries) + pos = list_entry(evlist->entries.next, struct perf_evsel, node); + else + pos = list_entry(pos->node.next, struct perf_evsel, node); + goto browse_hists; + case K_UNTAB: + if (pos->node.prev == &evlist->entries) + pos = list_entry(evlist->entries.prev, struct perf_evsel, node); + else + pos = list_entry(pos->node.prev, struct perf_evsel, node); + goto browse_hists; + case K_ESC: + if (!ui_browser__dialog_yesno(&menu->b, + "Do you really want to exit?")) + continue; + /* Fall thru */ + case K_SWITCH_INPUT_DATA: + case 'q': + case CTRL('c'): + goto out; + default: + continue; + } + case K_LEFT: + continue; + case K_ESC: + if (!ui_browser__dialog_yesno(&menu->b, + "Do you really want to exit?")) + continue; + /* Fall thru */ + case 'q': + case CTRL('c'): + goto out; + default: + continue; + } + } + +out: + ui_browser__hide(&menu->b); + return key; +} + +static bool filter_group_entries(struct ui_browser *self __maybe_unused, + void *entry) +{ + struct perf_evsel *evsel = list_entry(entry, struct perf_evsel, node); + + if (symbol_conf.event_group && !perf_evsel__is_group_leader(evsel)) + return true; + + return false; +} + +static int __perf_evlist__tui_browse_hists(struct perf_evlist *evlist, + int nr_entries, const char *help, + struct hist_browser_timer *hbt, + struct perf_session_env *env) +{ + struct perf_evsel *pos; + struct perf_evsel_menu menu = { + .b = { + .entries = &evlist->entries, + .refresh = ui_browser__list_head_refresh, + .seek = ui_browser__list_head_seek, + .write = perf_evsel_menu__write, + .filter = filter_group_entries, + .nr_entries = nr_entries, + .priv = evlist, + }, + .env = env, + }; + + ui_helpline__push("Press ESC to exit"); + + list_for_each_entry(pos, &evlist->entries, node) { + const char *ev_name = perf_evsel__name(pos); + size_t line_len = strlen(ev_name) + 7; + + if (menu.b.width < line_len) + menu.b.width = line_len; + } + + return perf_evsel_menu__run(&menu, nr_entries, help, hbt); +} + +int perf_evlist__tui_browse_hists(struct perf_evlist *evlist, const char *help, + struct hist_browser_timer *hbt, + struct perf_session_env *env) +{ + int nr_entries = evlist->nr_entries; + +single_entry: + if (nr_entries == 1) { + struct perf_evsel *first = list_entry(evlist->entries.next, + struct perf_evsel, node); + const char *ev_name = perf_evsel__name(first); + + return perf_evsel__hists_browse(first, nr_entries, help, + ev_name, false, hbt, env); + } + + if (symbol_conf.event_group) { + struct perf_evsel *pos; + + nr_entries = 0; + list_for_each_entry(pos, &evlist->entries, node) + if (perf_evsel__is_group_leader(pos)) + nr_entries++; + + if (nr_entries == 1) + goto single_entry; + } + + return __perf_evlist__tui_browse_hists(evlist, nr_entries, help, + hbt, env); +} diff --git a/tools/perf/ui/browsers/map.c b/tools/perf/ui/browsers/map.c new file mode 100644 index 00000000..98851d55 --- /dev/null +++ b/tools/perf/ui/browsers/map.c @@ -0,0 +1,154 @@ +#include "../libslang.h" +#include <elf.h> +#include <newt.h> +#include <inttypes.h> +#include <sys/ttydefaults.h> +#include <string.h> +#include <linux/bitops.h> +#include "../../util/util.h" +#include "../../util/debug.h" +#include "../../util/symbol.h" +#include "../browser.h" +#include "../helpline.h" +#include "map.h" + +static int ui_entry__read(const char *title, char *bf, size_t size, int width) +{ + struct newtExitStruct es; + newtComponent form, entry; + const char *result; + int err = -1; + + newtCenteredWindow(width, 1, title); + form = newtForm(NULL, NULL, 0); + if (form == NULL) + return -1; + + entry = newtEntry(0, 0, "0x", width, &result, NEWT_FLAG_SCROLL); + if (entry == NULL) + goto out_free_form; + + newtFormAddComponent(form, entry); + newtFormAddHotKey(form, NEWT_KEY_ENTER); + newtFormAddHotKey(form, NEWT_KEY_ESCAPE); + newtFormAddHotKey(form, NEWT_KEY_LEFT); + newtFormAddHotKey(form, CTRL('c')); + newtFormRun(form, &es); + + if (result != NULL) { + strncpy(bf, result, size); + err = 0; + } +out_free_form: + newtPopWindow(); + newtFormDestroy(form); + return err; +} + +struct map_browser { + struct ui_browser b; + struct map *map; + u8 addrlen; +}; + +static void map_browser__write(struct ui_browser *self, void *nd, int row) +{ + struct symbol *sym = rb_entry(nd, struct symbol, rb_node); + struct map_browser *mb = container_of(self, struct map_browser, b); + bool current_entry = ui_browser__is_current_entry(self, row); + int width; + + ui_browser__set_percent_color(self, 0, current_entry); + slsmg_printf("%*" PRIx64 " %*" PRIx64 " %c ", + mb->addrlen, sym->start, mb->addrlen, sym->end, + sym->binding == STB_GLOBAL ? 'g' : + sym->binding == STB_LOCAL ? 'l' : 'w'); + width = self->width - ((mb->addrlen * 2) + 4); + if (width > 0) + slsmg_write_nstring(sym->name, width); +} + +/* FIXME uber-kludgy, see comment on cmd_report... */ +static u32 *symbol__browser_index(struct symbol *self) +{ + return ((void *)self) - sizeof(struct rb_node) - sizeof(u32); +} + +static int map_browser__search(struct map_browser *self) +{ + char target[512]; + struct symbol *sym; + int err = ui_entry__read("Search by name/addr", target, sizeof(target), 40); + + if (err) + return err; + + if (target[0] == '0' && tolower(target[1]) == 'x') { + u64 addr = strtoull(target, NULL, 16); + sym = map__find_symbol(self->map, addr, NULL); + } else + sym = map__find_symbol_by_name(self->map, target, NULL); + + if (sym != NULL) { + u32 *idx = symbol__browser_index(sym); + + self->b.top = &sym->rb_node; + self->b.index = self->b.top_idx = *idx; + } else + ui_helpline__fpush("%s not found!", target); + + return 0; +} + +static int map_browser__run(struct map_browser *self) +{ + int key; + + if (ui_browser__show(&self->b, self->map->dso->long_name, + "Press <- or ESC to exit, %s / to search", + verbose ? "" : "restart with -v to use") < 0) + return -1; + + while (1) { + key = ui_browser__run(&self->b, 0); + + if (verbose && key == '/') + map_browser__search(self); + else + break; + } + + ui_browser__hide(&self->b); + return key; +} + +int map__browse(struct map *self) +{ + struct map_browser mb = { + .b = { + .entries = &self->dso->symbols[self->type], + .refresh = ui_browser__rb_tree_refresh, + .seek = ui_browser__rb_tree_seek, + .write = map_browser__write, + }, + .map = self, + }; + struct rb_node *nd; + char tmp[BITS_PER_LONG / 4]; + u64 maxaddr = 0; + + for (nd = rb_first(mb.b.entries); nd; nd = rb_next(nd)) { + struct symbol *pos = rb_entry(nd, struct symbol, rb_node); + + if (maxaddr < pos->end) + maxaddr = pos->end; + if (verbose) { + u32 *idx = symbol__browser_index(pos); + *idx = mb.b.nr_entries; + } + ++mb.b.nr_entries; + } + + mb.addrlen = snprintf(tmp, sizeof(tmp), "%" PRIx64, maxaddr); + return map_browser__run(&mb); +} diff --git a/tools/perf/ui/browsers/map.h b/tools/perf/ui/browsers/map.h new file mode 100644 index 00000000..df8581a4 --- /dev/null +++ b/tools/perf/ui/browsers/map.h @@ -0,0 +1,6 @@ +#ifndef _PERF_UI_MAP_BROWSER_H_ +#define _PERF_UI_MAP_BROWSER_H_ 1 +struct map; + +int map__browse(struct map *self); +#endif /* _PERF_UI_MAP_BROWSER_H_ */ diff --git a/tools/perf/ui/browsers/scripts.c b/tools/perf/ui/browsers/scripts.c new file mode 100644 index 00000000..cbbd44b0 --- /dev/null +++ b/tools/perf/ui/browsers/scripts.c @@ -0,0 +1,189 @@ +#include <elf.h> +#include <newt.h> +#include <inttypes.h> +#include <sys/ttydefaults.h> +#include <string.h> +#include "../../util/sort.h" +#include "../../util/util.h" +#include "../../util/hist.h" +#include "../../util/debug.h" +#include "../../util/symbol.h" +#include "../browser.h" +#include "../helpline.h" +#include "../libslang.h" + +/* 2048 lines should be enough for a script output */ +#define MAX_LINES 2048 + +/* 160 bytes for one output line */ +#define AVERAGE_LINE_LEN 160 + +struct script_line { + struct list_head node; + char line[AVERAGE_LINE_LEN]; +}; + +struct perf_script_browser { + struct ui_browser b; + struct list_head entries; + const char *script_name; + int nr_lines; +}; + +#define SCRIPT_NAMELEN 128 +#define SCRIPT_MAX_NO 64 +/* + * Usually the full path for a script is: + * /home/username/libexec/perf-core/scripts/python/xxx.py + * /home/username/libexec/perf-core/scripts/perl/xxx.pl + * So 256 should be long enough to contain the full path. + */ +#define SCRIPT_FULLPATH_LEN 256 + +/* + * When success, will copy the full path of the selected script + * into the buffer pointed by script_name, and return 0. + * Return -1 on failure. + */ +static int list_scripts(char *script_name) +{ + char *buf, *names[SCRIPT_MAX_NO], *paths[SCRIPT_MAX_NO]; + int i, num, choice, ret = -1; + + /* Preset the script name to SCRIPT_NAMELEN */ + buf = malloc(SCRIPT_MAX_NO * (SCRIPT_NAMELEN + SCRIPT_FULLPATH_LEN)); + if (!buf) + return ret; + + for (i = 0; i < SCRIPT_MAX_NO; i++) { + names[i] = buf + i * (SCRIPT_NAMELEN + SCRIPT_FULLPATH_LEN); + paths[i] = names[i] + SCRIPT_NAMELEN; + } + + num = find_scripts(names, paths); + if (num > 0) { + choice = ui__popup_menu(num, names); + if (choice < num && choice >= 0) { + strcpy(script_name, paths[choice]); + ret = 0; + } + } + + free(buf); + return ret; +} + +static void script_browser__write(struct ui_browser *browser, + void *entry, int row) +{ + struct script_line *sline = list_entry(entry, struct script_line, node); + bool current_entry = ui_browser__is_current_entry(browser, row); + + ui_browser__set_color(browser, current_entry ? HE_COLORSET_SELECTED : + HE_COLORSET_NORMAL); + + slsmg_write_nstring(sline->line, browser->width); +} + +static int script_browser__run(struct perf_script_browser *self) +{ + int key; + + if (ui_browser__show(&self->b, self->script_name, + "Press <- or ESC to exit") < 0) + return -1; + + while (1) { + key = ui_browser__run(&self->b, 0); + + /* We can add some special key handling here if needed */ + break; + } + + ui_browser__hide(&self->b); + return key; +} + + +int script_browse(const char *script_opt) +{ + char cmd[SCRIPT_FULLPATH_LEN*2], script_name[SCRIPT_FULLPATH_LEN]; + char *line = NULL; + size_t len = 0; + ssize_t retlen; + int ret = -1, nr_entries = 0; + FILE *fp; + void *buf; + struct script_line *sline; + + struct perf_script_browser script = { + .b = { + .refresh = ui_browser__list_head_refresh, + .seek = ui_browser__list_head_seek, + .write = script_browser__write, + }, + .script_name = script_name, + }; + + INIT_LIST_HEAD(&script.entries); + + /* Save each line of the output in one struct script_line object. */ + buf = zalloc((sizeof(*sline)) * MAX_LINES); + if (!buf) + return -1; + sline = buf; + + memset(script_name, 0, SCRIPT_FULLPATH_LEN); + if (list_scripts(script_name)) + goto exit; + + sprintf(cmd, "perf script -s %s ", script_name); + + if (script_opt) + strcat(cmd, script_opt); + + if (input_name) { + strcat(cmd, " -i "); + strcat(cmd, input_name); + } + + strcat(cmd, " 2>&1"); + + fp = popen(cmd, "r"); + if (!fp) + goto exit; + + while ((retlen = getline(&line, &len, fp)) != -1) { + strncpy(sline->line, line, AVERAGE_LINE_LEN); + + /* If one output line is very large, just cut it short */ + if (retlen >= AVERAGE_LINE_LEN) { + sline->line[AVERAGE_LINE_LEN - 1] = '\0'; + sline->line[AVERAGE_LINE_LEN - 2] = '\n'; + } + list_add_tail(&sline->node, &script.entries); + + if (script.b.width < retlen) + script.b.width = retlen; + + if (nr_entries++ >= MAX_LINES - 1) + break; + sline++; + } + + if (script.b.width > AVERAGE_LINE_LEN) + script.b.width = AVERAGE_LINE_LEN; + + if (line) + free(line); + pclose(fp); + + script.nr_lines = nr_entries; + script.b.nr_entries = nr_entries; + script.b.entries = &script.entries; + + ret = script_browser__run(&script); +exit: + free(buf); + return ret; +} diff --git a/tools/perf/ui/gtk/annotate.c b/tools/perf/ui/gtk/annotate.c new file mode 100644 index 00000000..7d8dc581 --- /dev/null +++ b/tools/perf/ui/gtk/annotate.c @@ -0,0 +1,229 @@ +#include "gtk.h" +#include "util/debug.h" +#include "util/annotate.h" +#include "ui/helpline.h" + + +enum { + ANN_COL__PERCENT, + ANN_COL__OFFSET, + ANN_COL__LINE, + + MAX_ANN_COLS +}; + +static const char *const col_names[] = { + "Overhead", + "Offset", + "Line" +}; + +static int perf_gtk__get_percent(char *buf, size_t size, struct symbol *sym, + struct disasm_line *dl, int evidx) +{ + struct sym_hist *symhist; + double percent = 0.0; + const char *markup; + int ret = 0; + + strcpy(buf, ""); + + if (dl->offset == (s64) -1) + return 0; + + symhist = annotation__histogram(symbol__annotation(sym), evidx); + if (!symhist->addr[dl->offset]) + return 0; + + percent = 100.0 * symhist->addr[dl->offset] / symhist->sum; + + markup = perf_gtk__get_percent_color(percent); + if (markup) + ret += scnprintf(buf, size, "%s", markup); + ret += scnprintf(buf + ret, size - ret, "%6.2f%%", percent); + if (markup) + ret += scnprintf(buf + ret, size - ret, "</span>"); + + return ret; +} + +static int perf_gtk__get_offset(char *buf, size_t size, struct symbol *sym, + struct map *map, struct disasm_line *dl) +{ + u64 start = map__rip_2objdump(map, sym->start); + + strcpy(buf, ""); + + if (dl->offset == (s64) -1) + return 0; + + return scnprintf(buf, size, "%"PRIx64, start + dl->offset); +} + +static int perf_gtk__get_line(char *buf, size_t size, struct disasm_line *dl) +{ + int ret = 0; + char *line = g_markup_escape_text(dl->line, -1); + const char *markup = "<span fgcolor='gray'>"; + + strcpy(buf, ""); + + if (!line) + return 0; + + if (dl->offset != (s64) -1) + markup = NULL; + + if (markup) + ret += scnprintf(buf, size, "%s", markup); + ret += scnprintf(buf + ret, size - ret, "%s", line); + if (markup) + ret += scnprintf(buf + ret, size - ret, "</span>"); + + g_free(line); + return ret; +} + +static int perf_gtk__annotate_symbol(GtkWidget *window, struct symbol *sym, + struct map *map, int evidx, + struct hist_browser_timer *hbt __maybe_unused) +{ + struct disasm_line *pos, *n; + struct annotation *notes; + GType col_types[MAX_ANN_COLS]; + GtkCellRenderer *renderer; + GtkListStore *store; + GtkWidget *view; + int i; + char s[512]; + + notes = symbol__annotation(sym); + + for (i = 0; i < MAX_ANN_COLS; i++) { + col_types[i] = G_TYPE_STRING; + } + store = gtk_list_store_newv(MAX_ANN_COLS, col_types); + + view = gtk_tree_view_new(); + renderer = gtk_cell_renderer_text_new(); + + for (i = 0; i < MAX_ANN_COLS; i++) { + gtk_tree_view_insert_column_with_attributes(GTK_TREE_VIEW(view), + -1, col_names[i], renderer, "markup", + i, NULL); + } + + gtk_tree_view_set_model(GTK_TREE_VIEW(view), GTK_TREE_MODEL(store)); + g_object_unref(GTK_TREE_MODEL(store)); + + list_for_each_entry(pos, ¬es->src->source, node) { + GtkTreeIter iter; + + gtk_list_store_append(store, &iter); + + if (perf_gtk__get_percent(s, sizeof(s), sym, pos, evidx)) + gtk_list_store_set(store, &iter, ANN_COL__PERCENT, s, -1); + if (perf_gtk__get_offset(s, sizeof(s), sym, map, pos)) + gtk_list_store_set(store, &iter, ANN_COL__OFFSET, s, -1); + if (perf_gtk__get_line(s, sizeof(s), pos)) + gtk_list_store_set(store, &iter, ANN_COL__LINE, s, -1); + } + + gtk_container_add(GTK_CONTAINER(window), view); + + list_for_each_entry_safe(pos, n, ¬es->src->source, node) { + list_del(&pos->node); + disasm_line__free(pos); + } + + return 0; +} + +int symbol__gtk_annotate(struct symbol *sym, struct map *map, int evidx, + struct hist_browser_timer *hbt) +{ + GtkWidget *window; + GtkWidget *notebook; + GtkWidget *scrolled_window; + GtkWidget *tab_label; + + if (map->dso->annotate_warned) + return -1; + + if (symbol__annotate(sym, map, 0) < 0) { + ui__error("%s", ui_helpline__current); + return -1; + } + + if (perf_gtk__is_active_context(pgctx)) { + window = pgctx->main_window; + notebook = pgctx->notebook; + } else { + GtkWidget *vbox; + GtkWidget *infobar; + GtkWidget *statbar; + + signal(SIGSEGV, perf_gtk__signal); + signal(SIGFPE, perf_gtk__signal); + signal(SIGINT, perf_gtk__signal); + signal(SIGQUIT, perf_gtk__signal); + signal(SIGTERM, perf_gtk__signal); + + window = gtk_window_new(GTK_WINDOW_TOPLEVEL); + gtk_window_set_title(GTK_WINDOW(window), "perf annotate"); + + g_signal_connect(window, "delete_event", gtk_main_quit, NULL); + + pgctx = perf_gtk__activate_context(window); + if (!pgctx) + return -1; + + vbox = gtk_vbox_new(FALSE, 0); + notebook = gtk_notebook_new(); + pgctx->notebook = notebook; + + gtk_box_pack_start(GTK_BOX(vbox), notebook, TRUE, TRUE, 0); + + infobar = perf_gtk__setup_info_bar(); + if (infobar) { + gtk_box_pack_start(GTK_BOX(vbox), infobar, + FALSE, FALSE, 0); + } + + statbar = perf_gtk__setup_statusbar(); + gtk_box_pack_start(GTK_BOX(vbox), statbar, FALSE, FALSE, 0); + + gtk_container_add(GTK_CONTAINER(window), vbox); + } + + scrolled_window = gtk_scrolled_window_new(NULL, NULL); + tab_label = gtk_label_new(sym->name); + + gtk_scrolled_window_set_policy(GTK_SCROLLED_WINDOW(scrolled_window), + GTK_POLICY_AUTOMATIC, + GTK_POLICY_AUTOMATIC); + + gtk_notebook_append_page(GTK_NOTEBOOK(notebook), scrolled_window, + tab_label); + + perf_gtk__annotate_symbol(scrolled_window, sym, map, evidx, hbt); + return 0; +} + +void perf_gtk__show_annotations(void) +{ + GtkWidget *window; + + if (!perf_gtk__is_active_context(pgctx)) + return; + + window = pgctx->main_window; + gtk_widget_show_all(window); + + perf_gtk__resize_window(window); + gtk_window_set_position(GTK_WINDOW(window), GTK_WIN_POS_CENTER); + + gtk_main(); + + perf_gtk__deactivate_context(&pgctx); +} diff --git a/tools/perf/ui/gtk/browser.c b/tools/perf/ui/gtk/browser.c new file mode 100644 index 00000000..c95012cd --- /dev/null +++ b/tools/perf/ui/gtk/browser.c @@ -0,0 +1,87 @@ +#include "../evlist.h" +#include "../cache.h" +#include "../evsel.h" +#include "../sort.h" +#include "../hist.h" +#include "../helpline.h" +#include "gtk.h" + +#include <signal.h> + +void perf_gtk__signal(int sig) +{ + perf_gtk__exit(false); + psignal(sig, "perf"); +} + +void perf_gtk__resize_window(GtkWidget *window) +{ + GdkRectangle rect; + GdkScreen *screen; + int monitor; + int height; + int width; + + screen = gtk_widget_get_screen(window); + + monitor = gdk_screen_get_monitor_at_window(screen, window->window); + + gdk_screen_get_monitor_geometry(screen, monitor, &rect); + + width = rect.width * 3 / 4; + height = rect.height * 3 / 4; + + gtk_window_resize(GTK_WINDOW(window), width, height); +} + +const char *perf_gtk__get_percent_color(double percent) +{ + if (percent >= MIN_RED) + return "<span fgcolor='red'>"; + if (percent >= MIN_GREEN) + return "<span fgcolor='dark green'>"; + return NULL; +} + +#ifdef HAVE_GTK_INFO_BAR +GtkWidget *perf_gtk__setup_info_bar(void) +{ + GtkWidget *info_bar; + GtkWidget *label; + GtkWidget *content_area; + + info_bar = gtk_info_bar_new(); + gtk_widget_set_no_show_all(info_bar, TRUE); + + label = gtk_label_new(""); + gtk_widget_show(label); + + content_area = gtk_info_bar_get_content_area(GTK_INFO_BAR(info_bar)); + gtk_container_add(GTK_CONTAINER(content_area), label); + + gtk_info_bar_add_button(GTK_INFO_BAR(info_bar), GTK_STOCK_OK, + GTK_RESPONSE_OK); + g_signal_connect(info_bar, "response", + G_CALLBACK(gtk_widget_hide), NULL); + + pgctx->info_bar = info_bar; + pgctx->message_label = label; + + return info_bar; +} +#endif + +GtkWidget *perf_gtk__setup_statusbar(void) +{ + GtkWidget *stbar; + unsigned ctxid; + + stbar = gtk_statusbar_new(); + + ctxid = gtk_statusbar_get_context_id(GTK_STATUSBAR(stbar), + "perf report"); + pgctx->statbar = stbar; + pgctx->statbar_ctx_id = ctxid; + + return stbar; +} diff --git a/tools/perf/ui/gtk/gtk.h b/tools/perf/ui/gtk/gtk.h new file mode 100644 index 00000000..3d96785e --- /dev/null +++ b/tools/perf/ui/gtk/gtk.h @@ -0,0 +1,51 @@ +#ifndef _PERF_GTK_H_ +#define _PERF_GTK_H_ 1 + +#include <stdbool.h> + +#pragma GCC diagnostic ignored "-Wstrict-prototypes" +#include <gtk/gtk.h> +#pragma GCC diagnostic error "-Wstrict-prototypes" + + +struct perf_gtk_context { + GtkWidget *main_window; + GtkWidget *notebook; + +#ifdef HAVE_GTK_INFO_BAR + GtkWidget *info_bar; + GtkWidget *message_label; +#endif + GtkWidget *statbar; + guint statbar_ctx_id; +}; + +extern struct perf_gtk_context *pgctx; + +static inline bool perf_gtk__is_active_context(struct perf_gtk_context *ctx) +{ + return ctx && ctx->main_window; +} + +struct perf_gtk_context *perf_gtk__activate_context(GtkWidget *window); +int perf_gtk__deactivate_context(struct perf_gtk_context **ctx); + +void perf_gtk__init_helpline(void); +void perf_gtk__init_progress(void); +void perf_gtk__init_hpp(void); + +void perf_gtk__signal(int sig); +void perf_gtk__resize_window(GtkWidget *window); +const char *perf_gtk__get_percent_color(double percent); +GtkWidget *perf_gtk__setup_statusbar(void); + +#ifdef HAVE_GTK_INFO_BAR +GtkWidget *perf_gtk__setup_info_bar(void); +#else +static inline GtkWidget *perf_gtk__setup_info_bar(void) +{ + return NULL; +} +#endif + +#endif /* _PERF_GTK_H_ */ diff --git a/tools/perf/ui/gtk/helpline.c b/tools/perf/ui/gtk/helpline.c new file mode 100644 index 00000000..3388cbd1 --- /dev/null +++ b/tools/perf/ui/gtk/helpline.c @@ -0,0 +1,57 @@ +#include <stdio.h> +#include <string.h> + +#include "gtk.h" +#include "../ui.h" +#include "../helpline.h" +#include "../../util/debug.h" + +static void gtk_helpline_pop(void) +{ + if (!perf_gtk__is_active_context(pgctx)) + return; + + gtk_statusbar_pop(GTK_STATUSBAR(pgctx->statbar), + pgctx->statbar_ctx_id); +} + +static void gtk_helpline_push(const char *msg) +{ + if (!perf_gtk__is_active_context(pgctx)) + return; + + gtk_statusbar_push(GTK_STATUSBAR(pgctx->statbar), + pgctx->statbar_ctx_id, msg); +} + +static int gtk_helpline_show(const char *fmt, va_list ap) +{ + int ret; + char *ptr; + static int backlog; + + ret = vscnprintf(ui_helpline__current + backlog, + sizeof(ui_helpline__current) - backlog, fmt, ap); + backlog += ret; + + /* only first line can be displayed */ + ptr = strchr(ui_helpline__current, '\n'); + if (ptr && (ptr - ui_helpline__current) <= backlog) { + *ptr = '\0'; + ui_helpline__puts(ui_helpline__current); + backlog = 0; + } + + return ret; +} + +static struct ui_helpline gtk_helpline_fns = { + .pop = gtk_helpline_pop, + .push = gtk_helpline_push, + .show = gtk_helpline_show, +}; + +void perf_gtk__init_helpline(void) +{ + helpline_fns = >k_helpline_fns; +} diff --git a/tools/perf/ui/gtk/hists.c b/tools/perf/ui/gtk/hists.c new file mode 100644 index 00000000..1e764a8a --- /dev/null +++ b/tools/perf/ui/gtk/hists.c @@ -0,0 +1,312 @@ +#include "../evlist.h" +#include "../cache.h" +#include "../evsel.h" +#include "../sort.h" +#include "../hist.h" +#include "../helpline.h" +#include "gtk.h" + +#define MAX_COLUMNS 32 + +static int __percent_color_snprintf(char *buf, size_t size, double percent) +{ + int ret = 0; + const char *markup; + + markup = perf_gtk__get_percent_color(percent); + if (markup) + ret += scnprintf(buf, size, markup); + + ret += scnprintf(buf + ret, size - ret, " %6.2f%%", percent); + + if (markup) + ret += scnprintf(buf + ret, size - ret, "</span>"); + + return ret; +} + + +static int __hpp__color_fmt(struct perf_hpp *hpp, struct hist_entry *he, + u64 (*get_field)(struct hist_entry *)) +{ + int ret; + double percent = 0.0; + struct hists *hists = he->hists; + + if (hists->stats.total_period) + percent = 100.0 * get_field(he) / hists->stats.total_period; + + ret = __percent_color_snprintf(hpp->buf, hpp->size, percent); + + if (symbol_conf.event_group) { + int prev_idx, idx_delta; + struct perf_evsel *evsel = hists_to_evsel(hists); + struct hist_entry *pair; + int nr_members = evsel->nr_members; + + if (nr_members <= 1) + return ret; + + prev_idx = perf_evsel__group_idx(evsel); + + list_for_each_entry(pair, &he->pairs.head, pairs.node) { + u64 period = get_field(pair); + u64 total = pair->hists->stats.total_period; + + evsel = hists_to_evsel(pair->hists); + idx_delta = perf_evsel__group_idx(evsel) - prev_idx - 1; + + while (idx_delta--) { + /* + * zero-fill group members in the middle which + * have no sample + */ + ret += __percent_color_snprintf(hpp->buf + ret, + hpp->size - ret, + 0.0); + } + + percent = 100.0 * period / total; + ret += __percent_color_snprintf(hpp->buf + ret, + hpp->size - ret, + percent); + + prev_idx = perf_evsel__group_idx(evsel); + } + + idx_delta = nr_members - prev_idx - 1; + + while (idx_delta--) { + /* + * zero-fill group members at last which have no sample + */ + ret += __percent_color_snprintf(hpp->buf + ret, + hpp->size - ret, + 0.0); + } + } + return ret; +} + +#define __HPP_COLOR_PERCENT_FN(_type, _field) \ +static u64 he_get_##_field(struct hist_entry *he) \ +{ \ + return he->stat._field; \ +} \ + \ +static int perf_gtk__hpp_color_##_type(struct perf_hpp *hpp, \ + struct hist_entry *he) \ +{ \ + return __hpp__color_fmt(hpp, he, he_get_##_field); \ +} + +__HPP_COLOR_PERCENT_FN(overhead, period) +__HPP_COLOR_PERCENT_FN(overhead_sys, period_sys) +__HPP_COLOR_PERCENT_FN(overhead_us, period_us) +__HPP_COLOR_PERCENT_FN(overhead_guest_sys, period_guest_sys) +__HPP_COLOR_PERCENT_FN(overhead_guest_us, period_guest_us) + +#undef __HPP_COLOR_PERCENT_FN + + +void perf_gtk__init_hpp(void) +{ + perf_hpp__column_enable(PERF_HPP__OVERHEAD); + + perf_hpp__init(); + + perf_hpp__format[PERF_HPP__OVERHEAD].color = + perf_gtk__hpp_color_overhead; + perf_hpp__format[PERF_HPP__OVERHEAD_SYS].color = + perf_gtk__hpp_color_overhead_sys; + perf_hpp__format[PERF_HPP__OVERHEAD_US].color = + perf_gtk__hpp_color_overhead_us; + perf_hpp__format[PERF_HPP__OVERHEAD_GUEST_SYS].color = + perf_gtk__hpp_color_overhead_guest_sys; + perf_hpp__format[PERF_HPP__OVERHEAD_GUEST_US].color = + perf_gtk__hpp_color_overhead_guest_us; +} + +static void perf_gtk__show_hists(GtkWidget *window, struct hists *hists) +{ + struct perf_hpp_fmt *fmt; + GType col_types[MAX_COLUMNS]; + GtkCellRenderer *renderer; + struct sort_entry *se; + GtkListStore *store; + struct rb_node *nd; + GtkWidget *view; + int col_idx; + int nr_cols; + char s[512]; + + struct perf_hpp hpp = { + .buf = s, + .size = sizeof(s), + .ptr = hists_to_evsel(hists), + }; + + nr_cols = 0; + + perf_hpp__for_each_format(fmt) + col_types[nr_cols++] = G_TYPE_STRING; + + list_for_each_entry(se, &hist_entry__sort_list, list) { + if (se->elide) + continue; + + col_types[nr_cols++] = G_TYPE_STRING; + } + + store = gtk_list_store_newv(nr_cols, col_types); + + view = gtk_tree_view_new(); + + renderer = gtk_cell_renderer_text_new(); + + col_idx = 0; + + perf_hpp__for_each_format(fmt) { + fmt->header(&hpp); + + gtk_tree_view_insert_column_with_attributes(GTK_TREE_VIEW(view), + -1, ltrim(s), + renderer, "markup", + col_idx++, NULL); + } + + list_for_each_entry(se, &hist_entry__sort_list, list) { + if (se->elide) + continue; + + gtk_tree_view_insert_column_with_attributes(GTK_TREE_VIEW(view), + -1, se->se_header, + renderer, "text", + col_idx++, NULL); + } + + gtk_tree_view_set_model(GTK_TREE_VIEW(view), GTK_TREE_MODEL(store)); + + g_object_unref(GTK_TREE_MODEL(store)); + + for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + GtkTreeIter iter; + + if (h->filtered) + continue; + + gtk_list_store_append(store, &iter); + + col_idx = 0; + + perf_hpp__for_each_format(fmt) { + if (fmt->color) + fmt->color(&hpp, h); + else + fmt->entry(&hpp, h); + + gtk_list_store_set(store, &iter, col_idx++, s, -1); + } + + list_for_each_entry(se, &hist_entry__sort_list, list) { + if (se->elide) + continue; + + se->se_snprintf(h, s, ARRAY_SIZE(s), + hists__col_len(hists, se->se_width_idx)); + + gtk_list_store_set(store, &iter, col_idx++, s, -1); + } + } + + gtk_container_add(GTK_CONTAINER(window), view); +} + +int perf_evlist__gtk_browse_hists(struct perf_evlist *evlist, + const char *help, + struct hist_browser_timer *hbt __maybe_unused) +{ + struct perf_evsel *pos; + GtkWidget *vbox; + GtkWidget *notebook; + GtkWidget *info_bar; + GtkWidget *statbar; + GtkWidget *window; + + signal(SIGSEGV, perf_gtk__signal); + signal(SIGFPE, perf_gtk__signal); + signal(SIGINT, perf_gtk__signal); + signal(SIGQUIT, perf_gtk__signal); + signal(SIGTERM, perf_gtk__signal); + + window = gtk_window_new(GTK_WINDOW_TOPLEVEL); + + gtk_window_set_title(GTK_WINDOW(window), "perf report"); + + g_signal_connect(window, "delete_event", gtk_main_quit, NULL); + + pgctx = perf_gtk__activate_context(window); + if (!pgctx) + return -1; + + vbox = gtk_vbox_new(FALSE, 0); + + notebook = gtk_notebook_new(); + + gtk_box_pack_start(GTK_BOX(vbox), notebook, TRUE, TRUE, 0); + + info_bar = perf_gtk__setup_info_bar(); + if (info_bar) + gtk_box_pack_start(GTK_BOX(vbox), info_bar, FALSE, FALSE, 0); + + statbar = perf_gtk__setup_statusbar(); + gtk_box_pack_start(GTK_BOX(vbox), statbar, FALSE, FALSE, 0); + + gtk_container_add(GTK_CONTAINER(window), vbox); + + list_for_each_entry(pos, &evlist->entries, node) { + struct hists *hists = &pos->hists; + const char *evname = perf_evsel__name(pos); + GtkWidget *scrolled_window; + GtkWidget *tab_label; + char buf[512]; + size_t size = sizeof(buf); + + if (symbol_conf.event_group) { + if (!perf_evsel__is_group_leader(pos)) + continue; + + if (pos->nr_members > 1) { + perf_evsel__group_desc(pos, buf, size); + evname = buf; + } + } + + scrolled_window = gtk_scrolled_window_new(NULL, NULL); + + gtk_scrolled_window_set_policy(GTK_SCROLLED_WINDOW(scrolled_window), + GTK_POLICY_AUTOMATIC, + GTK_POLICY_AUTOMATIC); + + perf_gtk__show_hists(scrolled_window, hists); + + tab_label = gtk_label_new(evname); + + gtk_notebook_append_page(GTK_NOTEBOOK(notebook), scrolled_window, tab_label); + } + + gtk_widget_show_all(window); + + perf_gtk__resize_window(window); + + gtk_window_set_position(GTK_WINDOW(window), GTK_WIN_POS_CENTER); + + ui_helpline__push(help); + + gtk_main(); + + perf_gtk__deactivate_context(&pgctx); + + return 0; +} diff --git a/tools/perf/ui/gtk/progress.c b/tools/perf/ui/gtk/progress.c new file mode 100644 index 00000000..482bcf3d --- /dev/null +++ b/tools/perf/ui/gtk/progress.c @@ -0,0 +1,59 @@ +#include <inttypes.h> + +#include "gtk.h" +#include "../progress.h" +#include "util.h" + +static GtkWidget *dialog; +static GtkWidget *progress; + +static void gtk_progress_update(u64 curr, u64 total, const char *title) +{ + double fraction = total ? 1.0 * curr / total : 0.0; + char buf[1024]; + + if (dialog == NULL) { + GtkWidget *vbox = gtk_vbox_new(TRUE, 5); + GtkWidget *label = gtk_label_new(title); + + dialog = gtk_window_new(GTK_WINDOW_TOPLEVEL); + progress = gtk_progress_bar_new(); + + gtk_box_pack_start(GTK_BOX(vbox), label, TRUE, FALSE, 3); + gtk_box_pack_start(GTK_BOX(vbox), progress, TRUE, TRUE, 3); + + gtk_container_add(GTK_CONTAINER(dialog), vbox); + + gtk_window_set_title(GTK_WINDOW(dialog), "perf"); + gtk_window_resize(GTK_WINDOW(dialog), 300, 80); + gtk_window_set_position(GTK_WINDOW(dialog), GTK_WIN_POS_CENTER); + + gtk_widget_show_all(dialog); + } + + gtk_progress_bar_set_fraction(GTK_PROGRESS_BAR(progress), fraction); + snprintf(buf, sizeof(buf), "%"PRIu64" / %"PRIu64, curr, total); + gtk_progress_bar_set_text(GTK_PROGRESS_BAR(progress), buf); + + /* we didn't call gtk_main yet, so do it manually */ + while (gtk_events_pending()) + gtk_main_iteration(); +} + +static void gtk_progress_finish(void) +{ + /* this will also destroy all of its children */ + gtk_widget_destroy(dialog); + + dialog = NULL; +} + +static struct ui_progress gtk_progress_fns = { + .update = gtk_progress_update, + .finish = gtk_progress_finish, +}; + +void perf_gtk__init_progress(void) +{ + progress_fns = >k_progress_fns; +} diff --git a/tools/perf/ui/gtk/setup.c b/tools/perf/ui/gtk/setup.c new file mode 100644 index 00000000..6c2dd2e4 --- /dev/null +++ b/tools/perf/ui/gtk/setup.c @@ -0,0 +1,23 @@ +#include "gtk.h" +#include "../../util/cache.h" +#include "../../util/debug.h" + +extern struct perf_error_ops perf_gtk_eops; + +int perf_gtk__init(void) +{ + perf_error__register(&perf_gtk_eops); + perf_gtk__init_helpline(); + perf_gtk__init_progress(); + perf_gtk__init_hpp(); + + return gtk_init_check(NULL, NULL) ? 0 : -1; +} + +void perf_gtk__exit(bool wait_for_ok __maybe_unused) +{ + if (!perf_gtk__is_active_context(pgctx)) + return; + perf_error__unregister(&perf_gtk_eops); + gtk_main_quit(); +} diff --git a/tools/perf/ui/gtk/util.c b/tools/perf/ui/gtk/util.c new file mode 100644 index 00000000..c06942a4 --- /dev/null +++ b/tools/perf/ui/gtk/util.c @@ -0,0 +1,113 @@ +#include "../util.h" +#include "../../util/debug.h" +#include "gtk.h" + +#include <string.h> + + +struct perf_gtk_context *pgctx; + +struct perf_gtk_context *perf_gtk__activate_context(GtkWidget *window) +{ + struct perf_gtk_context *ctx; + + ctx = malloc(sizeof(*pgctx)); + if (ctx) + ctx->main_window = window; + + return ctx; +} + +int perf_gtk__deactivate_context(struct perf_gtk_context **ctx) +{ + if (!perf_gtk__is_active_context(*ctx)) + return -1; + + free(*ctx); + *ctx = NULL; + return 0; +} + +static int perf_gtk__error(const char *format, va_list args) +{ + char *msg; + GtkWidget *dialog; + + if (!perf_gtk__is_active_context(pgctx) || + vasprintf(&msg, format, args) < 0) { + fprintf(stderr, "Error:\n"); + vfprintf(stderr, format, args); + fprintf(stderr, "\n"); + return -1; + } + + dialog = gtk_message_dialog_new_with_markup(GTK_WINDOW(pgctx->main_window), + GTK_DIALOG_DESTROY_WITH_PARENT, + GTK_MESSAGE_ERROR, + GTK_BUTTONS_CLOSE, + "<b>Error</b>\n\n%s", msg); + gtk_dialog_run(GTK_DIALOG(dialog)); + + gtk_widget_destroy(dialog); + free(msg); + return 0; +} + +#ifdef HAVE_GTK_INFO_BAR +static int perf_gtk__warning_info_bar(const char *format, va_list args) +{ + char *msg; + + if (!perf_gtk__is_active_context(pgctx) || + vasprintf(&msg, format, args) < 0) { + fprintf(stderr, "Warning:\n"); + vfprintf(stderr, format, args); + fprintf(stderr, "\n"); + return -1; + } + + gtk_label_set_text(GTK_LABEL(pgctx->message_label), msg); + gtk_info_bar_set_message_type(GTK_INFO_BAR(pgctx->info_bar), + GTK_MESSAGE_WARNING); + gtk_widget_show(pgctx->info_bar); + + free(msg); + return 0; +} +#else +static int perf_gtk__warning_statusbar(const char *format, va_list args) +{ + char *msg, *p; + + if (!perf_gtk__is_active_context(pgctx) || + vasprintf(&msg, format, args) < 0) { + fprintf(stderr, "Warning:\n"); + vfprintf(stderr, format, args); + fprintf(stderr, "\n"); + return -1; + } + + gtk_statusbar_pop(GTK_STATUSBAR(pgctx->statbar), + pgctx->statbar_ctx_id); + + /* Only first line can be displayed */ + p = strchr(msg, '\n'); + if (p) + *p = '\0'; + + gtk_statusbar_push(GTK_STATUSBAR(pgctx->statbar), + pgctx->statbar_ctx_id, msg); + + free(msg); + return 0; +} +#endif + +struct perf_error_ops perf_gtk_eops = { + .error = perf_gtk__error, +#ifdef HAVE_GTK_INFO_BAR + .warning = perf_gtk__warning_info_bar, +#else + .warning = perf_gtk__warning_statusbar, +#endif +}; diff --git a/tools/perf/ui/helpline.c b/tools/perf/ui/helpline.c new file mode 100644 index 00000000..700fb3cf --- /dev/null +++ b/tools/perf/ui/helpline.c @@ -0,0 +1,73 @@ +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include "../debug.h" +#include "helpline.h" +#include "ui.h" + +char ui_helpline__current[512]; + +static void nop_helpline__pop(void) +{ +} + +static void nop_helpline__push(const char *msg __maybe_unused) +{ +} + +static int nop_helpline__show(const char *fmt __maybe_unused, + va_list ap __maybe_unused) +{ + return 0; +} + +static struct ui_helpline default_helpline_fns = { + .pop = nop_helpline__pop, + .push = nop_helpline__push, + .show = nop_helpline__show, +}; + +struct ui_helpline *helpline_fns = &default_helpline_fns; + +void ui_helpline__pop(void) +{ + helpline_fns->pop(); +} + +void ui_helpline__push(const char *msg) +{ + helpline_fns->push(msg); +} + +void ui_helpline__vpush(const char *fmt, va_list ap) +{ + char *s; + + if (vasprintf(&s, fmt, ap) < 0) + vfprintf(stderr, fmt, ap); + else { + ui_helpline__push(s); + free(s); + } +} + +void ui_helpline__fpush(const char *fmt, ...) +{ + va_list ap; + + va_start(ap, fmt); + ui_helpline__vpush(fmt, ap); + va_end(ap); +} + +void ui_helpline__puts(const char *msg) +{ + ui_helpline__pop(); + ui_helpline__push(msg); +} + +int ui_helpline__vshow(const char *fmt, va_list ap) +{ + return helpline_fns->show(fmt, ap); +} diff --git a/tools/perf/ui/helpline.h b/tools/perf/ui/helpline.h new file mode 100644 index 00000000..46181f4f --- /dev/null +++ b/tools/perf/ui/helpline.h @@ -0,0 +1,29 @@ +#ifndef _PERF_UI_HELPLINE_H_ +#define _PERF_UI_HELPLINE_H_ 1 + +#include <stdio.h> +#include <stdarg.h> + +#include "../util/cache.h" + +struct ui_helpline { + void (*pop)(void); + void (*push)(const char *msg); + int (*show)(const char *fmt, va_list ap); +}; + +extern struct ui_helpline *helpline_fns; + +void ui_helpline__init(void); + +void ui_helpline__pop(void); +void ui_helpline__push(const char *msg); +void ui_helpline__vpush(const char *fmt, va_list ap); +void ui_helpline__fpush(const char *fmt, ...); +void ui_helpline__puts(const char *msg); +int ui_helpline__vshow(const char *fmt, va_list ap); + +extern char ui_helpline__current[512]; +extern char ui_helpline__last_msg[]; + +#endif /* _PERF_UI_HELPLINE_H_ */ diff --git a/tools/perf/ui/hist.c b/tools/perf/ui/hist.c new file mode 100644 index 00000000..d671e63a --- /dev/null +++ b/tools/perf/ui/hist.c @@ -0,0 +1,516 @@ +#include <math.h> + +#include "../util/hist.h" +#include "../util/util.h" +#include "../util/sort.h" +#include "../util/evsel.h" + +/* hist period print (hpp) functions */ + +typedef int (*hpp_snprint_fn)(char *buf, size_t size, const char *fmt, ...); + +static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he, + u64 (*get_field)(struct hist_entry *), + const char *fmt, hpp_snprint_fn print_fn, + bool fmt_percent) +{ + int ret; + struct hists *hists = he->hists; + + if (fmt_percent) { + double percent = 0.0; + + if (hists->stats.total_period) + percent = 100.0 * get_field(he) / + hists->stats.total_period; + + ret = print_fn(hpp->buf, hpp->size, fmt, percent); + } else + ret = print_fn(hpp->buf, hpp->size, fmt, get_field(he)); + + if (symbol_conf.event_group) { + int prev_idx, idx_delta; + struct perf_evsel *evsel = hists_to_evsel(hists); + struct hist_entry *pair; + int nr_members = evsel->nr_members; + + if (nr_members <= 1) + return ret; + + prev_idx = perf_evsel__group_idx(evsel); + + list_for_each_entry(pair, &he->pairs.head, pairs.node) { + u64 period = get_field(pair); + u64 total = pair->hists->stats.total_period; + + if (!total) + continue; + + evsel = hists_to_evsel(pair->hists); + idx_delta = perf_evsel__group_idx(evsel) - prev_idx - 1; + + while (idx_delta--) { + /* + * zero-fill group members in the middle which + * have no sample + */ + ret += print_fn(hpp->buf + ret, hpp->size - ret, + fmt, 0); + } + + if (fmt_percent) + ret += print_fn(hpp->buf + ret, hpp->size - ret, + fmt, 100.0 * period / total); + else + ret += print_fn(hpp->buf + ret, hpp->size - ret, + fmt, period); + + prev_idx = perf_evsel__group_idx(evsel); + } + + idx_delta = nr_members - prev_idx - 1; + + while (idx_delta--) { + /* + * zero-fill group members at last which have no sample + */ + ret += print_fn(hpp->buf + ret, hpp->size - ret, + fmt, 0); + } + } + return ret; +} + +#define __HPP_HEADER_FN(_type, _str, _min_width, _unit_width) \ +static int hpp__header_##_type(struct perf_hpp *hpp) \ +{ \ + int len = _min_width; \ + \ + if (symbol_conf.event_group) { \ + struct perf_evsel *evsel = hpp->ptr; \ + \ + len = max(len, evsel->nr_members * _unit_width); \ + } \ + return scnprintf(hpp->buf, hpp->size, "%*s", len, _str); \ +} + +#define __HPP_WIDTH_FN(_type, _min_width, _unit_width) \ +static int hpp__width_##_type(struct perf_hpp *hpp __maybe_unused) \ +{ \ + int len = _min_width; \ + \ + if (symbol_conf.event_group) { \ + struct perf_evsel *evsel = hpp->ptr; \ + \ + len = max(len, evsel->nr_members * _unit_width); \ + } \ + return len; \ +} + +#define __HPP_COLOR_PERCENT_FN(_type, _field) \ +static u64 he_get_##_field(struct hist_entry *he) \ +{ \ + return he->stat._field; \ +} \ + \ +static int hpp__color_##_type(struct perf_hpp *hpp, struct hist_entry *he) \ +{ \ + return __hpp__fmt(hpp, he, he_get_##_field, " %6.2f%%", \ + (hpp_snprint_fn)percent_color_snprintf, true); \ +} + +#define __HPP_ENTRY_PERCENT_FN(_type, _field) \ +static int hpp__entry_##_type(struct perf_hpp *hpp, struct hist_entry *he) \ +{ \ + const char *fmt = symbol_conf.field_sep ? " %.2f" : " %6.2f%%"; \ + return __hpp__fmt(hpp, he, he_get_##_field, fmt, \ + scnprintf, true); \ +} + +#define __HPP_ENTRY_RAW_FN(_type, _field) \ +static u64 he_get_raw_##_field(struct hist_entry *he) \ +{ \ + return he->stat._field; \ +} \ + \ +static int hpp__entry_##_type(struct perf_hpp *hpp, struct hist_entry *he) \ +{ \ + const char *fmt = symbol_conf.field_sep ? " %"PRIu64 : " %11"PRIu64; \ + return __hpp__fmt(hpp, he, he_get_raw_##_field, fmt, scnprintf, false); \ +} + +#define HPP_PERCENT_FNS(_type, _str, _field, _min_width, _unit_width) \ +__HPP_HEADER_FN(_type, _str, _min_width, _unit_width) \ +__HPP_WIDTH_FN(_type, _min_width, _unit_width) \ +__HPP_COLOR_PERCENT_FN(_type, _field) \ +__HPP_ENTRY_PERCENT_FN(_type, _field) + +#define HPP_RAW_FNS(_type, _str, _field, _min_width, _unit_width) \ +__HPP_HEADER_FN(_type, _str, _min_width, _unit_width) \ +__HPP_WIDTH_FN(_type, _min_width, _unit_width) \ +__HPP_ENTRY_RAW_FN(_type, _field) + + +HPP_PERCENT_FNS(overhead, "Overhead", period, 8, 8) +HPP_PERCENT_FNS(overhead_sys, "sys", period_sys, 8, 8) +HPP_PERCENT_FNS(overhead_us, "usr", period_us, 8, 8) +HPP_PERCENT_FNS(overhead_guest_sys, "guest sys", period_guest_sys, 9, 8) +HPP_PERCENT_FNS(overhead_guest_us, "guest usr", period_guest_us, 9, 8) + +HPP_RAW_FNS(samples, "Samples", nr_events, 12, 12) +HPP_RAW_FNS(period, "Period", period, 12, 12) + + +static int hpp__header_baseline(struct perf_hpp *hpp) +{ + return scnprintf(hpp->buf, hpp->size, "Baseline"); +} + +static int hpp__width_baseline(struct perf_hpp *hpp __maybe_unused) +{ + return 8; +} + +static double baseline_percent(struct hist_entry *he) +{ + struct hist_entry *pair = hist_entry__next_pair(he); + struct hists *pair_hists = pair ? pair->hists : NULL; + double percent = 0.0; + + if (pair) { + u64 total_period = pair_hists->stats.total_period; + u64 base_period = pair->stat.period; + + percent = 100.0 * base_period / total_period; + } + + return percent; +} + +static int hpp__color_baseline(struct perf_hpp *hpp, struct hist_entry *he) +{ + double percent = baseline_percent(he); + + if (hist_entry__has_pairs(he) || symbol_conf.field_sep) + return percent_color_snprintf(hpp->buf, hpp->size, " %6.2f%%", percent); + else + return scnprintf(hpp->buf, hpp->size, " "); +} + +static int hpp__entry_baseline(struct perf_hpp *hpp, struct hist_entry *he) +{ + double percent = baseline_percent(he); + const char *fmt = symbol_conf.field_sep ? "%.2f" : " %6.2f%%"; + + if (hist_entry__has_pairs(he) || symbol_conf.field_sep) + return scnprintf(hpp->buf, hpp->size, fmt, percent); + else + return scnprintf(hpp->buf, hpp->size, " "); +} + +static int hpp__header_period_baseline(struct perf_hpp *hpp) +{ + const char *fmt = symbol_conf.field_sep ? "%s" : "%12s"; + + return scnprintf(hpp->buf, hpp->size, fmt, "Period Base"); +} + +static int hpp__width_period_baseline(struct perf_hpp *hpp __maybe_unused) +{ + return 12; +} + +static int hpp__entry_period_baseline(struct perf_hpp *hpp, struct hist_entry *he) +{ + struct hist_entry *pair = hist_entry__next_pair(he); + u64 period = pair ? pair->stat.period : 0; + const char *fmt = symbol_conf.field_sep ? "%" PRIu64 : "%12" PRIu64; + + return scnprintf(hpp->buf, hpp->size, fmt, period); +} + +static int hpp__header_delta(struct perf_hpp *hpp) +{ + const char *fmt = symbol_conf.field_sep ? "%s" : "%7s"; + + return scnprintf(hpp->buf, hpp->size, fmt, "Delta"); +} + +static int hpp__width_delta(struct perf_hpp *hpp __maybe_unused) +{ + return 7; +} + +static int hpp__entry_delta(struct perf_hpp *hpp, struct hist_entry *he) +{ + struct hist_entry *pair = hist_entry__next_pair(he); + const char *fmt = symbol_conf.field_sep ? "%s" : "%7.7s"; + char buf[32] = " "; + double diff = 0.0; + + if (pair) { + if (he->diff.computed) + diff = he->diff.period_ratio_delta; + else + diff = perf_diff__compute_delta(he, pair); + } else + diff = perf_diff__period_percent(he, he->stat.period); + + if (fabs(diff) >= 0.01) + scnprintf(buf, sizeof(buf), "%+4.2F%%", diff); + + return scnprintf(hpp->buf, hpp->size, fmt, buf); +} + +static int hpp__header_ratio(struct perf_hpp *hpp) +{ + const char *fmt = symbol_conf.field_sep ? "%s" : "%14s"; + + return scnprintf(hpp->buf, hpp->size, fmt, "Ratio"); +} + +static int hpp__width_ratio(struct perf_hpp *hpp __maybe_unused) +{ + return 14; +} + +static int hpp__entry_ratio(struct perf_hpp *hpp, struct hist_entry *he) +{ + struct hist_entry *pair = hist_entry__next_pair(he); + const char *fmt = symbol_conf.field_sep ? "%s" : "%14s"; + char buf[32] = " "; + double ratio = 0.0; + + if (pair) { + if (he->diff.computed) + ratio = he->diff.period_ratio; + else + ratio = perf_diff__compute_ratio(he, pair); + } + + if (ratio > 0.0) + scnprintf(buf, sizeof(buf), "%+14.6F", ratio); + + return scnprintf(hpp->buf, hpp->size, fmt, buf); +} + +static int hpp__header_wdiff(struct perf_hpp *hpp) +{ + const char *fmt = symbol_conf.field_sep ? "%s" : "%14s"; + + return scnprintf(hpp->buf, hpp->size, fmt, "Weighted diff"); +} + +static int hpp__width_wdiff(struct perf_hpp *hpp __maybe_unused) +{ + return 14; +} + +static int hpp__entry_wdiff(struct perf_hpp *hpp, struct hist_entry *he) +{ + struct hist_entry *pair = hist_entry__next_pair(he); + const char *fmt = symbol_conf.field_sep ? "%s" : "%14s"; + char buf[32] = " "; + s64 wdiff = 0; + + if (pair) { + if (he->diff.computed) + wdiff = he->diff.wdiff; + else + wdiff = perf_diff__compute_wdiff(he, pair); + } + + if (wdiff != 0) + scnprintf(buf, sizeof(buf), "%14ld", wdiff); + + return scnprintf(hpp->buf, hpp->size, fmt, buf); +} + +static int hpp__header_formula(struct perf_hpp *hpp) +{ + const char *fmt = symbol_conf.field_sep ? "%s" : "%70s"; + + return scnprintf(hpp->buf, hpp->size, fmt, "Formula"); +} + +static int hpp__width_formula(struct perf_hpp *hpp __maybe_unused) +{ + return 70; +} + +static int hpp__entry_formula(struct perf_hpp *hpp, struct hist_entry *he) +{ + struct hist_entry *pair = hist_entry__next_pair(he); + const char *fmt = symbol_conf.field_sep ? "%s" : "%-70s"; + char buf[96] = " "; + + if (pair) + perf_diff__formula(he, pair, buf, sizeof(buf)); + + return scnprintf(hpp->buf, hpp->size, fmt, buf); +} + +#define HPP__COLOR_PRINT_FNS(_name) \ + { \ + .header = hpp__header_ ## _name, \ + .width = hpp__width_ ## _name, \ + .color = hpp__color_ ## _name, \ + .entry = hpp__entry_ ## _name \ + } + +#define HPP__PRINT_FNS(_name) \ + { \ + .header = hpp__header_ ## _name, \ + .width = hpp__width_ ## _name, \ + .entry = hpp__entry_ ## _name \ + } + +struct perf_hpp_fmt perf_hpp__format[] = { + HPP__COLOR_PRINT_FNS(baseline), + HPP__COLOR_PRINT_FNS(overhead), + HPP__COLOR_PRINT_FNS(overhead_sys), + HPP__COLOR_PRINT_FNS(overhead_us), + HPP__COLOR_PRINT_FNS(overhead_guest_sys), + HPP__COLOR_PRINT_FNS(overhead_guest_us), + HPP__PRINT_FNS(samples), + HPP__PRINT_FNS(period), + HPP__PRINT_FNS(period_baseline), + HPP__PRINT_FNS(delta), + HPP__PRINT_FNS(ratio), + HPP__PRINT_FNS(wdiff), + HPP__PRINT_FNS(formula) +}; + +LIST_HEAD(perf_hpp__list); + + +#undef HPP__COLOR_PRINT_FNS +#undef HPP__PRINT_FNS + +#undef HPP_PERCENT_FNS +#undef HPP_RAW_FNS + +#undef __HPP_HEADER_FN +#undef __HPP_WIDTH_FN +#undef __HPP_COLOR_PERCENT_FN +#undef __HPP_ENTRY_PERCENT_FN +#undef __HPP_ENTRY_RAW_FN + + +void perf_hpp__init(void) +{ + if (symbol_conf.show_cpu_utilization) { + perf_hpp__column_enable(PERF_HPP__OVERHEAD_SYS); + perf_hpp__column_enable(PERF_HPP__OVERHEAD_US); + + if (perf_guest) { + perf_hpp__column_enable(PERF_HPP__OVERHEAD_GUEST_SYS); + perf_hpp__column_enable(PERF_HPP__OVERHEAD_GUEST_US); + } + } + + if (symbol_conf.show_nr_samples) + perf_hpp__column_enable(PERF_HPP__SAMPLES); + + if (symbol_conf.show_total_period) + perf_hpp__column_enable(PERF_HPP__PERIOD); +} + +void perf_hpp__column_register(struct perf_hpp_fmt *format) +{ + list_add_tail(&format->list, &perf_hpp__list); +} + +void perf_hpp__column_enable(unsigned col) +{ + BUG_ON(col >= PERF_HPP__MAX_INDEX); + perf_hpp__column_register(&perf_hpp__format[col]); +} + +static inline void advance_hpp(struct perf_hpp *hpp, int inc) +{ + hpp->buf += inc; + hpp->size -= inc; +} + +int hist_entry__period_snprintf(struct perf_hpp *hpp, struct hist_entry *he, + bool color) +{ + const char *sep = symbol_conf.field_sep; + struct perf_hpp_fmt *fmt; + char *start = hpp->buf; + int ret; + bool first = true; + + if (symbol_conf.exclude_other && !he->parent) + return 0; + + perf_hpp__for_each_format(fmt) { + /* + * If there's no field_sep, we still need + * to display initial ' '. + */ + if (!sep || !first) { + ret = scnprintf(hpp->buf, hpp->size, "%s", sep ?: " "); + advance_hpp(hpp, ret); + } else + first = false; + + if (color && fmt->color) + ret = fmt->color(hpp, he); + else + ret = fmt->entry(hpp, he); + + advance_hpp(hpp, ret); + } + + return hpp->buf - start; +} + +int hist_entry__sort_snprintf(struct hist_entry *he, char *s, size_t size, + struct hists *hists) +{ + const char *sep = symbol_conf.field_sep; + struct sort_entry *se; + int ret = 0; + + list_for_each_entry(se, &hist_entry__sort_list, list) { + if (se->elide) + continue; + + ret += scnprintf(s + ret, size - ret, "%s", sep ?: " "); + ret += se->se_snprintf(he, s + ret, size - ret, + hists__col_len(hists, se->se_width_idx)); + } + + return ret; +} + +/* + * See hists__fprintf to match the column widths + */ +unsigned int hists__sort_list_width(struct hists *hists) +{ + struct perf_hpp_fmt *fmt; + struct sort_entry *se; + int i = 0, ret = 0; + struct perf_hpp dummy_hpp = { + .ptr = hists_to_evsel(hists), + }; + + perf_hpp__for_each_format(fmt) { + if (i) + ret += 2; + + ret += fmt->width(&dummy_hpp); + } + + list_for_each_entry(se, &hist_entry__sort_list, list) + if (!se->elide) + ret += 2 + hists__col_len(hists, se->se_width_idx); + + if (verbose) /* Addr + origin */ + ret += 3 + BITS_PER_LONG / 4; + + return ret; +} diff --git a/tools/perf/ui/keysyms.h b/tools/perf/ui/keysyms.h new file mode 100644 index 00000000..65092d57 --- /dev/null +++ b/tools/perf/ui/keysyms.h @@ -0,0 +1,28 @@ +#ifndef _PERF_KEYSYMS_H_ +#define _PERF_KEYSYMS_H_ 1 + +#include "libslang.h" + +#define K_DOWN SL_KEY_DOWN +#define K_END SL_KEY_END +#define K_ENTER '\r' +#define K_ESC 033 +#define K_F1 SL_KEY_F(1) +#define K_HOME SL_KEY_HOME +#define K_LEFT SL_KEY_LEFT +#define K_PGDN SL_KEY_NPAGE +#define K_PGUP SL_KEY_PPAGE +#define K_RIGHT SL_KEY_RIGHT +#define K_TAB '\t' +#define K_UNTAB SL_KEY_UNTAB +#define K_UP SL_KEY_UP +#define K_BKSPC 0x7f +#define K_DEL SL_KEY_DELETE + +/* Not really keys */ +#define K_TIMER -1 +#define K_ERROR -2 +#define K_RESIZE -3 +#define K_SWITCH_INPUT_DATA -4 + +#endif /* _PERF_KEYSYMS_H_ */ diff --git a/tools/perf/ui/libslang.h b/tools/perf/ui/libslang.h new file mode 100644 index 00000000..4d54b645 --- /dev/null +++ b/tools/perf/ui/libslang.h @@ -0,0 +1,29 @@ +#ifndef _PERF_UI_SLANG_H_ +#define _PERF_UI_SLANG_H_ 1 +/* + * slang versions <= 2.0.6 have a "#if HAVE_LONG_LONG" that breaks + * the build if it isn't defined. Use the equivalent one that glibc + * has on features.h. + */ +#include <features.h> +#ifndef HAVE_LONG_LONG +#define HAVE_LONG_LONG __GLIBC_HAVE_LONG_LONG +#endif +#include <slang.h> + +#if SLANG_VERSION < 20104 +#define slsmg_printf(msg, args...) \ + SLsmg_printf((char *)(msg), ##args) +#define slsmg_write_nstring(msg, len) \ + SLsmg_write_nstring((char *)(msg), len) +#define sltt_set_color(obj, name, fg, bg) \ + SLtt_set_color(obj,(char *)(name), (char *)(fg), (char *)(bg)) +#else +#define slsmg_printf SLsmg_printf +#define slsmg_write_nstring SLsmg_write_nstring +#define sltt_set_color SLtt_set_color +#endif + +#define SL_KEY_UNTAB 0x1000 + +#endif /* _PERF_UI_SLANG_H_ */ diff --git a/tools/perf/ui/progress.c b/tools/perf/ui/progress.c new file mode 100644 index 00000000..3ec69560 --- /dev/null +++ b/tools/perf/ui/progress.c @@ -0,0 +1,26 @@ +#include "../cache.h" +#include "progress.h" + +static void nop_progress_update(u64 curr __maybe_unused, + u64 total __maybe_unused, + const char *title __maybe_unused) +{ +} + +static struct ui_progress default_progress_fns = +{ + .update = nop_progress_update, +}; + +struct ui_progress *progress_fns = &default_progress_fns; + +void ui_progress__update(u64 curr, u64 total, const char *title) +{ + return progress_fns->update(curr, total, title); +} + +void ui_progress__finish(void) +{ + if (progress_fns->finish) + progress_fns->finish(); +} diff --git a/tools/perf/ui/progress.h b/tools/perf/ui/progress.h new file mode 100644 index 00000000..257cc224 --- /dev/null +++ b/tools/perf/ui/progress.h @@ -0,0 +1,18 @@ +#ifndef _PERF_UI_PROGRESS_H_ +#define _PERF_UI_PROGRESS_H_ 1 + +#include <../types.h> + +struct ui_progress { + void (*update)(u64, u64, const char *); + void (*finish)(void); +}; + +extern struct ui_progress *progress_fns; + +void ui_progress__init(void); + +void ui_progress__update(u64 curr, u64 total, const char *title); +void ui_progress__finish(void); + +#endif diff --git a/tools/perf/ui/setup.c b/tools/perf/ui/setup.c new file mode 100644 index 00000000..ae6a789c --- /dev/null +++ b/tools/perf/ui/setup.c @@ -0,0 +1,53 @@ +#include <pthread.h> + +#include "../util/cache.h" +#include "../util/debug.h" +#include "../util/hist.h" + +pthread_mutex_t ui__lock = PTHREAD_MUTEX_INITIALIZER; + +void setup_browser(bool fallback_to_pager) +{ + if (use_browser < 2 && (!isatty(1) || dump_trace)) + use_browser = 0; + + /* default to TUI */ + if (use_browser < 0) + use_browser = 1; + + switch (use_browser) { + case 2: + if (perf_gtk__init() == 0) + break; + /* fall through */ + case 1: + use_browser = 1; + if (ui__init() == 0) + break; + /* fall through */ + default: + use_browser = 0; + if (fallback_to_pager) + setup_pager(); + + perf_hpp__column_enable(PERF_HPP__OVERHEAD); + perf_hpp__init(); + break; + } +} + +void exit_browser(bool wait_for_ok) +{ + switch (use_browser) { + case 2: + perf_gtk__exit(wait_for_ok); + break; + + case 1: + ui__exit(wait_for_ok); + break; + + default: + break; + } +} diff --git a/tools/perf/ui/stdio/hist.c b/tools/perf/ui/stdio/hist.c new file mode 100644 index 00000000..ff1f60cf --- /dev/null +++ b/tools/perf/ui/stdio/hist.c @@ -0,0 +1,484 @@ +#include <stdio.h> + +#include "../../util/util.h" +#include "../../util/hist.h" +#include "../../util/sort.h" +#include "../../util/evsel.h" + + +static size_t callchain__fprintf_left_margin(FILE *fp, int left_margin) +{ + int i; + int ret = fprintf(fp, " "); + + for (i = 0; i < left_margin; i++) + ret += fprintf(fp, " "); + + return ret; +} + +static size_t ipchain__fprintf_graph_line(FILE *fp, int depth, int depth_mask, + int left_margin) +{ + int i; + size_t ret = callchain__fprintf_left_margin(fp, left_margin); + + for (i = 0; i < depth; i++) + if (depth_mask & (1 << i)) + ret += fprintf(fp, "| "); + else + ret += fprintf(fp, " "); + + ret += fprintf(fp, "\n"); + + return ret; +} + +static size_t ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain, + int depth, int depth_mask, int period, + u64 total_samples, u64 hits, + int left_margin) +{ + int i; + size_t ret = 0; + + ret += callchain__fprintf_left_margin(fp, left_margin); + for (i = 0; i < depth; i++) { + if (depth_mask & (1 << i)) + ret += fprintf(fp, "|"); + else + ret += fprintf(fp, " "); + if (!period && i == depth - 1) { + double percent; + + percent = hits * 100.0 / total_samples; + ret += percent_color_fprintf(fp, "--%2.2f%%-- ", percent); + } else + ret += fprintf(fp, "%s", " "); + } + if (chain->ms.sym) + ret += fprintf(fp, "%s\n", chain->ms.sym->name); + else + ret += fprintf(fp, "0x%0" PRIx64 "\n", chain->ip); + + return ret; +} + +static struct symbol *rem_sq_bracket; +static struct callchain_list rem_hits; + +static void init_rem_hits(void) +{ + rem_sq_bracket = malloc(sizeof(*rem_sq_bracket) + 6); + if (!rem_sq_bracket) { + fprintf(stderr, "Not enough memory to display remaining hits\n"); + return; + } + + strcpy(rem_sq_bracket->name, "[...]"); + rem_hits.ms.sym = rem_sq_bracket; +} + +static size_t __callchain__fprintf_graph(FILE *fp, struct rb_root *root, + u64 total_samples, int depth, + int depth_mask, int left_margin) +{ + struct rb_node *node, *next; + struct callchain_node *child; + struct callchain_list *chain; + int new_depth_mask = depth_mask; + u64 remaining; + size_t ret = 0; + int i; + uint entries_printed = 0; + + remaining = total_samples; + + node = rb_first(root); + while (node) { + u64 new_total; + u64 cumul; + + child = rb_entry(node, struct callchain_node, rb_node); + cumul = callchain_cumul_hits(child); + remaining -= cumul; + + /* + * The depth mask manages the output of pipes that show + * the depth. We don't want to keep the pipes of the current + * level for the last child of this depth. + * Except if we have remaining filtered hits. They will + * supersede the last child + */ + next = rb_next(node); + if (!next && (callchain_param.mode != CHAIN_GRAPH_REL || !remaining)) + new_depth_mask &= ~(1 << (depth - 1)); + + /* + * But we keep the older depth mask for the line separator + * to keep the level link until we reach the last child + */ + ret += ipchain__fprintf_graph_line(fp, depth, depth_mask, + left_margin); + i = 0; + list_for_each_entry(chain, &child->val, list) { + ret += ipchain__fprintf_graph(fp, chain, depth, + new_depth_mask, i++, + total_samples, + cumul, + left_margin); + } + + if (callchain_param.mode == CHAIN_GRAPH_REL) + new_total = child->children_hit; + else + new_total = total_samples; + + ret += __callchain__fprintf_graph(fp, &child->rb_root, new_total, + depth + 1, + new_depth_mask | (1 << depth), + left_margin); + node = next; + if (++entries_printed == callchain_param.print_limit) + break; + } + + if (callchain_param.mode == CHAIN_GRAPH_REL && + remaining && remaining != total_samples) { + + if (!rem_sq_bracket) + return ret; + + new_depth_mask &= ~(1 << (depth - 1)); + ret += ipchain__fprintf_graph(fp, &rem_hits, depth, + new_depth_mask, 0, total_samples, + remaining, left_margin); + } + + return ret; +} + +static size_t callchain__fprintf_graph(FILE *fp, struct rb_root *root, + u64 total_samples, int left_margin) +{ + struct callchain_node *cnode; + struct callchain_list *chain; + u32 entries_printed = 0; + bool printed = false; + struct rb_node *node; + int i = 0; + int ret = 0; + + /* + * If have one single callchain root, don't bother printing + * its percentage (100 % in fractal mode and the same percentage + * than the hist in graph mode). This also avoid one level of column. + */ + node = rb_first(root); + if (node && !rb_next(node)) { + cnode = rb_entry(node, struct callchain_node, rb_node); + list_for_each_entry(chain, &cnode->val, list) { + /* + * If we sort by symbol, the first entry is the same than + * the symbol. No need to print it otherwise it appears as + * displayed twice. + */ + if (!i++ && sort__first_dimension == SORT_SYM) + continue; + if (!printed) { + ret += callchain__fprintf_left_margin(fp, left_margin); + ret += fprintf(fp, "|\n"); + ret += callchain__fprintf_left_margin(fp, left_margin); + ret += fprintf(fp, "---"); + left_margin += 3; + printed = true; + } else + ret += callchain__fprintf_left_margin(fp, left_margin); + + if (chain->ms.sym) + ret += fprintf(fp, " %s\n", chain->ms.sym->name); + else + ret += fprintf(fp, " %p\n", (void *)(long)chain->ip); + + if (++entries_printed == callchain_param.print_limit) + break; + } + root = &cnode->rb_root; + } + + ret += __callchain__fprintf_graph(fp, root, total_samples, + 1, 1, left_margin); + ret += fprintf(fp, "\n"); + + return ret; +} + +static size_t __callchain__fprintf_flat(FILE *fp, + struct callchain_node *self, + u64 total_samples) +{ + struct callchain_list *chain; + size_t ret = 0; + + if (!self) + return 0; + + ret += __callchain__fprintf_flat(fp, self->parent, total_samples); + + + list_for_each_entry(chain, &self->val, list) { + if (chain->ip >= PERF_CONTEXT_MAX) + continue; + if (chain->ms.sym) + ret += fprintf(fp, " %s\n", chain->ms.sym->name); + else + ret += fprintf(fp, " %p\n", + (void *)(long)chain->ip); + } + + return ret; +} + +static size_t callchain__fprintf_flat(FILE *fp, struct rb_root *self, + u64 total_samples) +{ + size_t ret = 0; + u32 entries_printed = 0; + struct rb_node *rb_node; + struct callchain_node *chain; + + rb_node = rb_first(self); + while (rb_node) { + double percent; + + chain = rb_entry(rb_node, struct callchain_node, rb_node); + percent = chain->hit * 100.0 / total_samples; + + ret = percent_color_fprintf(fp, " %6.2f%%\n", percent); + ret += __callchain__fprintf_flat(fp, chain, total_samples); + ret += fprintf(fp, "\n"); + if (++entries_printed == callchain_param.print_limit) + break; + + rb_node = rb_next(rb_node); + } + + return ret; +} + +static size_t hist_entry_callchain__fprintf(struct hist_entry *he, + u64 total_samples, int left_margin, + FILE *fp) +{ + switch (callchain_param.mode) { + case CHAIN_GRAPH_REL: + return callchain__fprintf_graph(fp, &he->sorted_chain, he->stat.period, + left_margin); + break; + case CHAIN_GRAPH_ABS: + return callchain__fprintf_graph(fp, &he->sorted_chain, total_samples, + left_margin); + break; + case CHAIN_FLAT: + return callchain__fprintf_flat(fp, &he->sorted_chain, total_samples); + break; + case CHAIN_NONE: + break; + default: + pr_err("Bad callchain mode\n"); + } + + return 0; +} + +static size_t hist_entry__callchain_fprintf(struct hist_entry *he, + struct hists *hists, + FILE *fp) +{ + int left_margin = 0; + u64 total_period = hists->stats.total_period; + + if (sort__first_dimension == SORT_COMM) { + struct sort_entry *se = list_first_entry(&hist_entry__sort_list, + typeof(*se), list); + left_margin = hists__col_len(hists, se->se_width_idx); + left_margin -= thread__comm_len(he->thread); + } + + return hist_entry_callchain__fprintf(he, total_period, left_margin, fp); +} + +static int hist_entry__fprintf(struct hist_entry *he, size_t size, + struct hists *hists, FILE *fp) +{ + char bf[512]; + int ret; + struct perf_hpp hpp = { + .buf = bf, + .size = size, + }; + bool color = !symbol_conf.field_sep; + + if (size == 0 || size > sizeof(bf)) + size = hpp.size = sizeof(bf); + + ret = hist_entry__period_snprintf(&hpp, he, color); + hist_entry__sort_snprintf(he, bf + ret, size - ret, hists); + + ret = fprintf(fp, "%s\n", bf); + + if (symbol_conf.use_callchain) + ret += hist_entry__callchain_fprintf(he, hists, fp); + + return ret; +} + +size_t hists__fprintf(struct hists *hists, bool show_header, int max_rows, + int max_cols, FILE *fp) +{ + struct perf_hpp_fmt *fmt; + struct sort_entry *se; + struct rb_node *nd; + size_t ret = 0; + unsigned int width; + const char *sep = symbol_conf.field_sep; + const char *col_width = symbol_conf.col_width_list_str; + int nr_rows = 0; + char bf[96]; + struct perf_hpp dummy_hpp = { + .buf = bf, + .size = sizeof(bf), + .ptr = hists_to_evsel(hists), + }; + bool first = true; + + init_rem_hits(); + + if (!show_header) + goto print_entries; + + fprintf(fp, "# "); + + perf_hpp__for_each_format(fmt) { + if (!first) + fprintf(fp, "%s", sep ?: " "); + else + first = false; + + fmt->header(&dummy_hpp); + fprintf(fp, "%s", bf); + } + + list_for_each_entry(se, &hist_entry__sort_list, list) { + if (se->elide) + continue; + if (sep) { + fprintf(fp, "%c%s", *sep, se->se_header); + continue; + } + width = strlen(se->se_header); + if (symbol_conf.col_width_list_str) { + if (col_width) { + hists__set_col_len(hists, se->se_width_idx, + atoi(col_width)); + col_width = strchr(col_width, ','); + if (col_width) + ++col_width; + } + } + if (!hists__new_col_len(hists, se->se_width_idx, width)) + width = hists__col_len(hists, se->se_width_idx); + fprintf(fp, " %*s", width, se->se_header); + } + + fprintf(fp, "\n"); + if (max_rows && ++nr_rows >= max_rows) + goto out; + + if (sep) + goto print_entries; + + first = true; + + fprintf(fp, "# "); + + perf_hpp__for_each_format(fmt) { + unsigned int i; + + if (!first) + fprintf(fp, "%s", sep ?: " "); + else + first = false; + + width = fmt->width(&dummy_hpp); + for (i = 0; i < width; i++) + fprintf(fp, "."); + } + + list_for_each_entry(se, &hist_entry__sort_list, list) { + unsigned int i; + + if (se->elide) + continue; + + fprintf(fp, " "); + width = hists__col_len(hists, se->se_width_idx); + if (width == 0) + width = strlen(se->se_header); + for (i = 0; i < width; i++) + fprintf(fp, "."); + } + + fprintf(fp, "\n"); + if (max_rows && ++nr_rows >= max_rows) + goto out; + + fprintf(fp, "#\n"); + if (max_rows && ++nr_rows >= max_rows) + goto out; + +print_entries: + for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + + if (h->filtered) + continue; + + ret += hist_entry__fprintf(h, max_cols, hists, fp); + + if (max_rows && ++nr_rows >= max_rows) + goto out; + + if (h->ms.map == NULL && verbose > 1) { + __map_groups__fprintf_maps(&h->thread->mg, + MAP__FUNCTION, verbose, fp); + fprintf(fp, "%.10s end\n", graph_dotted_line); + } + } +out: + free(rem_sq_bracket); + + return ret; +} + +size_t events_stats__fprintf(struct events_stats *stats, FILE *fp) +{ + int i; + size_t ret = 0; + + for (i = 0; i < PERF_RECORD_HEADER_MAX; ++i) { + const char *name; + + if (stats->nr_events[i] == 0) + continue; + + name = perf_event__name(i); + if (!strcmp(name, "UNKNOWN")) + continue; + + ret += fprintf(fp, "%16s events: %10d\n", name, + stats->nr_events[i]); + } + + return ret; +} diff --git a/tools/perf/ui/tui/helpline.c b/tools/perf/ui/tui/helpline.c new file mode 100644 index 00000000..1c8b9afd --- /dev/null +++ b/tools/perf/ui/tui/helpline.c @@ -0,0 +1,58 @@ +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <pthread.h> + +#include "../../util/debug.h" +#include "../helpline.h" +#include "../ui.h" +#include "../libslang.h" + +char ui_helpline__last_msg[1024]; + +static void tui_helpline__pop(void) +{ +} + +static void tui_helpline__push(const char *msg) +{ + const size_t sz = sizeof(ui_helpline__current); + + SLsmg_gotorc(SLtt_Screen_Rows - 1, 0); + SLsmg_set_color(0); + SLsmg_write_nstring((char *)msg, SLtt_Screen_Cols); + SLsmg_refresh(); + strncpy(ui_helpline__current, msg, sz)[sz - 1] = '\0'; +} + +static int tui_helpline__show(const char *format, va_list ap) +{ + int ret; + static int backlog; + + pthread_mutex_lock(&ui__lock); + ret = vscnprintf(ui_helpline__last_msg + backlog, + sizeof(ui_helpline__last_msg) - backlog, format, ap); + backlog += ret; + + if (ui_helpline__last_msg[backlog - 1] == '\n') { + ui_helpline__puts(ui_helpline__last_msg); + SLsmg_refresh(); + backlog = 0; + } + pthread_mutex_unlock(&ui__lock); + + return ret; +} + +struct ui_helpline tui_helpline_fns = { + .pop = tui_helpline__pop, + .push = tui_helpline__push, + .show = tui_helpline__show, +}; + +void ui_helpline__init(void) +{ + helpline_fns = &tui_helpline_fns; + ui_helpline__puts(" "); +} diff --git a/tools/perf/ui/tui/progress.c b/tools/perf/ui/tui/progress.c new file mode 100644 index 00000000..6c2184d5 --- /dev/null +++ b/tools/perf/ui/tui/progress.c @@ -0,0 +1,42 @@ +#include "../cache.h" +#include "../progress.h" +#include "../libslang.h" +#include "../ui.h" +#include "../browser.h" + +static void tui_progress__update(u64 curr, u64 total, const char *title) +{ + int bar, y; + /* + * FIXME: We should have a per UI backend way of showing progress, + * stdio will just show a percentage as NN%, etc. + */ + if (use_browser <= 0) + return; + + if (total == 0) + return; + + ui__refresh_dimensions(true); + pthread_mutex_lock(&ui__lock); + y = SLtt_Screen_Rows / 2 - 2; + SLsmg_set_color(0); + SLsmg_draw_box(y, 0, 3, SLtt_Screen_Cols); + SLsmg_gotorc(y++, 1); + SLsmg_write_string((char *)title); + SLsmg_set_color(HE_COLORSET_SELECTED); + bar = ((SLtt_Screen_Cols - 2) * curr) / total; + SLsmg_fill_region(y, 1, 1, bar, ' '); + SLsmg_refresh(); + pthread_mutex_unlock(&ui__lock); +} + +static struct ui_progress tui_progress_fns = +{ + .update = tui_progress__update, +}; + +void ui_progress__init(void) +{ + progress_fns = &tui_progress_fns; +} diff --git a/tools/perf/ui/tui/setup.c b/tools/perf/ui/tui/setup.c new file mode 100644 index 00000000..81efa192 --- /dev/null +++ b/tools/perf/ui/tui/setup.c @@ -0,0 +1,149 @@ +#include <newt.h> +#include <signal.h> +#include <stdbool.h> + +#include "../../util/cache.h" +#include "../../util/debug.h" +#include "../browser.h" +#include "../helpline.h" +#include "../ui.h" +#include "../util.h" +#include "../libslang.h" +#include "../keysyms.h" + +static volatile int ui__need_resize; + +extern struct perf_error_ops perf_tui_eops; + +extern void hist_browser__init_hpp(void); + +void ui__refresh_dimensions(bool force) +{ + if (force || ui__need_resize) { + ui__need_resize = 0; + pthread_mutex_lock(&ui__lock); + SLtt_get_screen_size(); + SLsmg_reinit_smg(); + pthread_mutex_unlock(&ui__lock); + } +} + +static void ui__sigwinch(int sig __maybe_unused) +{ + ui__need_resize = 1; +} + +static void ui__setup_sigwinch(void) +{ + static bool done; + + if (done) + return; + + done = true; + pthread__unblock_sigwinch(); + signal(SIGWINCH, ui__sigwinch); +} + +int ui__getch(int delay_secs) +{ + struct timeval timeout, *ptimeout = delay_secs ? &timeout : NULL; + fd_set read_set; + int err, key; + + ui__setup_sigwinch(); + + FD_ZERO(&read_set); + FD_SET(0, &read_set); + + if (delay_secs) { + timeout.tv_sec = delay_secs; + timeout.tv_usec = 0; + } + + err = select(1, &read_set, NULL, NULL, ptimeout); + + if (err == 0) + return K_TIMER; + + if (err == -1) { + if (errno == EINTR) + return K_RESIZE; + return K_ERROR; + } + + key = SLang_getkey(); + if (key != K_ESC) + return key; + + FD_ZERO(&read_set); + FD_SET(0, &read_set); + timeout.tv_sec = 0; + timeout.tv_usec = 20; + err = select(1, &read_set, NULL, NULL, &timeout); + if (err == 0) + return K_ESC; + + SLang_ungetkey(key); + return SLkp_getkey(); +} + +static void newt_suspend(void *d __maybe_unused) +{ + newtSuspend(); + raise(SIGTSTP); + newtResume(); +} + +static void ui__signal(int sig) +{ + ui__exit(false); + psignal(sig, "perf"); + exit(0); +} + +int ui__init(void) +{ + int err; + + newtInit(); + err = SLkp_init(); + if (err < 0) { + pr_err("TUI initialization failed.\n"); + goto out; + } + + SLkp_define_keysym((char *)"^(kB)", SL_KEY_UNTAB); + + newtSetSuspendCallback(newt_suspend, NULL); + ui_helpline__init(); + ui_browser__init(); + ui_progress__init(); + + signal(SIGSEGV, ui__signal); + signal(SIGFPE, ui__signal); + signal(SIGINT, ui__signal); + signal(SIGQUIT, ui__signal); + signal(SIGTERM, ui__signal); + + perf_error__register(&perf_tui_eops); + + hist_browser__init_hpp(); +out: + return err; +} + +void ui__exit(bool wait_for_ok) +{ + if (wait_for_ok) + ui__question_window("Fatal Error", + ui_helpline__last_msg, + "Press any key...", 0); + + SLtt_set_cursor_visibility(1); + SLsmg_refresh(); + SLsmg_reset_smg(); + SLang_reset_tty(); + + perf_error__unregister(&perf_tui_eops); +} diff --git a/tools/perf/ui/tui/util.c b/tools/perf/ui/tui/util.c new file mode 100644 index 00000000..092902e3 --- /dev/null +++ b/tools/perf/ui/tui/util.c @@ -0,0 +1,243 @@ +#include "../../util/util.h" +#include <signal.h> +#include <stdbool.h> +#include <string.h> +#include <sys/ttydefaults.h> + +#include "../../util/cache.h" +#include "../../util/debug.h" +#include "../browser.h" +#include "../keysyms.h" +#include "../helpline.h" +#include "../ui.h" +#include "../util.h" +#include "../libslang.h" + +static void ui_browser__argv_write(struct ui_browser *browser, + void *entry, int row) +{ + char **arg = entry; + bool current_entry = ui_browser__is_current_entry(browser, row); + + ui_browser__set_color(browser, current_entry ? HE_COLORSET_SELECTED : + HE_COLORSET_NORMAL); + slsmg_write_nstring(*arg, browser->width); +} + +static int popup_menu__run(struct ui_browser *menu) +{ + int key; + + if (ui_browser__show(menu, " ", "ESC: exit, ENTER|->: Select option") < 0) + return -1; + + while (1) { + key = ui_browser__run(menu, 0); + + switch (key) { + case K_RIGHT: + case K_ENTER: + key = menu->index; + break; + case K_LEFT: + case K_ESC: + case 'q': + case CTRL('c'): + key = -1; + break; + default: + continue; + } + + break; + } + + ui_browser__hide(menu); + return key; +} + +int ui__popup_menu(int argc, char * const argv[]) +{ + struct ui_browser menu = { + .entries = (void *)argv, + .refresh = ui_browser__argv_refresh, + .seek = ui_browser__argv_seek, + .write = ui_browser__argv_write, + .nr_entries = argc, + }; + + return popup_menu__run(&menu); +} + +int ui_browser__input_window(const char *title, const char *text, char *input, + const char *exit_msg, int delay_secs) +{ + int x, y, len, key; + int max_len = 60, nr_lines = 0; + static char buf[50]; + const char *t; + + t = text; + while (1) { + const char *sep = strchr(t, '\n'); + + if (sep == NULL) + sep = strchr(t, '\0'); + len = sep - t; + if (max_len < len) + max_len = len; + ++nr_lines; + if (*sep == '\0') + break; + t = sep + 1; + } + + max_len += 2; + nr_lines += 8; + y = SLtt_Screen_Rows / 2 - nr_lines / 2; + x = SLtt_Screen_Cols / 2 - max_len / 2; + + SLsmg_set_color(0); + SLsmg_draw_box(y, x++, nr_lines, max_len); + if (title) { + SLsmg_gotorc(y, x + 1); + SLsmg_write_string((char *)title); + } + SLsmg_gotorc(++y, x); + nr_lines -= 7; + max_len -= 2; + SLsmg_write_wrapped_string((unsigned char *)text, y, x, + nr_lines, max_len, 1); + y += nr_lines; + len = 5; + while (len--) { + SLsmg_gotorc(y + len - 1, x); + SLsmg_write_nstring((char *)" ", max_len); + } + SLsmg_draw_box(y++, x + 1, 3, max_len - 2); + + SLsmg_gotorc(y + 3, x); + SLsmg_write_nstring((char *)exit_msg, max_len); + SLsmg_refresh(); + + x += 2; + len = 0; + key = ui__getch(delay_secs); + while (key != K_TIMER && key != K_ENTER && key != K_ESC) { + if (key == K_BKSPC) { + if (len == 0) + goto next_key; + SLsmg_gotorc(y, x + --len); + SLsmg_write_char(' '); + } else { + buf[len] = key; + SLsmg_gotorc(y, x + len++); + SLsmg_write_char(key); + } + SLsmg_refresh(); + + /* XXX more graceful overflow handling needed */ + if (len == sizeof(buf) - 1) { + ui_helpline__push("maximum size of symbol name reached!"); + key = K_ENTER; + break; + } +next_key: + key = ui__getch(delay_secs); + } + + buf[len] = '\0'; + strncpy(input, buf, len+1); + return key; +} + +int ui__question_window(const char *title, const char *text, + const char *exit_msg, int delay_secs) +{ + int x, y; + int max_len = 0, nr_lines = 0; + const char *t; + + t = text; + while (1) { + const char *sep = strchr(t, '\n'); + int len; + + if (sep == NULL) + sep = strchr(t, '\0'); + len = sep - t; + if (max_len < len) + max_len = len; + ++nr_lines; + if (*sep == '\0') + break; + t = sep + 1; + } + + max_len += 2; + nr_lines += 4; + y = SLtt_Screen_Rows / 2 - nr_lines / 2, + x = SLtt_Screen_Cols / 2 - max_len / 2; + + SLsmg_set_color(0); + SLsmg_draw_box(y, x++, nr_lines, max_len); + if (title) { + SLsmg_gotorc(y, x + 1); + SLsmg_write_string((char *)title); + } + SLsmg_gotorc(++y, x); + nr_lines -= 2; + max_len -= 2; + SLsmg_write_wrapped_string((unsigned char *)text, y, x, + nr_lines, max_len, 1); + SLsmg_gotorc(y + nr_lines - 2, x); + SLsmg_write_nstring((char *)" ", max_len); + SLsmg_gotorc(y + nr_lines - 1, x); + SLsmg_write_nstring((char *)exit_msg, max_len); + SLsmg_refresh(); + return ui__getch(delay_secs); +} + +int ui__help_window(const char *text) +{ + return ui__question_window("Help", text, "Press any key...", 0); +} + +int ui__dialog_yesno(const char *msg) +{ + return ui__question_window(NULL, msg, "Enter: Yes, ESC: No", 0); +} + +static int __ui__warning(const char *title, const char *format, va_list args) +{ + char *s; + + if (vasprintf(&s, format, args) > 0) { + int key; + + pthread_mutex_lock(&ui__lock); + key = ui__question_window(title, s, "Press any key...", 0); + pthread_mutex_unlock(&ui__lock); + free(s); + return key; + } + + fprintf(stderr, "%s\n", title); + vfprintf(stderr, format, args); + return K_ESC; +} + +static int perf_tui__error(const char *format, va_list args) +{ + return __ui__warning("Error:", format, args); +} + +static int perf_tui__warning(const char *format, va_list args) +{ + return __ui__warning("Warning:", format, args); +} + +struct perf_error_ops perf_tui_eops = { + .error = perf_tui__error, + .warning = perf_tui__warning, +}; diff --git a/tools/perf/ui/ui.h b/tools/perf/ui/ui.h new file mode 100644 index 00000000..d86359c9 --- /dev/null +++ b/tools/perf/ui/ui.h @@ -0,0 +1,39 @@ +#ifndef _PERF_UI_H_ +#define _PERF_UI_H_ 1 + +#include <pthread.h> +#include <stdbool.h> +#include <linux/compiler.h> + +extern pthread_mutex_t ui__lock; + +extern int use_browser; + +void setup_browser(bool fallback_to_pager); +void exit_browser(bool wait_for_ok); + +#ifdef NEWT_SUPPORT +int ui__init(void); +void ui__exit(bool wait_for_ok); +#else +static inline int ui__init(void) +{ + return -1; +} +static inline void ui__exit(bool wait_for_ok __maybe_unused) {} +#endif + +#ifdef GTK2_SUPPORT +int perf_gtk__init(void); +void perf_gtk__exit(bool wait_for_ok); +#else +static inline int perf_gtk__init(void) +{ + return -1; +} +static inline void perf_gtk__exit(bool wait_for_ok __maybe_unused) {} +#endif + +void ui__refresh_dimensions(bool force); + +#endif /* _PERF_UI_H_ */ diff --git a/tools/perf/ui/util.c b/tools/perf/ui/util.c new file mode 100644 index 00000000..e3e0a963 --- /dev/null +++ b/tools/perf/ui/util.c @@ -0,0 +1,84 @@ +#include "util.h" +#include "../debug.h" + + +/* + * Default error logging functions + */ +static int perf_stdio__error(const char *format, va_list args) +{ + fprintf(stderr, "Error:\n"); + vfprintf(stderr, format, args); + return 0; +} + +static int perf_stdio__warning(const char *format, va_list args) +{ + fprintf(stderr, "Warning:\n"); + vfprintf(stderr, format, args); + return 0; +} + +static struct perf_error_ops default_eops = +{ + .error = perf_stdio__error, + .warning = perf_stdio__warning, +}; + +static struct perf_error_ops *perf_eops = &default_eops; + + +int ui__error(const char *format, ...) +{ + int ret; + va_list args; + + va_start(args, format); + ret = perf_eops->error(format, args); + va_end(args); + + return ret; +} + +int ui__warning(const char *format, ...) +{ + int ret; + va_list args; + + va_start(args, format); + ret = perf_eops->warning(format, args); + va_end(args); + + return ret; +} + +/** + * perf_error__register - Register error logging functions + * @eops: The pointer to error logging function struct + * + * Register UI-specific error logging functions. Before calling this, + * other logging functions should be unregistered, if any. + */ +int perf_error__register(struct perf_error_ops *eops) +{ + if (perf_eops != &default_eops) + return -1; + + perf_eops = eops; + return 0; +} + +/** + * perf_error__unregister - Unregister error logging functions + * @eops: The pointer to error logging function struct + * + * Unregister already registered error logging functions. + */ +int perf_error__unregister(struct perf_error_ops *eops) +{ + if (perf_eops != eops) + return -1; + + perf_eops = &default_eops; + return 0; +} diff --git a/tools/perf/ui/util.h b/tools/perf/ui/util.h new file mode 100644 index 00000000..361f08c5 --- /dev/null +++ b/tools/perf/ui/util.h @@ -0,0 +1,21 @@ +#ifndef _PERF_UI_UTIL_H_ +#define _PERF_UI_UTIL_H_ 1 + +#include <stdarg.h> + +int ui__getch(int delay_secs); +int ui__popup_menu(int argc, char * const argv[]); +int ui__help_window(const char *text); +int ui__dialog_yesno(const char *msg); +int ui__question_window(const char *title, const char *text, + const char *exit_msg, int delay_secs); + +struct perf_error_ops { + int (*error)(const char *format, va_list args); + int (*warning)(const char *format, va_list args); +}; + +int perf_error__register(struct perf_error_ops *eops); +int perf_error__unregister(struct perf_error_ops *eops); + +#endif /* _PERF_UI_UTIL_H_ */ diff --git a/tools/perf/util/PERF-VERSION-GEN b/tools/perf/util/PERF-VERSION-GEN new file mode 100755 index 00000000..055fef34 --- /dev/null +++ b/tools/perf/util/PERF-VERSION-GEN @@ -0,0 +1,38 @@ +#!/bin/sh + +if [ $# -eq 1 ] ; then + OUTPUT=$1 +fi + +GVF=${OUTPUT}PERF-VERSION-FILE + +LF=' +' + +# +# First check if there is a .git to get the version from git describe +# otherwise try to get the version from the kernel Makefile +# +if test -d ../../.git -o -f ../../.git && + VN=$(git tag 2>/dev/null | tail -1 | grep -E "v[0-9].[0-9]*") +then + VN=$(echo $VN"-g"$(git log -1 --abbrev=4 --pretty=format:"%h" HEAD)) + VN=$(echo "$VN" | sed -e 's/-/./g'); +else + VN=$(MAKEFLAGS= make -sC ../.. kernelversion) +fi + +VN=$(expr "$VN" : v*'\(.*\)') + +if test -r $GVF +then + VC=$(sed -e 's/^#define PERF_VERSION "\(.*\)"/\1/' <$GVF) +else + VC=unset +fi +test "$VN" = "$VC" || { + echo >&2 "PERF_VERSION = $VN" + echo "#define PERF_VERSION \"$VN\"" >$GVF +} + + diff --git a/tools/perf/util/abspath.c b/tools/perf/util/abspath.c new file mode 100644 index 00000000..0e76affe --- /dev/null +++ b/tools/perf/util/abspath.c @@ -0,0 +1,37 @@ +#include "cache.h" + +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 00000000..e6d13477 --- /dev/null +++ b/tools/perf/util/alias.c @@ -0,0 +1,78 @@ +#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 __maybe_unused) +{ + 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/annotate.c b/tools/perf/util/annotate.c new file mode 100644 index 00000000..d33fe937 --- /dev/null +++ b/tools/perf/util/annotate.c @@ -0,0 +1,1194 @@ +/* + * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com> + * + * Parts came from builtin-annotate.c, see those files for further + * copyright notes. + * + * Released under the GPL v2. (and only v2, not any later version) + */ + +#include "util.h" +#include "build-id.h" +#include "color.h" +#include "cache.h" +#include "symbol.h" +#include "debug.h" +#include "annotate.h" +#include <pthread.h> +#include <linux/bitops.h> + +const char *disassembler_style; +const char *objdump_path; + +static struct ins *ins__find(const char *name); +static int disasm_line__parse(char *line, char **namep, char **rawp); + +static void ins__delete(struct ins_operands *ops) +{ + free(ops->source.raw); + free(ops->source.name); + free(ops->target.raw); + free(ops->target.name); +} + +static int ins__raw_scnprintf(struct ins *ins, char *bf, size_t size, + struct ins_operands *ops) +{ + return scnprintf(bf, size, "%-6.6s %s", ins->name, ops->raw); +} + +int ins__scnprintf(struct ins *ins, char *bf, size_t size, + struct ins_operands *ops) +{ + if (ins->ops->scnprintf) + return ins->ops->scnprintf(ins, bf, size, ops); + + return ins__raw_scnprintf(ins, bf, size, ops); +} + +static int call__parse(struct ins_operands *ops) +{ + char *endptr, *tok, *name; + + ops->target.addr = strtoull(ops->raw, &endptr, 16); + + name = strchr(endptr, '<'); + if (name == NULL) + goto indirect_call; + + name++; + + tok = strchr(name, '>'); + if (tok == NULL) + return -1; + + *tok = '\0'; + ops->target.name = strdup(name); + *tok = '>'; + + return ops->target.name == NULL ? -1 : 0; + +indirect_call: + tok = strchr(endptr, '('); + if (tok != NULL) { + ops->target.addr = 0; + return 0; + } + + tok = strchr(endptr, '*'); + if (tok == NULL) + return -1; + + ops->target.addr = strtoull(tok + 1, NULL, 16); + return 0; +} + +static int call__scnprintf(struct ins *ins, char *bf, size_t size, + struct ins_operands *ops) +{ + if (ops->target.name) + return scnprintf(bf, size, "%-6.6s %s", ins->name, ops->target.name); + + if (ops->target.addr == 0) + return ins__raw_scnprintf(ins, bf, size, ops); + + return scnprintf(bf, size, "%-6.6s *%" PRIx64, ins->name, ops->target.addr); +} + +static struct ins_ops call_ops = { + .parse = call__parse, + .scnprintf = call__scnprintf, +}; + +bool ins__is_call(const struct ins *ins) +{ + return ins->ops == &call_ops; +} + +static int jump__parse(struct ins_operands *ops) +{ + const char *s = strchr(ops->raw, '+'); + + ops->target.addr = strtoll(ops->raw, NULL, 16); + + if (s++ != NULL) + ops->target.offset = strtoll(s, NULL, 16); + else + ops->target.offset = UINT64_MAX; + + return 0; +} + +static int jump__scnprintf(struct ins *ins, char *bf, size_t size, + struct ins_operands *ops) +{ + return scnprintf(bf, size, "%-6.6s %" PRIx64, ins->name, ops->target.offset); +} + +static struct ins_ops jump_ops = { + .parse = jump__parse, + .scnprintf = jump__scnprintf, +}; + +bool ins__is_jump(const struct ins *ins) +{ + return ins->ops == &jump_ops; +} + +static int comment__symbol(char *raw, char *comment, u64 *addrp, char **namep) +{ + char *endptr, *name, *t; + + if (strstr(raw, "(%rip)") == NULL) + return 0; + + *addrp = strtoull(comment, &endptr, 16); + name = strchr(endptr, '<'); + if (name == NULL) + return -1; + + name++; + + t = strchr(name, '>'); + if (t == NULL) + return 0; + + *t = '\0'; + *namep = strdup(name); + *t = '>'; + + return 0; +} + +static int lock__parse(struct ins_operands *ops) +{ + char *name; + + ops->locked.ops = zalloc(sizeof(*ops->locked.ops)); + if (ops->locked.ops == NULL) + return 0; + + if (disasm_line__parse(ops->raw, &name, &ops->locked.ops->raw) < 0) + goto out_free_ops; + + ops->locked.ins = ins__find(name); + if (ops->locked.ins == NULL) + goto out_free_ops; + + if (!ops->locked.ins->ops) + return 0; + + if (ops->locked.ins->ops->parse) + ops->locked.ins->ops->parse(ops->locked.ops); + + return 0; + +out_free_ops: + free(ops->locked.ops); + ops->locked.ops = NULL; + return 0; +} + +static int lock__scnprintf(struct ins *ins, char *bf, size_t size, + struct ins_operands *ops) +{ + int printed; + + if (ops->locked.ins == NULL) + return ins__raw_scnprintf(ins, bf, size, ops); + + printed = scnprintf(bf, size, "%-6.6s ", ins->name); + return printed + ins__scnprintf(ops->locked.ins, bf + printed, + size - printed, ops->locked.ops); +} + +static void lock__delete(struct ins_operands *ops) +{ + free(ops->locked.ops); + free(ops->target.raw); + free(ops->target.name); +} + +static struct ins_ops lock_ops = { + .free = lock__delete, + .parse = lock__parse, + .scnprintf = lock__scnprintf, +}; + +static int mov__parse(struct ins_operands *ops) +{ + char *s = strchr(ops->raw, ','), *target, *comment, prev; + + if (s == NULL) + return -1; + + *s = '\0'; + ops->source.raw = strdup(ops->raw); + *s = ','; + + if (ops->source.raw == NULL) + return -1; + + target = ++s; + + while (s[0] != '\0' && !isspace(s[0])) + ++s; + prev = *s; + *s = '\0'; + + ops->target.raw = strdup(target); + *s = prev; + + if (ops->target.raw == NULL) + goto out_free_source; + + comment = strchr(s, '#'); + if (comment == NULL) + return 0; + + while (comment[0] != '\0' && isspace(comment[0])) + ++comment; + + comment__symbol(ops->source.raw, comment, &ops->source.addr, &ops->source.name); + comment__symbol(ops->target.raw, comment, &ops->target.addr, &ops->target.name); + + return 0; + +out_free_source: + free(ops->source.raw); + ops->source.raw = NULL; + return -1; +} + +static int mov__scnprintf(struct ins *ins, char *bf, size_t size, + struct ins_operands *ops) +{ + return scnprintf(bf, size, "%-6.6s %s,%s", ins->name, + ops->source.name ?: ops->source.raw, + ops->target.name ?: ops->target.raw); +} + +static struct ins_ops mov_ops = { + .parse = mov__parse, + .scnprintf = mov__scnprintf, +}; + +static int dec__parse(struct ins_operands *ops) +{ + char *target, *comment, *s, prev; + + target = s = ops->raw; + + while (s[0] != '\0' && !isspace(s[0])) + ++s; + prev = *s; + *s = '\0'; + + ops->target.raw = strdup(target); + *s = prev; + + if (ops->target.raw == NULL) + return -1; + + comment = strchr(s, '#'); + if (comment == NULL) + return 0; + + while (comment[0] != '\0' && isspace(comment[0])) + ++comment; + + comment__symbol(ops->target.raw, comment, &ops->target.addr, &ops->target.name); + + return 0; +} + +static int dec__scnprintf(struct ins *ins, char *bf, size_t size, + struct ins_operands *ops) +{ + return scnprintf(bf, size, "%-6.6s %s", ins->name, + ops->target.name ?: ops->target.raw); +} + +static struct ins_ops dec_ops = { + .parse = dec__parse, + .scnprintf = dec__scnprintf, +}; + +static int nop__scnprintf(struct ins *ins __maybe_unused, char *bf, size_t size, + struct ins_operands *ops __maybe_unused) +{ + return scnprintf(bf, size, "%-6.6s", "nop"); +} + +static struct ins_ops nop_ops = { + .scnprintf = nop__scnprintf, +}; + +/* + * Must be sorted by name! + */ +static struct ins instructions[] = { + { .name = "add", .ops = &mov_ops, }, + { .name = "addl", .ops = &mov_ops, }, + { .name = "addq", .ops = &mov_ops, }, + { .name = "addw", .ops = &mov_ops, }, + { .name = "and", .ops = &mov_ops, }, + { .name = "bts", .ops = &mov_ops, }, + { .name = "call", .ops = &call_ops, }, + { .name = "callq", .ops = &call_ops, }, + { .name = "cmp", .ops = &mov_ops, }, + { .name = "cmpb", .ops = &mov_ops, }, + { .name = "cmpl", .ops = &mov_ops, }, + { .name = "cmpq", .ops = &mov_ops, }, + { .name = "cmpw", .ops = &mov_ops, }, + { .name = "cmpxch", .ops = &mov_ops, }, + { .name = "dec", .ops = &dec_ops, }, + { .name = "decl", .ops = &dec_ops, }, + { .name = "imul", .ops = &mov_ops, }, + { .name = "inc", .ops = &dec_ops, }, + { .name = "incl", .ops = &dec_ops, }, + { .name = "ja", .ops = &jump_ops, }, + { .name = "jae", .ops = &jump_ops, }, + { .name = "jb", .ops = &jump_ops, }, + { .name = "jbe", .ops = &jump_ops, }, + { .name = "jc", .ops = &jump_ops, }, + { .name = "jcxz", .ops = &jump_ops, }, + { .name = "je", .ops = &jump_ops, }, + { .name = "jecxz", .ops = &jump_ops, }, + { .name = "jg", .ops = &jump_ops, }, + { .name = "jge", .ops = &jump_ops, }, + { .name = "jl", .ops = &jump_ops, }, + { .name = "jle", .ops = &jump_ops, }, + { .name = "jmp", .ops = &jump_ops, }, + { .name = "jmpq", .ops = &jump_ops, }, + { .name = "jna", .ops = &jump_ops, }, + { .name = "jnae", .ops = &jump_ops, }, + { .name = "jnb", .ops = &jump_ops, }, + { .name = "jnbe", .ops = &jump_ops, }, + { .name = "jnc", .ops = &jump_ops, }, + { .name = "jne", .ops = &jump_ops, }, + { .name = "jng", .ops = &jump_ops, }, + { .name = "jnge", .ops = &jump_ops, }, + { .name = "jnl", .ops = &jump_ops, }, + { .name = "jnle", .ops = &jump_ops, }, + { .name = "jno", .ops = &jump_ops, }, + { .name = "jnp", .ops = &jump_ops, }, + { .name = "jns", .ops = &jump_ops, }, + { .name = "jnz", .ops = &jump_ops, }, + { .name = "jo", .ops = &jump_ops, }, + { .name = "jp", .ops = &jump_ops, }, + { .name = "jpe", .ops = &jump_ops, }, + { .name = "jpo", .ops = &jump_ops, }, + { .name = "jrcxz", .ops = &jump_ops, }, + { .name = "js", .ops = &jump_ops, }, + { .name = "jz", .ops = &jump_ops, }, + { .name = "lea", .ops = &mov_ops, }, + { .name = "lock", .ops = &lock_ops, }, + { .name = "mov", .ops = &mov_ops, }, + { .name = "movb", .ops = &mov_ops, }, + { .name = "movdqa",.ops = &mov_ops, }, + { .name = "movl", .ops = &mov_ops, }, + { .name = "movq", .ops = &mov_ops, }, + { .name = "movslq", .ops = &mov_ops, }, + { .name = "movzbl", .ops = &mov_ops, }, + { .name = "movzwl", .ops = &mov_ops, }, + { .name = "nop", .ops = &nop_ops, }, + { .name = "nopl", .ops = &nop_ops, }, + { .name = "nopw", .ops = &nop_ops, }, + { .name = "or", .ops = &mov_ops, }, + { .name = "orl", .ops = &mov_ops, }, + { .name = "test", .ops = &mov_ops, }, + { .name = "testb", .ops = &mov_ops, }, + { .name = "testl", .ops = &mov_ops, }, + { .name = "xadd", .ops = &mov_ops, }, + { .name = "xbeginl", .ops = &jump_ops, }, + { .name = "xbeginq", .ops = &jump_ops, }, +}; + +static int ins__cmp(const void *name, const void *insp) +{ + const struct ins *ins = insp; + + return strcmp(name, ins->name); +} + +static struct ins *ins__find(const char *name) +{ + const int nmemb = ARRAY_SIZE(instructions); + + return bsearch(name, instructions, nmemb, sizeof(struct ins), ins__cmp); +} + +int symbol__annotate_init(struct map *map __maybe_unused, struct symbol *sym) +{ + struct annotation *notes = symbol__annotation(sym); + pthread_mutex_init(¬es->lock, NULL); + return 0; +} + +int symbol__alloc_hist(struct symbol *sym) +{ + struct annotation *notes = symbol__annotation(sym); + const size_t size = symbol__size(sym); + size_t sizeof_sym_hist; + + /* Check for overflow when calculating sizeof_sym_hist */ + if (size > (SIZE_MAX - sizeof(struct sym_hist)) / sizeof(u64)) + return -1; + + sizeof_sym_hist = (sizeof(struct sym_hist) + size * sizeof(u64)); + + /* Check for overflow in zalloc argument */ + if (sizeof_sym_hist > (SIZE_MAX - sizeof(*notes->src)) + / symbol_conf.nr_events) + return -1; + + notes->src = zalloc(sizeof(*notes->src) + symbol_conf.nr_events * sizeof_sym_hist); + if (notes->src == NULL) + return -1; + notes->src->sizeof_sym_hist = sizeof_sym_hist; + notes->src->nr_histograms = symbol_conf.nr_events; + INIT_LIST_HEAD(¬es->src->source); + return 0; +} + +void symbol__annotate_zero_histograms(struct symbol *sym) +{ + struct annotation *notes = symbol__annotation(sym); + + pthread_mutex_lock(¬es->lock); + if (notes->src != NULL) + memset(notes->src->histograms, 0, + notes->src->nr_histograms * notes->src->sizeof_sym_hist); + pthread_mutex_unlock(¬es->lock); +} + +int symbol__inc_addr_samples(struct symbol *sym, struct map *map, + int evidx, u64 addr) +{ + unsigned offset; + struct annotation *notes; + struct sym_hist *h; + + notes = symbol__annotation(sym); + if (notes->src == NULL) + return -ENOMEM; + + pr_debug3("%s: addr=%#" PRIx64 "\n", __func__, map->unmap_ip(map, addr)); + + if (addr < sym->start || addr > sym->end) + return -ERANGE; + + offset = addr - sym->start; + h = annotation__histogram(notes, evidx); + h->sum++; + h->addr[offset]++; + + pr_debug3("%#" PRIx64 " %s: period++ [addr: %#" PRIx64 ", %#" PRIx64 + ", evidx=%d] => %" PRIu64 "\n", sym->start, sym->name, + addr, addr - sym->start, evidx, h->addr[offset]); + return 0; +} + +static void disasm_line__init_ins(struct disasm_line *dl) +{ + dl->ins = ins__find(dl->name); + + if (dl->ins == NULL) + return; + + if (!dl->ins->ops) + return; + + if (dl->ins->ops->parse) + dl->ins->ops->parse(&dl->ops); +} + +static int disasm_line__parse(char *line, char **namep, char **rawp) +{ + char *name = line, tmp; + + while (isspace(name[0])) + ++name; + + if (name[0] == '\0') + return -1; + + *rawp = name + 1; + + while ((*rawp)[0] != '\0' && !isspace((*rawp)[0])) + ++*rawp; + + tmp = (*rawp)[0]; + (*rawp)[0] = '\0'; + *namep = strdup(name); + + if (*namep == NULL) + goto out_free_name; + + (*rawp)[0] = tmp; + + if ((*rawp)[0] != '\0') { + (*rawp)++; + while (isspace((*rawp)[0])) + ++(*rawp); + } + + return 0; + +out_free_name: + free(*namep); + *namep = NULL; + return -1; +} + +static struct disasm_line *disasm_line__new(s64 offset, char *line, size_t privsize) +{ + struct disasm_line *dl = zalloc(sizeof(*dl) + privsize); + + if (dl != NULL) { + dl->offset = offset; + dl->line = strdup(line); + if (dl->line == NULL) + goto out_delete; + + if (offset != -1) { + if (disasm_line__parse(dl->line, &dl->name, &dl->ops.raw) < 0) + goto out_free_line; + + disasm_line__init_ins(dl); + } + } + + return dl; + +out_free_line: + free(dl->line); +out_delete: + free(dl); + return NULL; +} + +void disasm_line__free(struct disasm_line *dl) +{ + free(dl->line); + free(dl->name); + if (dl->ins && dl->ins->ops->free) + dl->ins->ops->free(&dl->ops); + else + ins__delete(&dl->ops); + free(dl); +} + +int disasm_line__scnprintf(struct disasm_line *dl, char *bf, size_t size, bool raw) +{ + if (raw || !dl->ins) + return scnprintf(bf, size, "%-6.6s %s", dl->name, dl->ops.raw); + + return ins__scnprintf(dl->ins, bf, size, &dl->ops); +} + +static void disasm__add(struct list_head *head, struct disasm_line *line) +{ + list_add_tail(&line->node, head); +} + +struct disasm_line *disasm__get_next_ip_line(struct list_head *head, struct disasm_line *pos) +{ + list_for_each_entry_continue(pos, head, node) + if (pos->offset >= 0) + return pos; + + return NULL; +} + +static int disasm_line__print(struct disasm_line *dl, struct symbol *sym, u64 start, + int evidx, u64 len, int min_pcnt, int printed, + int max_lines, struct disasm_line *queue) +{ + static const char *prev_line; + static const char *prev_color; + + if (dl->offset != -1) { + const char *path = NULL; + unsigned int hits = 0; + double percent = 0.0; + const char *color; + struct annotation *notes = symbol__annotation(sym); + struct source_line *src_line = notes->src->lines; + struct sym_hist *h = annotation__histogram(notes, evidx); + s64 offset = dl->offset; + const u64 addr = start + offset; + struct disasm_line *next; + + next = disasm__get_next_ip_line(¬es->src->source, dl); + + while (offset < (s64)len && + (next == NULL || offset < next->offset)) { + if (src_line) { + if (path == NULL) + path = src_line[offset].path; + percent += src_line[offset].percent; + } else + hits += h->addr[offset]; + + ++offset; + } + + if (src_line == NULL && h->sum) + percent = 100.0 * hits / h->sum; + + if (percent < min_pcnt) + return -1; + + if (max_lines && printed >= max_lines) + return 1; + + if (queue != NULL) { + list_for_each_entry_from(queue, ¬es->src->source, node) { + if (queue == dl) + break; + disasm_line__print(queue, sym, start, evidx, len, + 0, 0, 1, NULL); + } + } + + color = get_percent_color(percent); + + /* + * Also color the filename and line if needed, with + * the same color than the percentage. Don't print it + * twice for close colored addr with the same filename:line + */ + if (path) { + if (!prev_line || strcmp(prev_line, path) + || color != prev_color) { + color_fprintf(stdout, color, " %s", path); + prev_line = path; + prev_color = color; + } + } + + color_fprintf(stdout, color, " %7.2f", percent); + printf(" : "); + color_fprintf(stdout, PERF_COLOR_MAGENTA, " %" PRIx64 ":", addr); + color_fprintf(stdout, PERF_COLOR_BLUE, "%s\n", dl->line); + } else if (max_lines && printed >= max_lines) + return 1; + else { + if (queue) + return -1; + + if (!*dl->line) + printf(" :\n"); + else + printf(" : %s\n", dl->line); + } + + return 0; +} + +static int symbol__parse_objdump_line(struct symbol *sym, struct map *map, + FILE *file, size_t privsize) +{ + struct annotation *notes = symbol__annotation(sym); + struct disasm_line *dl; + char *line = NULL, *parsed_line, *tmp, *tmp2, *c; + size_t line_len; + s64 line_ip, offset = -1; + + if (getline(&line, &line_len, file) < 0) + return -1; + + if (!line) + return -1; + + while (line_len != 0 && isspace(line[line_len - 1])) + line[--line_len] = '\0'; + + c = strchr(line, '\n'); + if (c) + *c = 0; + + line_ip = -1; + parsed_line = line; + + /* + * Strip leading spaces: + */ + tmp = line; + while (*tmp) { + if (*tmp != ' ') + break; + tmp++; + } + + if (*tmp) { + /* + * Parse hexa addresses followed by ':' + */ + line_ip = strtoull(tmp, &tmp2, 16); + if (*tmp2 != ':' || tmp == tmp2 || tmp2[1] == '\0') + line_ip = -1; + } + + if (line_ip != -1) { + u64 start = map__rip_2objdump(map, sym->start), + end = map__rip_2objdump(map, sym->end); + + offset = line_ip - start; + if (offset < 0 || (u64)line_ip > end) + offset = -1; + else + parsed_line = tmp2 + 1; + } + + dl = disasm_line__new(offset, parsed_line, privsize); + free(line); + + if (dl == NULL) + return -1; + + disasm__add(¬es->src->source, dl); + + return 0; +} + +int symbol__annotate(struct symbol *sym, struct map *map, size_t privsize) +{ + struct dso *dso = map->dso; + char *filename = dso__build_id_filename(dso, NULL, 0); + bool free_filename = true; + char command[PATH_MAX * 2]; + FILE *file; + int err = 0; + char symfs_filename[PATH_MAX]; + + if (filename) { + snprintf(symfs_filename, sizeof(symfs_filename), "%s%s", + symbol_conf.symfs, filename); + } + + if (filename == NULL) { + if (dso->has_build_id) { + pr_err("Can't annotate %s: not enough memory\n", + sym->name); + return -ENOMEM; + } + goto fallback; + } else if (readlink(symfs_filename, command, sizeof(command)) < 0 || + strstr(command, "[kernel.kallsyms]") || + access(symfs_filename, R_OK)) { + free(filename); +fallback: + /* + * If we don't have build-ids or the build-id file isn't in the + * cache, or is just a kallsyms file, well, lets hope that this + * DSO is the same as when 'perf record' ran. + */ + filename = dso->long_name; + snprintf(symfs_filename, sizeof(symfs_filename), "%s%s", + symbol_conf.symfs, filename); + free_filename = false; + } + + if (dso->symtab_type == DSO_BINARY_TYPE__KALLSYMS) { + char bf[BUILD_ID_SIZE * 2 + 16] = " with build id "; + char *build_id_msg = NULL; + + if (dso->annotate_warned) + goto out_free_filename; + + if (dso->has_build_id) { + build_id__sprintf(dso->build_id, + sizeof(dso->build_id), bf + 15); + build_id_msg = bf; + } + err = -ENOENT; + dso->annotate_warned = 1; + pr_err("Can't annotate %s:\n\n" + "No vmlinux file%s\nwas found in the path.\n\n" + "Please use:\n\n" + " perf buildid-cache -vu vmlinux\n\n" + "or:\n\n" + " --vmlinux vmlinux\n", + sym->name, build_id_msg ?: ""); + goto out_free_filename; + } + + pr_debug("%s: filename=%s, sym=%s, start=%#" PRIx64 ", end=%#" PRIx64 "\n", __func__, + filename, sym->name, map->unmap_ip(map, sym->start), + map->unmap_ip(map, sym->end)); + + pr_debug("annotating [%p] %30s : [%p] %30s\n", + dso, dso->long_name, sym, sym->name); + + snprintf(command, sizeof(command), + "%s %s%s --start-address=0x%016" PRIx64 + " --stop-address=0x%016" PRIx64 + " -d %s %s -C %s|grep -v %s|expand", + objdump_path ? objdump_path : "objdump", + disassembler_style ? "-M " : "", + disassembler_style ? disassembler_style : "", + map__rip_2objdump(map, sym->start), + map__rip_2objdump(map, sym->end+1), + symbol_conf.annotate_asm_raw ? "" : "--no-show-raw", + symbol_conf.annotate_src ? "-S" : "", + symfs_filename, filename); + + pr_debug("Executing: %s\n", command); + + file = popen(command, "r"); + if (!file) + goto out_free_filename; + + while (!feof(file)) + if (symbol__parse_objdump_line(sym, map, file, privsize) < 0) + break; + + pclose(file); +out_free_filename: + if (free_filename) + free(filename); + return err; +} + +static void insert_source_line(struct rb_root *root, struct source_line *src_line) +{ + struct source_line *iter; + struct rb_node **p = &root->rb_node; + struct rb_node *parent = NULL; + int ret; + + while (*p != NULL) { + parent = *p; + iter = rb_entry(parent, struct source_line, node); + + ret = strcmp(iter->path, src_line->path); + if (ret == 0) { + iter->percent_sum += src_line->percent; + return; + } + + if (ret < 0) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + src_line->percent_sum = src_line->percent; + + rb_link_node(&src_line->node, parent, p); + rb_insert_color(&src_line->node, root); +} + +static void __resort_source_line(struct rb_root *root, struct source_line *src_line) +{ + struct source_line *iter; + struct rb_node **p = &root->rb_node; + struct rb_node *parent = NULL; + + while (*p != NULL) { + parent = *p; + iter = rb_entry(parent, struct source_line, node); + + if (src_line->percent_sum > iter->percent_sum) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + rb_link_node(&src_line->node, parent, p); + rb_insert_color(&src_line->node, root); +} + +static void resort_source_line(struct rb_root *dest_root, struct rb_root *src_root) +{ + struct source_line *src_line; + struct rb_node *node; + + node = rb_first(src_root); + while (node) { + struct rb_node *next; + + src_line = rb_entry(node, struct source_line, node); + next = rb_next(node); + rb_erase(node, src_root); + + __resort_source_line(dest_root, src_line); + node = next; + } +} + +static void symbol__free_source_line(struct symbol *sym, int len) +{ + struct annotation *notes = symbol__annotation(sym); + struct source_line *src_line = notes->src->lines; + int i; + + for (i = 0; i < len; i++) + free(src_line[i].path); + + free(src_line); + notes->src->lines = NULL; +} + +/* Get the filename:line for the colored entries */ +static int symbol__get_source_line(struct symbol *sym, struct map *map, + int evidx, struct rb_root *root, int len, + const char *filename) +{ + u64 start; + int i; + char cmd[PATH_MAX * 2]; + struct source_line *src_line; + struct annotation *notes = symbol__annotation(sym); + struct sym_hist *h = annotation__histogram(notes, evidx); + struct rb_root tmp_root = RB_ROOT; + + if (!h->sum) + return 0; + + src_line = notes->src->lines = calloc(len, sizeof(struct source_line)); + if (!notes->src->lines) + return -1; + + start = map__rip_2objdump(map, sym->start); + + for (i = 0; i < len; i++) { + char *path = NULL; + size_t line_len; + u64 offset; + FILE *fp; + + src_line[i].percent = 100.0 * h->addr[i] / h->sum; + if (src_line[i].percent <= 0.5) + continue; + + offset = start + i; + sprintf(cmd, "addr2line -e %s %016" PRIx64, filename, offset); + fp = popen(cmd, "r"); + if (!fp) + continue; + + if (getline(&path, &line_len, fp) < 0 || !line_len) + goto next; + + src_line[i].path = malloc(sizeof(char) * line_len + 1); + if (!src_line[i].path) + goto next; + + strcpy(src_line[i].path, path); + insert_source_line(&tmp_root, &src_line[i]); + + next: + pclose(fp); + } + + resort_source_line(root, &tmp_root); + return 0; +} + +static void print_summary(struct rb_root *root, const char *filename) +{ + struct source_line *src_line; + struct rb_node *node; + + printf("\nSorted summary for file %s\n", filename); + printf("----------------------------------------------\n\n"); + + if (RB_EMPTY_ROOT(root)) { + printf(" Nothing higher than %1.1f%%\n", MIN_GREEN); + return; + } + + node = rb_first(root); + while (node) { + double percent; + const char *color; + char *path; + + src_line = rb_entry(node, struct source_line, node); + percent = src_line->percent_sum; + color = get_percent_color(percent); + path = src_line->path; + + color_fprintf(stdout, color, " %7.2f %s", percent, path); + node = rb_next(node); + } +} + +static void symbol__annotate_hits(struct symbol *sym, int evidx) +{ + struct annotation *notes = symbol__annotation(sym); + struct sym_hist *h = annotation__histogram(notes, evidx); + u64 len = symbol__size(sym), offset; + + for (offset = 0; offset < len; ++offset) + if (h->addr[offset] != 0) + printf("%*" PRIx64 ": %" PRIu64 "\n", BITS_PER_LONG / 2, + sym->start + offset, h->addr[offset]); + printf("%*s: %" PRIu64 "\n", BITS_PER_LONG / 2, "h->sum", h->sum); +} + +int symbol__annotate_printf(struct symbol *sym, struct map *map, int evidx, + bool full_paths, int min_pcnt, int max_lines, + int context) +{ + struct dso *dso = map->dso; + char *filename; + const char *d_filename; + struct annotation *notes = symbol__annotation(sym); + struct disasm_line *pos, *queue = NULL; + u64 start = map__rip_2objdump(map, sym->start); + int printed = 2, queue_len = 0; + int more = 0; + u64 len; + + filename = strdup(dso->long_name); + if (!filename) + return -ENOMEM; + + if (full_paths) + d_filename = filename; + else + d_filename = basename(filename); + + len = symbol__size(sym); + + printf(" Percent | Source code & Disassembly of %s\n", d_filename); + printf("------------------------------------------------\n"); + + if (verbose) + symbol__annotate_hits(sym, evidx); + + list_for_each_entry(pos, ¬es->src->source, node) { + if (context && queue == NULL) { + queue = pos; + queue_len = 0; + } + + switch (disasm_line__print(pos, sym, start, evidx, len, + min_pcnt, printed, max_lines, + queue)) { + case 0: + ++printed; + if (context) { + printed += queue_len; + queue = NULL; + queue_len = 0; + } + break; + case 1: + /* filtered by max_lines */ + ++more; + break; + case -1: + default: + /* + * Filtered by min_pcnt or non IP lines when + * context != 0 + */ + if (!context) + break; + if (queue_len == context) + queue = list_entry(queue->node.next, typeof(*queue), node); + else + ++queue_len; + break; + } + } + + free(filename); + + return more; +} + +void symbol__annotate_zero_histogram(struct symbol *sym, int evidx) +{ + struct annotation *notes = symbol__annotation(sym); + struct sym_hist *h = annotation__histogram(notes, evidx); + + memset(h, 0, notes->src->sizeof_sym_hist); +} + +void symbol__annotate_decay_histogram(struct symbol *sym, int evidx) +{ + struct annotation *notes = symbol__annotation(sym); + struct sym_hist *h = annotation__histogram(notes, evidx); + int len = symbol__size(sym), offset; + + h->sum = 0; + for (offset = 0; offset < len; ++offset) { + h->addr[offset] = h->addr[offset] * 7 / 8; + h->sum += h->addr[offset]; + } +} + +void disasm__purge(struct list_head *head) +{ + struct disasm_line *pos, *n; + + list_for_each_entry_safe(pos, n, head, node) { + list_del(&pos->node); + disasm_line__free(pos); + } +} + +static size_t disasm_line__fprintf(struct disasm_line *dl, FILE *fp) +{ + size_t printed; + + if (dl->offset == -1) + return fprintf(fp, "%s\n", dl->line); + + printed = fprintf(fp, "%#" PRIx64 " %s", dl->offset, dl->name); + + if (dl->ops.raw[0] != '\0') { + printed += fprintf(fp, "%.*s %s\n", 6 - (int)printed, " ", + dl->ops.raw); + } + + return printed + fprintf(fp, "\n"); +} + +size_t disasm__fprintf(struct list_head *head, FILE *fp) +{ + struct disasm_line *pos; + size_t printed = 0; + + list_for_each_entry(pos, head, node) + printed += disasm_line__fprintf(pos, fp); + + return printed; +} + +int symbol__tty_annotate(struct symbol *sym, struct map *map, int evidx, + bool print_lines, bool full_paths, int min_pcnt, + int max_lines) +{ + struct dso *dso = map->dso; + const char *filename = dso->long_name; + struct rb_root source_line = RB_ROOT; + u64 len; + + if (symbol__annotate(sym, map, 0) < 0) + return -1; + + len = symbol__size(sym); + + if (print_lines) { + symbol__get_source_line(sym, map, evidx, &source_line, + len, filename); + print_summary(&source_line, filename); + } + + symbol__annotate_printf(sym, map, evidx, full_paths, + min_pcnt, max_lines, 0); + if (print_lines) + symbol__free_source_line(sym, len); + + disasm__purge(&symbol__annotation(sym)->src->source); + + return 0; +} diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h new file mode 100644 index 00000000..c422440f --- /dev/null +++ b/tools/perf/util/annotate.h @@ -0,0 +1,183 @@ +#ifndef __PERF_ANNOTATE_H +#define __PERF_ANNOTATE_H + +#include <stdbool.h> +#include <stdint.h> +#include "types.h" +#include "symbol.h" +#include "hist.h" +#include "sort.h" +#include <linux/list.h> +#include <linux/rbtree.h> +#include <pthread.h> + +struct ins; + +struct ins_operands { + char *raw; + struct { + char *raw; + char *name; + u64 addr; + u64 offset; + } target; + union { + struct { + char *raw; + char *name; + u64 addr; + } source; + struct { + struct ins *ins; + struct ins_operands *ops; + } locked; + }; +}; + +struct ins_ops { + void (*free)(struct ins_operands *ops); + int (*parse)(struct ins_operands *ops); + int (*scnprintf)(struct ins *ins, char *bf, size_t size, + struct ins_operands *ops); +}; + +struct ins { + const char *name; + struct ins_ops *ops; +}; + +bool ins__is_jump(const struct ins *ins); +bool ins__is_call(const struct ins *ins); +int ins__scnprintf(struct ins *ins, char *bf, size_t size, struct ins_operands *ops); + +struct disasm_line { + struct list_head node; + s64 offset; + char *line; + char *name; + struct ins *ins; + struct ins_operands ops; +}; + +static inline bool disasm_line__has_offset(const struct disasm_line *dl) +{ + return dl->ops.target.offset != UINT64_MAX; +} + +void disasm_line__free(struct disasm_line *dl); +struct disasm_line *disasm__get_next_ip_line(struct list_head *head, struct disasm_line *pos); +int disasm_line__scnprintf(struct disasm_line *dl, char *bf, size_t size, bool raw); +size_t disasm__fprintf(struct list_head *head, FILE *fp); + +struct sym_hist { + u64 sum; + u64 addr[0]; +}; + +struct source_line { + struct rb_node node; + double percent; + double percent_sum; + char *path; +}; + +/** struct annotated_source - symbols with hits have this attached as in sannotation + * + * @histogram: Array of addr hit histograms per event being monitored + * @lines: If 'print_lines' is specified, per source code line percentages + * @source: source parsed from a disassembler like objdump -dS + * + * lines is allocated, percentages calculated and all sorted by percentage + * when the annotation is about to be presented, so the percentages are for + * one of the entries in the histogram array, i.e. for the event/counter being + * presented. It is deallocated right after symbol__{tui,tty,etc}_annotate + * returns. + */ +struct annotated_source { + struct list_head source; + struct source_line *lines; + int nr_histograms; + int sizeof_sym_hist; + struct sym_hist histograms[0]; +}; + +struct annotation { + pthread_mutex_t lock; + struct annotated_source *src; +}; + +struct sannotation { + struct annotation annotation; + struct symbol symbol; +}; + +static inline struct sym_hist *annotation__histogram(struct annotation *notes, int idx) +{ + return (((void *)¬es->src->histograms) + + (notes->src->sizeof_sym_hist * idx)); +} + +static inline struct annotation *symbol__annotation(struct symbol *sym) +{ + struct sannotation *a = container_of(sym, struct sannotation, symbol); + return &a->annotation; +} + +int symbol__inc_addr_samples(struct symbol *sym, struct map *map, + int evidx, u64 addr); +int symbol__alloc_hist(struct symbol *sym); +void symbol__annotate_zero_histograms(struct symbol *sym); + +int symbol__annotate(struct symbol *sym, struct map *map, size_t privsize); +int symbol__annotate_init(struct map *map __maybe_unused, struct symbol *sym); +int symbol__annotate_printf(struct symbol *sym, struct map *map, int evidx, + bool full_paths, int min_pcnt, int max_lines, + int context); +void symbol__annotate_zero_histogram(struct symbol *sym, int evidx); +void symbol__annotate_decay_histogram(struct symbol *sym, int evidx); +void disasm__purge(struct list_head *head); + +int symbol__tty_annotate(struct symbol *sym, struct map *map, int evidx, + bool print_lines, bool full_paths, int min_pcnt, + int max_lines); + +#ifdef NEWT_SUPPORT +int symbol__tui_annotate(struct symbol *sym, struct map *map, int evidx, + struct hist_browser_timer *hbt); +#else +static inline int symbol__tui_annotate(struct symbol *sym __maybe_unused, + struct map *map __maybe_unused, + int evidx __maybe_unused, + struct hist_browser_timer *hbt + __maybe_unused) +{ + return 0; +} +#endif + +#ifdef GTK2_SUPPORT +int symbol__gtk_annotate(struct symbol *sym, struct map *map, int evidx, + struct hist_browser_timer *hbt); + +static inline int hist_entry__gtk_annotate(struct hist_entry *he, int evidx, + struct hist_browser_timer *hbt) +{ + return symbol__gtk_annotate(he->ms.sym, he->ms.map, evidx, hbt); +} + +void perf_gtk__show_annotations(void); +#else +static inline int hist_entry__gtk_annotate(struct hist_entry *he __maybe_unused, + int evidx __maybe_unused, + struct hist_browser_timer *hbt + __maybe_unused) +{ + return 0; +} + +static inline void perf_gtk__show_annotations(void) {} +#endif + +extern const char *disassembler_style; + +#endif /* __PERF_ANNOTATE_H */ diff --git a/tools/perf/util/bitmap.c b/tools/perf/util/bitmap.c new file mode 100644 index 00000000..0a1adc11 --- /dev/null +++ b/tools/perf/util/bitmap.c @@ -0,0 +1,31 @@ +/* + * From lib/bitmap.c + * Helper functions for bitmap.h. + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ +#include <linux/bitmap.h> + +int __bitmap_weight(const unsigned long *bitmap, int bits) +{ + int k, w = 0, lim = bits/BITS_PER_LONG; + + for (k = 0; k < lim; k++) + w += hweight_long(bitmap[k]); + + if (bits % BITS_PER_LONG) + w += hweight_long(bitmap[k] & BITMAP_LAST_WORD_MASK(bits)); + + return w; +} + +void __bitmap_or(unsigned long *dst, const unsigned long *bitmap1, + const unsigned long *bitmap2, int bits) +{ + int k; + int nr = BITS_TO_LONGS(bits); + + for (k = 0; k < nr; k++) + dst[k] = bitmap1[k] | bitmap2[k]; +} diff --git a/tools/perf/util/build-id.c b/tools/perf/util/build-id.c new file mode 100644 index 00000000..5295625c --- /dev/null +++ b/tools/perf/util/build-id.c @@ -0,0 +1,104 @@ +/* + * build-id.c + * + * build-id support + * + * Copyright (C) 2009, 2010 Red Hat Inc. + * Copyright (C) 2009, 2010 Arnaldo Carvalho de Melo <acme@redhat.com> + */ +#include "util.h" +#include <stdio.h> +#include "build-id.h" +#include "event.h" +#include "symbol.h" +#include <linux/kernel.h> +#include "debug.h" +#include "session.h" +#include "tool.h" + +int build_id__mark_dso_hit(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct perf_evsel *evsel __maybe_unused, + struct machine *machine) +{ + struct addr_location al; + u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; + struct thread *thread = machine__findnew_thread(machine, event->ip.pid); + + if (thread == NULL) { + pr_err("problem processing %d event, skipping it.\n", + event->header.type); + return -1; + } + + thread__find_addr_map(thread, machine, cpumode, MAP__FUNCTION, + event->ip.ip, &al); + + if (al.map != NULL) + al.map->dso->hit = 1; + + return 0; +} + +static int perf_event__exit_del_thread(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample + __maybe_unused, + struct machine *machine) +{ + struct thread *thread = machine__findnew_thread(machine, event->fork.tid); + + dump_printf("(%d:%d):(%d:%d)\n", event->fork.pid, event->fork.tid, + event->fork.ppid, event->fork.ptid); + + if (thread) { + rb_erase(&thread->rb_node, &machine->threads); + machine->last_match = NULL; + thread__delete(thread); + } + + return 0; +} + +struct perf_tool build_id__mark_dso_hit_ops = { + .sample = build_id__mark_dso_hit, + .mmap = perf_event__process_mmap, + .fork = perf_event__process_fork, + .exit = perf_event__exit_del_thread, + .attr = perf_event__process_attr, + .build_id = perf_event__process_build_id, +}; + +int build_id__sprintf(const u8 *build_id, int len, char *bf) +{ + char *bid = bf; + const u8 *raw = build_id; + int i; + + for (i = 0; i < len; ++i) { + sprintf(bid, "%02x", *raw); + ++raw; + bid += 2; + } + + return raw - build_id; +} + +char *dso__build_id_filename(struct dso *self, char *bf, size_t size) +{ + char build_id_hex[BUILD_ID_SIZE * 2 + 1]; + + if (!self->has_build_id) + return NULL; + + build_id__sprintf(self->build_id, sizeof(self->build_id), build_id_hex); + if (bf == NULL) { + if (asprintf(&bf, "%s/.build-id/%.2s/%s", buildid_dir, + build_id_hex, build_id_hex + 2) < 0) + return NULL; + } else + snprintf(bf, size, "%s/.build-id/%.2s/%s", buildid_dir, + build_id_hex, build_id_hex + 2); + return bf; +} diff --git a/tools/perf/util/build-id.h b/tools/perf/util/build-id.h new file mode 100644 index 00000000..a811f5c6 --- /dev/null +++ b/tools/perf/util/build-id.h @@ -0,0 +1,19 @@ +#ifndef PERF_BUILD_ID_H_ +#define PERF_BUILD_ID_H_ 1 + +#define BUILD_ID_SIZE 20 + +#include "tool.h" +#include "types.h" + +extern struct perf_tool build_id__mark_dso_hit_ops; +struct dso; + +int build_id__sprintf(const u8 *build_id, int len, char *bf); +char *dso__build_id_filename(struct dso *self, char *bf, size_t size); + +int build_id__mark_dso_hit(struct perf_tool *tool, union perf_event *event, + struct perf_sample *sample, struct perf_evsel *evsel, + struct machine *machine); + +#endif diff --git a/tools/perf/util/cache.h b/tools/perf/util/cache.h new file mode 100644 index 00000000..26e36723 --- /dev/null +++ b/tools/perf/util/cache.h @@ -0,0 +1,77 @@ +#ifndef __PERF_CACHE_H +#define __PERF_CACHE_H + +#include <stdbool.h> +#include "util.h" +#include "strbuf.h" +#include "../perf.h" +#include "../ui/ui.h" + +#define CMD_EXEC_PATH "--exec-path" +#define CMD_PERF_DIR "--perf-dir=" +#define CMD_WORK_TREE "--work-tree=" +#define CMD_DEBUGFS_DIR "--debugfs-dir=" + +#define PERF_DIR_ENVIRONMENT "PERF_DIR" +#define PERF_WORK_TREE_ENVIRONMENT "PERF_WORK_TREE" +#define EXEC_PATH_ENVIRONMENT "PERF_EXEC_PATH" +#define DEFAULT_PERF_DIR_ENVIRONMENT ".perf" +#define PERF_DEBUGFS_ENVIRONMENT "PERF_DEBUGFS_DIR" + +typedef int (*config_fn_t)(const char *, const char *, void *); +extern int perf_default_config(const char *, const char *, void *); +extern int perf_config(config_fn_t fn, void *); +extern int perf_config_int(const char *, const char *); +extern int perf_config_bool(const char *, const char *); +extern int config_error_nonbool(const char *); +extern const char *perf_config_dirname(const char *, 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; + +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_nonrelative_path(const char *path); +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))); + +extern char *perf_pathdup(const char *fmt, ...) + __attribute__((format (printf, 1, 2))); + +#ifndef HAVE_STRLCPY +extern size_t strlcpy(char *dest, const char *src, size_t size); +#endif + +#endif /* __PERF_CACHE_H */ diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c new file mode 100644 index 00000000..42b6a632 --- /dev/null +++ b/tools/perf/util/callchain.c @@ -0,0 +1,463 @@ +/* + * Copyright (C) 2009-2011, Frederic Weisbecker <fweisbec@gmail.com> + * + * Handle the callchains from the stream in an ad-hoc radix tree and then + * sort them in an rbtree. + * + * Using a radix for code path provides a fast retrieval and factorizes + * memory use. Also that lets us use the paths in a hierarchical graph view. + * + */ + +#include <stdlib.h> +#include <stdio.h> +#include <stdbool.h> +#include <errno.h> +#include <math.h> + +#include "util.h" +#include "callchain.h" + +__thread struct callchain_cursor callchain_cursor; + +bool ip_callchain__valid(struct ip_callchain *chain, + const union perf_event *event) +{ + unsigned int chain_size = event->header.size; + chain_size -= (unsigned long)&event->ip.__more_data - (unsigned long)event; + return chain->nr * sizeof(u64) <= chain_size; +} + +#define chain_for_each_child(child, parent) \ + list_for_each_entry(child, &parent->children, siblings) + +#define chain_for_each_child_safe(child, next, parent) \ + list_for_each_entry_safe(child, next, &parent->children, siblings) + +static void +rb_insert_callchain(struct rb_root *root, struct callchain_node *chain, + enum chain_mode mode) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent = NULL; + struct callchain_node *rnode; + u64 chain_cumul = callchain_cumul_hits(chain); + + while (*p) { + u64 rnode_cumul; + + parent = *p; + rnode = rb_entry(parent, struct callchain_node, rb_node); + rnode_cumul = callchain_cumul_hits(rnode); + + switch (mode) { + case CHAIN_FLAT: + if (rnode->hit < chain->hit) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + break; + case CHAIN_GRAPH_ABS: /* Falldown */ + case CHAIN_GRAPH_REL: + if (rnode_cumul < chain_cumul) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + break; + case CHAIN_NONE: + default: + break; + } + } + + rb_link_node(&chain->rb_node, parent, p); + rb_insert_color(&chain->rb_node, root); +} + +static void +__sort_chain_flat(struct rb_root *rb_root, struct callchain_node *node, + u64 min_hit) +{ + struct callchain_node *child; + + chain_for_each_child(child, node) + __sort_chain_flat(rb_root, child, min_hit); + + if (node->hit && node->hit >= min_hit) + rb_insert_callchain(rb_root, node, CHAIN_FLAT); +} + +/* + * Once we get every callchains from the stream, we can now + * sort them by hit + */ +static void +sort_chain_flat(struct rb_root *rb_root, struct callchain_root *root, + u64 min_hit, struct callchain_param *param __maybe_unused) +{ + __sort_chain_flat(rb_root, &root->node, min_hit); +} + +static void __sort_chain_graph_abs(struct callchain_node *node, + u64 min_hit) +{ + struct callchain_node *child; + + node->rb_root = RB_ROOT; + + chain_for_each_child(child, node) { + __sort_chain_graph_abs(child, min_hit); + if (callchain_cumul_hits(child) >= min_hit) + rb_insert_callchain(&node->rb_root, child, + CHAIN_GRAPH_ABS); + } +} + +static void +sort_chain_graph_abs(struct rb_root *rb_root, struct callchain_root *chain_root, + u64 min_hit, struct callchain_param *param __maybe_unused) +{ + __sort_chain_graph_abs(&chain_root->node, min_hit); + rb_root->rb_node = chain_root->node.rb_root.rb_node; +} + +static void __sort_chain_graph_rel(struct callchain_node *node, + double min_percent) +{ + struct callchain_node *child; + u64 min_hit; + + node->rb_root = RB_ROOT; + min_hit = ceil(node->children_hit * min_percent); + + chain_for_each_child(child, node) { + __sort_chain_graph_rel(child, min_percent); + if (callchain_cumul_hits(child) >= min_hit) + rb_insert_callchain(&node->rb_root, child, + CHAIN_GRAPH_REL); + } +} + +static void +sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_root *chain_root, + u64 min_hit __maybe_unused, struct callchain_param *param) +{ + __sort_chain_graph_rel(&chain_root->node, param->min_percent / 100.0); + rb_root->rb_node = chain_root->node.rb_root.rb_node; +} + +int callchain_register_param(struct callchain_param *param) +{ + switch (param->mode) { + case CHAIN_GRAPH_ABS: + param->sort = sort_chain_graph_abs; + break; + case CHAIN_GRAPH_REL: + param->sort = sort_chain_graph_rel; + break; + case CHAIN_FLAT: + param->sort = sort_chain_flat; + break; + case CHAIN_NONE: + default: + return -1; + } + return 0; +} + +/* + * Create a child for a parent. If inherit_children, then the new child + * will become the new parent of it's parent children + */ +static struct callchain_node * +create_child(struct callchain_node *parent, bool inherit_children) +{ + struct callchain_node *new; + + new = zalloc(sizeof(*new)); + if (!new) { + perror("not enough memory to create child for code path tree"); + return NULL; + } + new->parent = parent; + INIT_LIST_HEAD(&new->children); + INIT_LIST_HEAD(&new->val); + + if (inherit_children) { + struct callchain_node *next; + + list_splice(&parent->children, &new->children); + INIT_LIST_HEAD(&parent->children); + + chain_for_each_child(next, new) + next->parent = new; + } + list_add_tail(&new->siblings, &parent->children); + + return new; +} + + +/* + * Fill the node with callchain values + */ +static void +fill_node(struct callchain_node *node, struct callchain_cursor *cursor) +{ + struct callchain_cursor_node *cursor_node; + + node->val_nr = cursor->nr - cursor->pos; + if (!node->val_nr) + pr_warning("Warning: empty node in callchain tree\n"); + + cursor_node = callchain_cursor_current(cursor); + + while (cursor_node) { + struct callchain_list *call; + + call = zalloc(sizeof(*call)); + if (!call) { + perror("not enough memory for the code path tree"); + return; + } + call->ip = cursor_node->ip; + call->ms.sym = cursor_node->sym; + call->ms.map = cursor_node->map; + list_add_tail(&call->list, &node->val); + + callchain_cursor_advance(cursor); + cursor_node = callchain_cursor_current(cursor); + } +} + +static void +add_child(struct callchain_node *parent, + struct callchain_cursor *cursor, + u64 period) +{ + struct callchain_node *new; + + new = create_child(parent, false); + fill_node(new, cursor); + + new->children_hit = 0; + new->hit = period; +} + +/* + * Split the parent in two parts (a new child is created) and + * give a part of its callchain to the created child. + * Then create another child to host the given callchain of new branch + */ +static void +split_add_child(struct callchain_node *parent, + struct callchain_cursor *cursor, + struct callchain_list *to_split, + u64 idx_parents, u64 idx_local, u64 period) +{ + struct callchain_node *new; + struct list_head *old_tail; + unsigned int idx_total = idx_parents + idx_local; + + /* split */ + new = create_child(parent, true); + + /* split the callchain and move a part to the new child */ + old_tail = parent->val.prev; + list_del_range(&to_split->list, old_tail); + new->val.next = &to_split->list; + new->val.prev = old_tail; + to_split->list.prev = &new->val; + old_tail->next = &new->val; + + /* split the hits */ + new->hit = parent->hit; + new->children_hit = parent->children_hit; + parent->children_hit = callchain_cumul_hits(new); + new->val_nr = parent->val_nr - idx_local; + parent->val_nr = idx_local; + + /* create a new child for the new branch if any */ + if (idx_total < cursor->nr) { + parent->hit = 0; + add_child(parent, cursor, period); + parent->children_hit += period; + } else { + parent->hit = period; + } +} + +static int +append_chain(struct callchain_node *root, + struct callchain_cursor *cursor, + u64 period); + +static void +append_chain_children(struct callchain_node *root, + struct callchain_cursor *cursor, + u64 period) +{ + struct callchain_node *rnode; + + /* lookup in childrens */ + chain_for_each_child(rnode, root) { + unsigned int ret = append_chain(rnode, cursor, period); + + if (!ret) + goto inc_children_hit; + } + /* nothing in children, add to the current node */ + add_child(root, cursor, period); + +inc_children_hit: + root->children_hit += period; +} + +static int +append_chain(struct callchain_node *root, + struct callchain_cursor *cursor, + u64 period) +{ + struct callchain_cursor_node *curr_snap = cursor->curr; + struct callchain_list *cnode; + u64 start = cursor->pos; + bool found = false; + u64 matches; + + /* + * Lookup in the current node + * If we have a symbol, then compare the start to match + * anywhere inside a function. + */ + list_for_each_entry(cnode, &root->val, list) { + struct callchain_cursor_node *node; + struct symbol *sym; + + node = callchain_cursor_current(cursor); + if (!node) + break; + + sym = node->sym; + + if (cnode->ms.sym && sym) { + if (cnode->ms.sym->start != sym->start) + break; + } else if (cnode->ip != node->ip) + break; + + if (!found) + found = true; + + callchain_cursor_advance(cursor); + } + + /* matches not, relay on the parent */ + if (!found) { + cursor->curr = curr_snap; + cursor->pos = start; + return -1; + } + + matches = cursor->pos - start; + + /* we match only a part of the node. Split it and add the new chain */ + if (matches < root->val_nr) { + split_add_child(root, cursor, cnode, start, matches, period); + return 0; + } + + /* we match 100% of the path, increment the hit */ + if (matches == root->val_nr && cursor->pos == cursor->nr) { + root->hit += period; + return 0; + } + + /* We match the node and still have a part remaining */ + append_chain_children(root, cursor, period); + + return 0; +} + +int callchain_append(struct callchain_root *root, + struct callchain_cursor *cursor, + u64 period) +{ + if (!cursor->nr) + return 0; + + callchain_cursor_commit(cursor); + + append_chain_children(&root->node, cursor, period); + + if (cursor->nr > root->max_depth) + root->max_depth = cursor->nr; + + return 0; +} + +static int +merge_chain_branch(struct callchain_cursor *cursor, + struct callchain_node *dst, struct callchain_node *src) +{ + struct callchain_cursor_node **old_last = cursor->last; + struct callchain_node *child, *next_child; + struct callchain_list *list, *next_list; + int old_pos = cursor->nr; + int err = 0; + + list_for_each_entry_safe(list, next_list, &src->val, list) { + callchain_cursor_append(cursor, list->ip, + list->ms.map, list->ms.sym); + list_del(&list->list); + free(list); + } + + if (src->hit) { + callchain_cursor_commit(cursor); + append_chain_children(dst, cursor, src->hit); + } + + chain_for_each_child_safe(child, next_child, src) { + err = merge_chain_branch(cursor, dst, child); + if (err) + break; + + list_del(&child->siblings); + free(child); + } + + cursor->nr = old_pos; + cursor->last = old_last; + + return err; +} + +int callchain_merge(struct callchain_cursor *cursor, + struct callchain_root *dst, struct callchain_root *src) +{ + return merge_chain_branch(cursor, &dst->node, &src->node); +} + +int callchain_cursor_append(struct callchain_cursor *cursor, + u64 ip, struct map *map, struct symbol *sym) +{ + struct callchain_cursor_node *node = *cursor->last; + + if (!node) { + node = calloc(1, sizeof(*node)); + if (!node) + return -ENOMEM; + + *cursor->last = node; + } + + node->ip = ip; + node->map = map; + node->sym = sym; + + cursor->nr++; + + cursor->last = &node->next; + + return 0; +} diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h new file mode 100644 index 00000000..3ee9f67d --- /dev/null +++ b/tools/perf/util/callchain.h @@ -0,0 +1,151 @@ +#ifndef __PERF_CALLCHAIN_H +#define __PERF_CALLCHAIN_H + +#include "../perf.h" +#include <linux/list.h> +#include <linux/rbtree.h> +#include "event.h" +#include "symbol.h" + +enum chain_mode { + CHAIN_NONE, + CHAIN_FLAT, + CHAIN_GRAPH_ABS, + CHAIN_GRAPH_REL +}; + +enum chain_order { + ORDER_CALLER, + ORDER_CALLEE +}; + +struct callchain_node { + struct callchain_node *parent; + struct list_head siblings; + struct list_head children; + struct list_head val; + struct rb_node rb_node; /* to sort nodes in an rbtree */ + struct rb_root rb_root; /* sorted tree of children */ + unsigned int val_nr; + u64 hit; + u64 children_hit; +}; + +struct callchain_root { + u64 max_depth; + struct callchain_node node; +}; + +struct callchain_param; + +typedef void (*sort_chain_func_t)(struct rb_root *, struct callchain_root *, + u64, struct callchain_param *); + +struct callchain_param { + enum chain_mode mode; + u32 print_limit; + double min_percent; + sort_chain_func_t sort; + enum chain_order order; +}; + +struct callchain_list { + u64 ip; + struct map_symbol ms; + struct list_head list; +}; + +/* + * A callchain cursor is a single linked list that + * let one feed a callchain progressively. + * It keeps persistent allocated entries to minimize + * allocations. + */ +struct callchain_cursor_node { + u64 ip; + struct map *map; + struct symbol *sym; + struct callchain_cursor_node *next; +}; + +struct callchain_cursor { + u64 nr; + struct callchain_cursor_node *first; + struct callchain_cursor_node **last; + u64 pos; + struct callchain_cursor_node *curr; +}; + +extern __thread struct callchain_cursor callchain_cursor; + +static inline void callchain_init(struct callchain_root *root) +{ + INIT_LIST_HEAD(&root->node.siblings); + INIT_LIST_HEAD(&root->node.children); + INIT_LIST_HEAD(&root->node.val); + + root->node.parent = NULL; + root->node.hit = 0; + root->node.children_hit = 0; + root->max_depth = 0; +} + +static inline u64 callchain_cumul_hits(struct callchain_node *node) +{ + return node->hit + node->children_hit; +} + +int callchain_register_param(struct callchain_param *param); +int callchain_append(struct callchain_root *root, + struct callchain_cursor *cursor, + u64 period); + +int callchain_merge(struct callchain_cursor *cursor, + struct callchain_root *dst, struct callchain_root *src); + +struct ip_callchain; +union perf_event; + +bool ip_callchain__valid(struct ip_callchain *chain, + const union perf_event *event); +/* + * Initialize a cursor before adding entries inside, but keep + * the previously allocated entries as a cache. + */ +static inline void callchain_cursor_reset(struct callchain_cursor *cursor) +{ + cursor->nr = 0; + cursor->last = &cursor->first; +} + +int callchain_cursor_append(struct callchain_cursor *cursor, u64 ip, + struct map *map, struct symbol *sym); + +/* Close a cursor writing session. Initialize for the reader */ +static inline void callchain_cursor_commit(struct callchain_cursor *cursor) +{ + cursor->curr = cursor->first; + cursor->pos = 0; +} + +/* Cursor reading iteration helpers */ +static inline struct callchain_cursor_node * +callchain_cursor_current(struct callchain_cursor *cursor) +{ + if (cursor->pos == cursor->nr) + return NULL; + + return cursor->curr; +} + +static inline void callchain_cursor_advance(struct callchain_cursor *cursor) +{ + cursor->curr = cursor->curr->next; + cursor->pos++; +} + +struct option; + +int record_parse_callchain_opt(const struct option *opt, const char *arg, int unset); +extern const char record_callchain_help[]; +#endif /* __PERF_CALLCHAIN_H */ diff --git a/tools/perf/util/cgroup.c b/tools/perf/util/cgroup.c new file mode 100644 index 00000000..96bbda1d --- /dev/null +++ b/tools/perf/util/cgroup.c @@ -0,0 +1,177 @@ +#include "util.h" +#include "../perf.h" +#include "parse-options.h" +#include "evsel.h" +#include "cgroup.h" +#include "evlist.h" + +int nr_cgroups; + +static int +cgroupfs_find_mountpoint(char *buf, size_t maxlen) +{ + FILE *fp; + char mountpoint[PATH_MAX + 1], tokens[PATH_MAX + 1], type[PATH_MAX + 1]; + char *token, *saved_ptr = NULL; + int found = 0; + + fp = fopen("/proc/mounts", "r"); + if (!fp) + return -1; + + /* + * in order to handle split hierarchy, we need to scan /proc/mounts + * and inspect every cgroupfs mount point to find one that has + * perf_event subsystem + */ + while (fscanf(fp, "%*s %"STR(PATH_MAX)"s %"STR(PATH_MAX)"s %" + STR(PATH_MAX)"s %*d %*d\n", + mountpoint, type, tokens) == 3) { + + if (!strcmp(type, "cgroup")) { + + token = strtok_r(tokens, ",", &saved_ptr); + + while (token != NULL) { + if (!strcmp(token, "perf_event")) { + found = 1; + break; + } + token = strtok_r(NULL, ",", &saved_ptr); + } + } + if (found) + break; + } + fclose(fp); + if (!found) + return -1; + + if (strlen(mountpoint) < maxlen) { + strcpy(buf, mountpoint); + return 0; + } + return -1; +} + +static int open_cgroup(char *name) +{ + char path[PATH_MAX + 1]; + char mnt[PATH_MAX + 1]; + int fd; + + + if (cgroupfs_find_mountpoint(mnt, PATH_MAX + 1)) + return -1; + + snprintf(path, PATH_MAX, "%s/%s", mnt, name); + + fd = open(path, O_RDONLY); + if (fd == -1) + fprintf(stderr, "no access to cgroup %s\n", path); + + return fd; +} + +static int add_cgroup(struct perf_evlist *evlist, char *str) +{ + struct perf_evsel *counter; + struct cgroup_sel *cgrp = NULL; + int n; + /* + * check if cgrp is already defined, if so we reuse it + */ + list_for_each_entry(counter, &evlist->entries, node) { + cgrp = counter->cgrp; + if (!cgrp) + continue; + if (!strcmp(cgrp->name, str)) + break; + + cgrp = NULL; + } + + if (!cgrp) { + cgrp = zalloc(sizeof(*cgrp)); + if (!cgrp) + return -1; + + cgrp->name = str; + + cgrp->fd = open_cgroup(str); + if (cgrp->fd == -1) { + free(cgrp); + return -1; + } + } + + /* + * find corresponding event + * if add cgroup N, then need to find event N + */ + n = 0; + list_for_each_entry(counter, &evlist->entries, node) { + if (n == nr_cgroups) + goto found; + n++; + } + if (cgrp->refcnt == 0) + free(cgrp); + + return -1; +found: + cgrp->refcnt++; + counter->cgrp = cgrp; + return 0; +} + +void close_cgroup(struct cgroup_sel *cgrp) +{ + if (!cgrp) + return; + + /* XXX: not reentrant */ + if (--cgrp->refcnt == 0) { + close(cgrp->fd); + free(cgrp->name); + free(cgrp); + } +} + +int parse_cgroups(const struct option *opt __maybe_unused, const char *str, + int unset __maybe_unused) +{ + struct perf_evlist *evlist = *(struct perf_evlist **)opt->value; + const char *p, *e, *eos = str + strlen(str); + char *s; + int ret; + + if (list_empty(&evlist->entries)) { + fprintf(stderr, "must define events before cgroups\n"); + return -1; + } + + for (;;) { + p = strchr(str, ','); + e = p ? p : eos; + + /* allow empty cgroups, i.e., skip */ + if (e - str) { + /* termination added */ + s = strndup(str, e - str); + if (!s) + return -1; + ret = add_cgroup(evlist, s); + if (ret) { + free(s); + return -1; + } + } + /* nr_cgroups is increased een for empty cgroups */ + nr_cgroups++; + if (!p) + break; + str = p+1; + } + return 0; +} diff --git a/tools/perf/util/cgroup.h b/tools/perf/util/cgroup.h new file mode 100644 index 00000000..89acd6de --- /dev/null +++ b/tools/perf/util/cgroup.h @@ -0,0 +1,17 @@ +#ifndef __CGROUP_H__ +#define __CGROUP_H__ + +struct option; + +struct cgroup_sel { + char *name; + int fd; + int refcnt; +}; + + +extern int nr_cgroups; /* number of explicit cgroups defined */ +extern void close_cgroup(struct cgroup_sel *cgrp); +extern int parse_cgroups(const struct option *opt, const char *str, int unset); + +#endif /* __CGROUP_H__ */ diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c new file mode 100644 index 00000000..11e46da1 --- /dev/null +++ b/tools/perf/util/color.c @@ -0,0 +1,325 @@ +#include <linux/kernel.h> +#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 < (int)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" + }; + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(attr_names); i++) { + const char *str = attr_names[i]; + if (!strncasecmp(name, str, len) && !str[len]) + 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_vsnprintf(char *bf, size_t size, const char *color, + const char *fmt, va_list args, const char *trail) +{ + int r = 0; + + /* + * Auto-detect: + */ + if (perf_use_color_default < 0) { + if (isatty(1) || pager_in_use()) + perf_use_color_default = 1; + else + perf_use_color_default = 0; + } + + if (perf_use_color_default && *color) + r += scnprintf(bf, size, "%s", color); + r += vscnprintf(bf + r, size - r, fmt, args); + if (perf_use_color_default && *color) + r += scnprintf(bf + r, size - r, "%s", PERF_COLOR_RESET); + if (trail) + r += scnprintf(bf + r, size - r, "%s", trail); + return r; +} + +static int __color_vfprintf(FILE *fp, const char *color, const char *fmt, + va_list args, const char *trail) +{ + int r = 0; + + /* + * Auto-detect: + */ + if (perf_use_color_default < 0) { + if (isatty(fileno(fp)) || 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_vsnprintf(char *bf, size_t size, const char *color, + const char *fmt, va_list args) +{ + return __color_vsnprintf(bf, size, color, fmt, args, NULL); +} + +int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args) +{ + return __color_vfprintf(fp, color, fmt, args, NULL); +} + +int color_snprintf(char *bf, size_t size, const char *color, + const char *fmt, ...) +{ + va_list args; + int r; + + va_start(args, fmt); + r = color_vsnprintf(bf, size, color, fmt, args); + va_end(args); + return r; +} + +int color_fprintf(FILE *fp, const char *color, const char *fmt, ...) +{ + va_list args; + int r; + + va_start(args, fmt); + r = color_vfprintf(fp, color, fmt, args); + 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 ? (size_t)(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; +} + +const char *get_percent_color(double percent) +{ + const char *color = PERF_COLOR_NORMAL; + + /* + * We color high-overhead entries in red, mid-overhead + * entries in green - and keep the low overhead places + * normal: + */ + if (percent >= MIN_RED) + color = PERF_COLOR_RED; + else { + if (percent > MIN_GREEN) + color = PERF_COLOR_GREEN; + } + return color; +} + +int percent_color_fprintf(FILE *fp, const char *fmt, double percent) +{ + int r; + const char *color; + + color = get_percent_color(percent); + r = color_fprintf(fp, color, fmt, percent); + + return r; +} + +int percent_color_snprintf(char *bf, size_t size, const char *fmt, double percent) +{ + const char *color = get_percent_color(percent); + return color_snprintf(bf, size, color, fmt, percent); +} diff --git a/tools/perf/util/color.h b/tools/perf/util/color.h new file mode 100644 index 00000000..dea082b7 --- /dev/null +++ b/tools/perf/util/color.h @@ -0,0 +1,46 @@ +#ifndef __PERF_COLOR_H +#define __PERF_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" + +#define MIN_GREEN 0.5 +#define MIN_RED 5.0 + +/* + * 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_vsnprintf(char *bf, size_t size, const char *color, + const char *fmt, va_list args); +int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args); +int color_fprintf(FILE *fp, const char *color, const char *fmt, ...); +int color_snprintf(char *bf, size_t size, const char *color, const char *fmt, ...); +int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...); +int color_fwrite_lines(FILE *fp, const char *color, size_t count, const char *buf); +int percent_color_snprintf(char *bf, size_t size, const char *fmt, double percent); +int percent_color_fprintf(FILE *fp, const char *fmt, double percent); +const char *get_percent_color(double percent); + +#endif /* __PERF_COLOR_H */ diff --git a/tools/perf/util/config.c b/tools/perf/util/config.c new file mode 100644 index 00000000..3e0fdd36 --- /dev/null +++ b/tools/perf/util/config.c @@ -0,0 +1,509 @@ +/* + * config.c + * + * Helper functions for parsing config items. + * Originally copied from GIT source. + * + * Copyright (C) Linus Torvalds, 2005 + * Copyright (C) Johannes Schindelin, 2005 + * + */ +#include "util.h" +#include "cache.h" +#include "exec_cmd.h" + +#define MAXNAME (256) + +#define DEBUG_CACHE_DIR ".debug" + + +char buildid_dir[MAXPATHLEN]; /* root dir for buildid, binary cache */ + +static FILE *config_file; +static const char *config_file_name; +static int config_linenr; +static int config_file_eof; + +static const char *config_exclusive_filename; + +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, space = 0; + size_t len = 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 == '-' || 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++] = 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 ch = get_next_char(); + + if (ch == '\n') + return -1; + if (ch == '"') + break; + if (ch == '\\') { + ch = get_next_char(); + if (ch == '\n') + return -1; + } + name[baselen++] = ch; + 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; +} + +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; +} + +static 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); +} + +const char *perf_config_dirname(const char *name, const char *value) +{ + if (!name) + return NULL; + return value; +} + +static int perf_default_core_config(const char *var __maybe_unused, + const char *value __maybe_unused) +{ + /* Add other config variables here. */ + return 0; +} + +int perf_default_config(const char *var, const char *value, + void *dummy __maybe_unused) +{ + if (!prefixcmp(var, "core.")) + return perf_default_core_config(var, value); + + /* Add other config variables here. */ + return 0; +} + +static 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; +} + +static 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; +} + +static int perf_config_system(void) +{ + return !perf_env_bool("PERF_CONFIG_NOSYSTEM", 0); +} + +static 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; + 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)); + struct stat st; + + if (user_config == NULL) { + warning("Not enough memory to process %s/.perfconfig, " + "ignoring it.", home); + goto out; + } + + if (stat(user_config, &st) < 0) + goto out_free; + + if (st.st_uid && (st.st_uid != geteuid())) { + warning("File %s not owned by current user or root, " + "ignoring it.", user_config); + goto out_free; + } + + if (!st.st_size) + goto out_free; + + ret += perf_config_from_file(fn, user_config, data); + found += 1; +out_free: + free(user_config); + } +out: + if (found == 0) + return -1; + return ret; +} + +/* + * 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); +} + +struct buildid_dir_config { + char *dir; +}; + +static int buildid_dir_command_config(const char *var, const char *value, + void *data) +{ + struct buildid_dir_config *c = data; + const char *v; + + /* same dir for all commands */ + if (!prefixcmp(var, "buildid.") && !strcmp(var + 8, "dir")) { + v = perf_config_dirname(var, value); + if (!v) + return -1; + strncpy(c->dir, v, MAXPATHLEN-1); + c->dir[MAXPATHLEN-1] = '\0'; + } + return 0; +} + +static void check_buildid_dir_config(void) +{ + struct buildid_dir_config c; + c.dir = buildid_dir; + perf_config(buildid_dir_command_config, &c); +} + +void set_buildid_dir(void) +{ + buildid_dir[0] = '\0'; + + /* try config file */ + check_buildid_dir_config(); + + /* default to $HOME/.debug */ + if (buildid_dir[0] == '\0') { + char *v = getenv("HOME"); + if (v) { + snprintf(buildid_dir, MAXPATHLEN-1, "%s/%s", + v, DEBUG_CACHE_DIR); + } else { + strncpy(buildid_dir, DEBUG_CACHE_DIR, MAXPATHLEN-1); + } + buildid_dir[MAXPATHLEN-1] = '\0'; + } + /* for communicating with external commands */ + setenv("PERF_BUILDID_DIR", buildid_dir, 1); +} diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c new file mode 100644 index 00000000..f817046e --- /dev/null +++ b/tools/perf/util/cpumap.c @@ -0,0 +1,257 @@ +#include "util.h" +#include "sysfs.h" +#include "../perf.h" +#include "cpumap.h" +#include <assert.h> +#include <stdio.h> + +static struct cpu_map *cpu_map__default_new(void) +{ + struct cpu_map *cpus; + int nr_cpus; + + nr_cpus = sysconf(_SC_NPROCESSORS_ONLN); + if (nr_cpus < 0) + return NULL; + + cpus = malloc(sizeof(*cpus) + nr_cpus * sizeof(int)); + if (cpus != NULL) { + int i; + for (i = 0; i < nr_cpus; ++i) + cpus->map[i] = i; + + cpus->nr = nr_cpus; + } + + return cpus; +} + +static struct cpu_map *cpu_map__trim_new(int nr_cpus, int *tmp_cpus) +{ + size_t payload_size = nr_cpus * sizeof(int); + struct cpu_map *cpus = malloc(sizeof(*cpus) + payload_size); + + if (cpus != NULL) { + cpus->nr = nr_cpus; + memcpy(cpus->map, tmp_cpus, payload_size); + } + + return cpus; +} + +struct cpu_map *cpu_map__read(FILE *file) +{ + struct cpu_map *cpus = NULL; + int nr_cpus = 0; + int *tmp_cpus = NULL, *tmp; + int max_entries = 0; + int n, cpu, prev; + char sep; + + sep = 0; + prev = -1; + for (;;) { + n = fscanf(file, "%u%c", &cpu, &sep); + if (n <= 0) + break; + if (prev >= 0) { + int new_max = nr_cpus + cpu - prev - 1; + + if (new_max >= max_entries) { + max_entries = new_max + MAX_NR_CPUS / 2; + tmp = realloc(tmp_cpus, max_entries * sizeof(int)); + if (tmp == NULL) + goto out_free_tmp; + tmp_cpus = tmp; + } + + while (++prev < cpu) + tmp_cpus[nr_cpus++] = prev; + } + if (nr_cpus == max_entries) { + max_entries += MAX_NR_CPUS; + tmp = realloc(tmp_cpus, max_entries * sizeof(int)); + if (tmp == NULL) + goto out_free_tmp; + tmp_cpus = tmp; + } + + tmp_cpus[nr_cpus++] = cpu; + if (n == 2 && sep == '-') + prev = cpu; + else + prev = -1; + if (n == 1 || sep == '\n') + break; + } + + if (nr_cpus > 0) + cpus = cpu_map__trim_new(nr_cpus, tmp_cpus); + else + cpus = cpu_map__default_new(); +out_free_tmp: + free(tmp_cpus); + return cpus; +} + +static struct cpu_map *cpu_map__read_all_cpu_map(void) +{ + struct cpu_map *cpus = NULL; + FILE *onlnf; + + onlnf = fopen("/sys/devices/system/cpu/online", "r"); + if (!onlnf) + return cpu_map__default_new(); + + cpus = cpu_map__read(onlnf); + fclose(onlnf); + return cpus; +} + +struct cpu_map *cpu_map__new(const char *cpu_list) +{ + struct cpu_map *cpus = NULL; + unsigned long start_cpu, end_cpu = 0; + char *p = NULL; + int i, nr_cpus = 0; + int *tmp_cpus = NULL, *tmp; + int max_entries = 0; + + if (!cpu_list) + return cpu_map__read_all_cpu_map(); + + if (!isdigit(*cpu_list)) + goto out; + + while (isdigit(*cpu_list)) { + p = NULL; + start_cpu = strtoul(cpu_list, &p, 0); + if (start_cpu >= INT_MAX + || (*p != '\0' && *p != ',' && *p != '-')) + goto invalid; + + if (*p == '-') { + cpu_list = ++p; + p = NULL; + end_cpu = strtoul(cpu_list, &p, 0); + + if (end_cpu >= INT_MAX || (*p != '\0' && *p != ',')) + goto invalid; + + if (end_cpu < start_cpu) + goto invalid; + } else { + end_cpu = start_cpu; + } + + for (; start_cpu <= end_cpu; start_cpu++) { + /* check for duplicates */ + for (i = 0; i < nr_cpus; i++) + if (tmp_cpus[i] == (int)start_cpu) + goto invalid; + + if (nr_cpus == max_entries) { + max_entries += MAX_NR_CPUS; + tmp = realloc(tmp_cpus, max_entries * sizeof(int)); + if (tmp == NULL) + goto invalid; + tmp_cpus = tmp; + } + tmp_cpus[nr_cpus++] = (int)start_cpu; + } + if (*p) + ++p; + + cpu_list = p; + } + + if (nr_cpus > 0) + cpus = cpu_map__trim_new(nr_cpus, tmp_cpus); + else + cpus = cpu_map__default_new(); +invalid: + free(tmp_cpus); +out: + return cpus; +} + +size_t cpu_map__fprintf(struct cpu_map *map, FILE *fp) +{ + int i; + size_t printed = fprintf(fp, "%d cpu%s: ", + map->nr, map->nr > 1 ? "s" : ""); + for (i = 0; i < map->nr; ++i) + printed += fprintf(fp, "%s%d", i ? ", " : "", map->map[i]); + + return printed + fprintf(fp, "\n"); +} + +struct cpu_map *cpu_map__dummy_new(void) +{ + struct cpu_map *cpus = malloc(sizeof(*cpus) + sizeof(int)); + + if (cpus != NULL) { + cpus->nr = 1; + cpus->map[0] = -1; + } + + return cpus; +} + +void cpu_map__delete(struct cpu_map *map) +{ + free(map); +} + +int cpu_map__get_socket(struct cpu_map *map, int idx) +{ + FILE *fp; + const char *mnt; + char path[PATH_MAX]; + int cpu, ret; + + if (idx > map->nr) + return -1; + + cpu = map->map[idx]; + + mnt = sysfs_find_mountpoint(); + if (!mnt) + return -1; + + sprintf(path, + "%s/devices/system/cpu/cpu%d/topology/physical_package_id", + mnt, cpu); + + fp = fopen(path, "r"); + if (!fp) + return -1; + ret = fscanf(fp, "%d", &cpu); + fclose(fp); + return ret == 1 ? cpu : -1; +} + +int cpu_map__build_socket_map(struct cpu_map *cpus, struct cpu_map **sockp) +{ + struct cpu_map *sock; + int nr = cpus->nr; + int cpu, s1, s2; + + sock = calloc(1, sizeof(*sock) + nr * sizeof(int)); + if (!sock) + return -1; + + for (cpu = 0; cpu < nr; cpu++) { + s1 = cpu_map__get_socket(cpus, cpu); + for (s2 = 0; s2 < sock->nr; s2++) { + if (s1 == sock->map[s2]) + break; + } + if (s2 == sock->nr) { + sock->map[sock->nr] = s1; + sock->nr++; + } + } + *sockp = sock; + return 0; +} diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h new file mode 100644 index 00000000..161b0075 --- /dev/null +++ b/tools/perf/util/cpumap.h @@ -0,0 +1,37 @@ +#ifndef __PERF_CPUMAP_H +#define __PERF_CPUMAP_H + +#include <stdio.h> +#include <stdbool.h> + +struct cpu_map { + int nr; + int map[]; +}; + +struct cpu_map *cpu_map__new(const char *cpu_list); +struct cpu_map *cpu_map__dummy_new(void); +void cpu_map__delete(struct cpu_map *map); +struct cpu_map *cpu_map__read(FILE *file); +size_t cpu_map__fprintf(struct cpu_map *map, FILE *fp); +int cpu_map__get_socket(struct cpu_map *map, int idx); +int cpu_map__build_socket_map(struct cpu_map *cpus, struct cpu_map **sockp); + +static inline int cpu_map__socket(struct cpu_map *sock, int s) +{ + if (!sock || s > sock->nr || s < 0) + return 0; + return sock->map[s]; +} + +static inline int cpu_map__nr(const struct cpu_map *map) +{ + return map ? map->nr : 1; +} + +static inline bool cpu_map__all(const struct cpu_map *map) +{ + return map ? map->map[0] == -1 : true; +} + +#endif /* __PERF_CPUMAP_H */ diff --git a/tools/perf/util/ctype.c b/tools/perf/util/ctype.c new file mode 100644 index 00000000..aada3ac5 --- /dev/null +++ b/tools/perf/util/ctype.c @@ -0,0 +1,39 @@ +/* + * Sane locale-independent, ASCII ctype. + * + * No surprises, and works with signed and unsigned chars. + */ +#include "util.h" + +enum { + S = GIT_SPACE, + A = GIT_ALPHA, + D = GIT_DIGIT, + G = GIT_GLOB_SPECIAL, /* *, ?, [, \\ */ + R = GIT_REGEX_SPECIAL, /* $, (, ), +, ., ^, {, | * */ + P = GIT_PRINT_EXTRA, /* printable - alpha - digit - glob - regex */ + + PS = GIT_SPACE | GIT_PRINT_EXTRA, +}; + +unsigned char sane_ctype[256] = { +/* 0 1 2 3 4 5 6 7 8 9 A B C D E F */ + + 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 */ + PS,P, P, P, R, P, P, P, R, R, G, R, P, P, R, P, /* 32.. 47 */ + D, D, D, D, D, D, D, D, D, D, P, P, P, P, P, G, /* 48.. 63 */ + P, 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, P, R, P, /* 80.. 95 */ + P, 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, P, P, 0, /* 112..127 */ + /* Nothing in the 128.. range */ +}; + +const char *graph_line = + "_____________________________________________________________________" + "_____________________________________________________________________"; +const char *graph_dotted_line = + "---------------------------------------------------------------------" + "---------------------------------------------------------------------" + "---------------------------------------------------------------------"; diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c new file mode 100644 index 00000000..399e74c3 --- /dev/null +++ b/tools/perf/util/debug.c @@ -0,0 +1,84 @@ +/* For general debugging purposes */ + +#include "../perf.h" + +#include <string.h> +#include <stdarg.h> +#include <stdio.h> + +#include "cache.h" +#include "color.h" +#include "event.h" +#include "debug.h" +#include "util.h" +#include "target.h" + +int verbose; +bool dump_trace = false, quiet = false; + +int eprintf(int level, const char *fmt, ...) +{ + va_list args; + int ret = 0; + + if (verbose >= level) { + va_start(args, fmt); + if (use_browser >= 1) + ui_helpline__vshow(fmt, args); + else + ret = vfprintf(stderr, fmt, args); + va_end(args); + } + + return ret; +} + +int dump_printf(const char *fmt, ...) +{ + va_list args; + int ret = 0; + + if (dump_trace) { + va_start(args, fmt); + ret = vprintf(fmt, args); + va_end(args); + } + + return ret; +} + +void trace_event(union perf_event *event) +{ + unsigned char *raw_event = (void *)event; + const char *color = PERF_COLOR_BLUE; + int i, j; + + if (!dump_trace) + return; + + printf("."); + color_fprintf(stdout, color, "\n. ... raw event: size %d bytes\n", + event->header.size); + + for (i = 0; i < event->header.size; i++) { + if ((i & 15) == 0) { + printf("."); + color_fprintf(stdout, color, " %04x: ", i); + } + + color_fprintf(stdout, color, " %02x", raw_event[i]); + + if (((i & 15) == 15) || i == event->header.size-1) { + color_fprintf(stdout, color, " "); + for (j = 0; j < 15-(i & 15); j++) + color_fprintf(stdout, color, " "); + for (j = i & ~15; j <= i; j++) { + color_fprintf(stdout, color, "%c", + isprint(raw_event[j]) ? + raw_event[j] : '.'); + } + color_fprintf(stdout, color, "\n"); + } + } + printf(".\n"); +} diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h new file mode 100644 index 00000000..efbd9880 --- /dev/null +++ b/tools/perf/util/debug.h @@ -0,0 +1,20 @@ +/* For debugging general purposes */ +#ifndef __PERF_DEBUG_H +#define __PERF_DEBUG_H + +#include <stdbool.h> +#include "event.h" +#include "../ui/helpline.h" +#include "../ui/progress.h" +#include "../ui/util.h" + +extern int verbose; +extern bool quiet, dump_trace; + +int dump_printf(const char *fmt, ...) __attribute__((format(printf, 1, 2))); +void trace_event(union perf_event *event); + +int ui__error(const char *format, ...) __attribute__((format(printf, 1, 2))); +int ui__warning(const char *format, ...) __attribute__((format(printf, 1, 2))); + +#endif /* __PERF_DEBUG_H */ diff --git a/tools/perf/util/debugfs.c b/tools/perf/util/debugfs.c new file mode 100644 index 00000000..dd8b1931 --- /dev/null +++ b/tools/perf/util/debugfs.c @@ -0,0 +1,114 @@ +#include "util.h" +#include "debugfs.h" +#include "cache.h" + +#include <linux/kernel.h> +#include <sys/mount.h> + +static int debugfs_premounted; +char debugfs_mountpoint[PATH_MAX + 1] = "/sys/kernel/debug"; +char tracing_events_path[PATH_MAX + 1] = "/sys/kernel/debug/tracing/events"; + +static const char *debugfs_known_mountpoints[] = { + "/sys/kernel/debug/", + "/debug/", + 0, +}; + +static int debugfs_found; + +/* find the path to the mounted debugfs */ +const char *debugfs_find_mountpoint(void) +{ + const char **ptr; + char type[100]; + FILE *fp; + + if (debugfs_found) + return (const char *) debugfs_mountpoint; + + ptr = debugfs_known_mountpoints; + while (*ptr) { + if (debugfs_valid_mountpoint(*ptr) == 0) { + debugfs_found = 1; + strcpy(debugfs_mountpoint, *ptr); + return debugfs_mountpoint; + } + ptr++; + } + + /* give up and parse /proc/mounts */ + fp = fopen("/proc/mounts", "r"); + if (fp == NULL) + return NULL; + + while (fscanf(fp, "%*s %" STR(PATH_MAX) "s %99s %*s %*d %*d\n", + debugfs_mountpoint, type) == 2) { + if (strcmp(type, "debugfs") == 0) + break; + } + fclose(fp); + + if (strcmp(type, "debugfs") != 0) + return NULL; + + debugfs_found = 1; + + return debugfs_mountpoint; +} + +/* verify that a mountpoint is actually a debugfs instance */ + +int debugfs_valid_mountpoint(const char *debugfs) +{ + struct statfs st_fs; + + if (statfs(debugfs, &st_fs) < 0) + return -ENOENT; + else if (st_fs.f_type != (long) DEBUGFS_MAGIC) + return -ENOENT; + + return 0; +} + +static void debugfs_set_tracing_events_path(const char *mountpoint) +{ + snprintf(tracing_events_path, sizeof(tracing_events_path), "%s/%s", + mountpoint, "tracing/events"); +} + +/* mount the debugfs somewhere if it's not mounted */ + +char *debugfs_mount(const char *mountpoint) +{ + /* see if it's already mounted */ + if (debugfs_find_mountpoint()) { + debugfs_premounted = 1; + goto out; + } + + /* if not mounted and no argument */ + if (mountpoint == NULL) { + /* see if environment variable set */ + mountpoint = getenv(PERF_DEBUGFS_ENVIRONMENT); + /* if no environment variable, use default */ + if (mountpoint == NULL) + mountpoint = "/sys/kernel/debug"; + } + + if (mount(NULL, mountpoint, "debugfs", 0, NULL) < 0) + return NULL; + + /* save the mountpoint */ + debugfs_found = 1; + strncpy(debugfs_mountpoint, mountpoint, sizeof(debugfs_mountpoint)); +out: + debugfs_set_tracing_events_path(debugfs_mountpoint); + return debugfs_mountpoint; +} + +void debugfs_set_path(const char *mountpoint) +{ + snprintf(debugfs_mountpoint, sizeof(debugfs_mountpoint), "%s", mountpoint); + debugfs_set_tracing_events_path(mountpoint); +} diff --git a/tools/perf/util/debugfs.h b/tools/perf/util/debugfs.h new file mode 100644 index 00000000..68f3e87e --- /dev/null +++ b/tools/perf/util/debugfs.h @@ -0,0 +1,12 @@ +#ifndef __DEBUGFS_H__ +#define __DEBUGFS_H__ + +const char *debugfs_find_mountpoint(void); +int debugfs_valid_mountpoint(const char *debugfs); +char *debugfs_mount(const char *mountpoint); +void debugfs_set_path(const char *mountpoint); + +extern char debugfs_mountpoint[]; +extern char tracing_events_path[]; + +#endif /* __DEBUGFS_H__ */ diff --git a/tools/perf/util/dso.c b/tools/perf/util/dso.c new file mode 100644 index 00000000..6f7d5a9d --- /dev/null +++ b/tools/perf/util/dso.c @@ -0,0 +1,595 @@ +#include "symbol.h" +#include "dso.h" +#include "machine.h" +#include "util.h" +#include "debug.h" + +char dso__symtab_origin(const struct dso *dso) +{ + static const char origin[] = { + [DSO_BINARY_TYPE__KALLSYMS] = 'k', + [DSO_BINARY_TYPE__VMLINUX] = 'v', + [DSO_BINARY_TYPE__JAVA_JIT] = 'j', + [DSO_BINARY_TYPE__DEBUGLINK] = 'l', + [DSO_BINARY_TYPE__BUILD_ID_CACHE] = 'B', + [DSO_BINARY_TYPE__FEDORA_DEBUGINFO] = 'f', + [DSO_BINARY_TYPE__UBUNTU_DEBUGINFO] = 'u', + [DSO_BINARY_TYPE__BUILDID_DEBUGINFO] = 'b', + [DSO_BINARY_TYPE__SYSTEM_PATH_DSO] = 'd', + [DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE] = 'K', + [DSO_BINARY_TYPE__GUEST_KALLSYMS] = 'g', + [DSO_BINARY_TYPE__GUEST_KMODULE] = 'G', + [DSO_BINARY_TYPE__GUEST_VMLINUX] = 'V', + }; + + if (dso == NULL || dso->symtab_type == DSO_BINARY_TYPE__NOT_FOUND) + return '!'; + return origin[dso->symtab_type]; +} + +int dso__binary_type_file(struct dso *dso, enum dso_binary_type type, + char *root_dir, char *file, size_t size) +{ + char build_id_hex[BUILD_ID_SIZE * 2 + 1]; + int ret = 0; + + switch (type) { + case DSO_BINARY_TYPE__DEBUGLINK: { + char *debuglink; + + strncpy(file, dso->long_name, size); + debuglink = file + dso->long_name_len; + while (debuglink != file && *debuglink != '/') + debuglink--; + if (*debuglink == '/') + debuglink++; + filename__read_debuglink(dso->long_name, debuglink, + size - (debuglink - file)); + } + break; + case DSO_BINARY_TYPE__BUILD_ID_CACHE: + /* skip the locally configured cache if a symfs is given */ + if (symbol_conf.symfs[0] || + (dso__build_id_filename(dso, file, size) == NULL)) + ret = -1; + break; + + case DSO_BINARY_TYPE__FEDORA_DEBUGINFO: + snprintf(file, size, "%s/usr/lib/debug%s.debug", + symbol_conf.symfs, dso->long_name); + break; + + case DSO_BINARY_TYPE__UBUNTU_DEBUGINFO: + snprintf(file, size, "%s/usr/lib/debug%s", + symbol_conf.symfs, dso->long_name); + break; + + case DSO_BINARY_TYPE__BUILDID_DEBUGINFO: + if (!dso->has_build_id) { + ret = -1; + break; + } + + build_id__sprintf(dso->build_id, + sizeof(dso->build_id), + build_id_hex); + snprintf(file, size, + "%s/usr/lib/debug/.build-id/%.2s/%s.debug", + symbol_conf.symfs, build_id_hex, build_id_hex + 2); + break; + + case DSO_BINARY_TYPE__SYSTEM_PATH_DSO: + snprintf(file, size, "%s%s", + symbol_conf.symfs, dso->long_name); + break; + + case DSO_BINARY_TYPE__GUEST_KMODULE: + snprintf(file, size, "%s%s%s", symbol_conf.symfs, + root_dir, dso->long_name); + break; + + case DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE: + snprintf(file, size, "%s%s", symbol_conf.symfs, + dso->long_name); + break; + + default: + case DSO_BINARY_TYPE__KALLSYMS: + case DSO_BINARY_TYPE__VMLINUX: + case DSO_BINARY_TYPE__GUEST_KALLSYMS: + case DSO_BINARY_TYPE__GUEST_VMLINUX: + case DSO_BINARY_TYPE__JAVA_JIT: + case DSO_BINARY_TYPE__NOT_FOUND: + ret = -1; + break; + } + + return ret; +} + +static int open_dso(struct dso *dso, struct machine *machine) +{ + char *root_dir = (char *) ""; + char *name; + int fd; + + name = malloc(PATH_MAX); + if (!name) + return -ENOMEM; + + if (machine) + root_dir = machine->root_dir; + + if (dso__binary_type_file(dso, dso->data_type, + root_dir, name, PATH_MAX)) { + free(name); + return -EINVAL; + } + + fd = open(name, O_RDONLY); + free(name); + return fd; +} + +int dso__data_fd(struct dso *dso, struct machine *machine) +{ + static enum dso_binary_type binary_type_data[] = { + DSO_BINARY_TYPE__BUILD_ID_CACHE, + DSO_BINARY_TYPE__SYSTEM_PATH_DSO, + DSO_BINARY_TYPE__NOT_FOUND, + }; + int i = 0; + + if (dso->data_type != DSO_BINARY_TYPE__NOT_FOUND) + return open_dso(dso, machine); + + do { + int fd; + + dso->data_type = binary_type_data[i++]; + + fd = open_dso(dso, machine); + if (fd >= 0) + return fd; + + } while (dso->data_type != DSO_BINARY_TYPE__NOT_FOUND); + + return -EINVAL; +} + +static void +dso_cache__free(struct rb_root *root) +{ + struct rb_node *next = rb_first(root); + + while (next) { + struct dso_cache *cache; + + cache = rb_entry(next, struct dso_cache, rb_node); + next = rb_next(&cache->rb_node); + rb_erase(&cache->rb_node, root); + free(cache); + } +} + +static struct dso_cache* +dso_cache__find(struct rb_root *root, u64 offset) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent = NULL; + struct dso_cache *cache; + + while (*p != NULL) { + u64 end; + + parent = *p; + cache = rb_entry(parent, struct dso_cache, rb_node); + end = cache->offset + DSO__DATA_CACHE_SIZE; + + if (offset < cache->offset) + p = &(*p)->rb_left; + else if (offset >= end) + p = &(*p)->rb_right; + else + return cache; + } + return NULL; +} + +static void +dso_cache__insert(struct rb_root *root, struct dso_cache *new) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent = NULL; + struct dso_cache *cache; + u64 offset = new->offset; + + while (*p != NULL) { + u64 end; + + parent = *p; + cache = rb_entry(parent, struct dso_cache, rb_node); + end = cache->offset + DSO__DATA_CACHE_SIZE; + + if (offset < cache->offset) + p = &(*p)->rb_left; + else if (offset >= end) + p = &(*p)->rb_right; + } + + rb_link_node(&new->rb_node, parent, p); + rb_insert_color(&new->rb_node, root); +} + +static ssize_t +dso_cache__memcpy(struct dso_cache *cache, u64 offset, + u8 *data, u64 size) +{ + u64 cache_offset = offset - cache->offset; + u64 cache_size = min(cache->size - cache_offset, size); + + memcpy(data, cache->data + cache_offset, cache_size); + return cache_size; +} + +static ssize_t +dso_cache__read(struct dso *dso, struct machine *machine, + u64 offset, u8 *data, ssize_t size) +{ + struct dso_cache *cache; + ssize_t ret; + int fd; + + fd = dso__data_fd(dso, machine); + if (fd < 0) + return -1; + + do { + u64 cache_offset; + + ret = -ENOMEM; + + cache = zalloc(sizeof(*cache) + DSO__DATA_CACHE_SIZE); + if (!cache) + break; + + cache_offset = offset & DSO__DATA_CACHE_MASK; + ret = -EINVAL; + + if (-1 == lseek(fd, cache_offset, SEEK_SET)) + break; + + ret = read(fd, cache->data, DSO__DATA_CACHE_SIZE); + if (ret <= 0) + break; + + cache->offset = cache_offset; + cache->size = ret; + dso_cache__insert(&dso->cache, cache); + + ret = dso_cache__memcpy(cache, offset, data, size); + + } while (0); + + if (ret <= 0) + free(cache); + + close(fd); + return ret; +} + +static ssize_t dso_cache_read(struct dso *dso, struct machine *machine, + u64 offset, u8 *data, ssize_t size) +{ + struct dso_cache *cache; + + cache = dso_cache__find(&dso->cache, offset); + if (cache) + return dso_cache__memcpy(cache, offset, data, size); + else + return dso_cache__read(dso, machine, offset, data, size); +} + +ssize_t dso__data_read_offset(struct dso *dso, struct machine *machine, + u64 offset, u8 *data, ssize_t size) +{ + ssize_t r = 0; + u8 *p = data; + + do { + ssize_t ret; + + ret = dso_cache_read(dso, machine, offset, p, size); + if (ret < 0) + return ret; + + /* Reached EOF, return what we have. */ + if (!ret) + break; + + BUG_ON(ret > size); + + r += ret; + p += ret; + offset += ret; + size -= ret; + + } while (size); + + return r; +} + +ssize_t dso__data_read_addr(struct dso *dso, struct map *map, + struct machine *machine, u64 addr, + u8 *data, ssize_t size) +{ + u64 offset = map->map_ip(map, addr); + return dso__data_read_offset(dso, machine, offset, data, size); +} + +struct map *dso__new_map(const char *name) +{ + struct map *map = NULL; + struct dso *dso = dso__new(name); + + if (dso) + map = map__new2(0, dso, MAP__FUNCTION); + + return map; +} + +struct dso *dso__kernel_findnew(struct machine *machine, const char *name, + const char *short_name, int dso_type) +{ + /* + * The kernel dso could be created by build_id processing. + */ + struct dso *dso = __dsos__findnew(&machine->kernel_dsos, name); + + /* + * We need to run this in all cases, since during the build_id + * processing we had no idea this was the kernel dso. + */ + if (dso != NULL) { + dso__set_short_name(dso, short_name); + dso->kernel = dso_type; + } + + return dso; +} + +void dso__set_long_name(struct dso *dso, char *name) +{ + if (name == NULL) + return; + dso->long_name = name; + dso->long_name_len = strlen(name); +} + +void dso__set_short_name(struct dso *dso, const char *name) +{ + if (name == NULL) + return; + dso->short_name = name; + dso->short_name_len = strlen(name); +} + +static void dso__set_basename(struct dso *dso) +{ + dso__set_short_name(dso, basename(dso->long_name)); +} + +int dso__name_len(const struct dso *dso) +{ + if (!dso) + return strlen("[unknown]"); + if (verbose) + return dso->long_name_len; + + return dso->short_name_len; +} + +bool dso__loaded(const struct dso *dso, enum map_type type) +{ + return dso->loaded & (1 << type); +} + +bool dso__sorted_by_name(const struct dso *dso, enum map_type type) +{ + return dso->sorted_by_name & (1 << type); +} + +void dso__set_sorted_by_name(struct dso *dso, enum map_type type) +{ + dso->sorted_by_name |= (1 << type); +} + +struct dso *dso__new(const char *name) +{ + struct dso *dso = calloc(1, sizeof(*dso) + strlen(name) + 1); + + if (dso != NULL) { + int i; + strcpy(dso->name, name); + dso__set_long_name(dso, dso->name); + dso__set_short_name(dso, dso->name); + for (i = 0; i < MAP__NR_TYPES; ++i) + dso->symbols[i] = dso->symbol_names[i] = RB_ROOT; + dso->cache = RB_ROOT; + dso->symtab_type = DSO_BINARY_TYPE__NOT_FOUND; + dso->data_type = DSO_BINARY_TYPE__NOT_FOUND; + dso->loaded = 0; + dso->sorted_by_name = 0; + dso->has_build_id = 0; + dso->kernel = DSO_TYPE_USER; + dso->needs_swap = DSO_SWAP__UNSET; + INIT_LIST_HEAD(&dso->node); + } + + return dso; +} + +void dso__delete(struct dso *dso) +{ + int i; + for (i = 0; i < MAP__NR_TYPES; ++i) + symbols__delete(&dso->symbols[i]); + if (dso->sname_alloc) + free((char *)dso->short_name); + if (dso->lname_alloc) + free(dso->long_name); + dso_cache__free(&dso->cache); + free(dso); +} + +void dso__set_build_id(struct dso *dso, void *build_id) +{ + memcpy(dso->build_id, build_id, sizeof(dso->build_id)); + dso->has_build_id = 1; +} + +bool dso__build_id_equal(const struct dso *dso, u8 *build_id) +{ + return memcmp(dso->build_id, build_id, sizeof(dso->build_id)) == 0; +} + +void dso__read_running_kernel_build_id(struct dso *dso, struct machine *machine) +{ + char path[PATH_MAX]; + + if (machine__is_default_guest(machine)) + return; + sprintf(path, "%s/sys/kernel/notes", machine->root_dir); + if (sysfs__read_build_id(path, dso->build_id, + sizeof(dso->build_id)) == 0) + dso->has_build_id = true; +} + +int dso__kernel_module_get_build_id(struct dso *dso, + const char *root_dir) +{ + char filename[PATH_MAX]; + /* + * kernel module short names are of the form "[module]" and + * we need just "module" here. + */ + const char *name = dso->short_name + 1; + + snprintf(filename, sizeof(filename), + "%s/sys/module/%.*s/notes/.note.gnu.build-id", + root_dir, (int)strlen(name) - 1, name); + + if (sysfs__read_build_id(filename, dso->build_id, + sizeof(dso->build_id)) == 0) + dso->has_build_id = true; + + return 0; +} + +bool __dsos__read_build_ids(struct list_head *head, bool with_hits) +{ + bool have_build_id = false; + struct dso *pos; + + list_for_each_entry(pos, head, node) { + if (with_hits && !pos->hit) + continue; + if (pos->has_build_id) { + have_build_id = true; + continue; + } + if (filename__read_build_id(pos->long_name, pos->build_id, + sizeof(pos->build_id)) > 0) { + have_build_id = true; + pos->has_build_id = true; + } + } + + return have_build_id; +} + +void dsos__add(struct list_head *head, struct dso *dso) +{ + list_add_tail(&dso->node, head); +} + +struct dso *dsos__find(struct list_head *head, const char *name) +{ + struct dso *pos; + + list_for_each_entry(pos, head, node) + if (strcmp(pos->long_name, name) == 0) + return pos; + return NULL; +} + +struct dso *__dsos__findnew(struct list_head *head, const char *name) +{ + struct dso *dso = dsos__find(head, name); + + if (!dso) { + dso = dso__new(name); + if (dso != NULL) { + dsos__add(head, dso); + dso__set_basename(dso); + } + } + + return dso; +} + +size_t __dsos__fprintf_buildid(struct list_head *head, FILE *fp, + bool (skip)(struct dso *dso, int parm), int parm) +{ + struct dso *pos; + size_t ret = 0; + + list_for_each_entry(pos, head, node) { + if (skip && skip(pos, parm)) + continue; + ret += dso__fprintf_buildid(pos, fp); + ret += fprintf(fp, " %s\n", pos->long_name); + } + return ret; +} + +size_t __dsos__fprintf(struct list_head *head, FILE *fp) +{ + struct dso *pos; + size_t ret = 0; + + list_for_each_entry(pos, head, node) { + int i; + for (i = 0; i < MAP__NR_TYPES; ++i) + ret += dso__fprintf(pos, i, fp); + } + + return ret; +} + +size_t dso__fprintf_buildid(struct dso *dso, FILE *fp) +{ + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + build_id__sprintf(dso->build_id, sizeof(dso->build_id), sbuild_id); + return fprintf(fp, "%s", sbuild_id); +} + +size_t dso__fprintf(struct dso *dso, enum map_type type, FILE *fp) +{ + struct rb_node *nd; + size_t ret = fprintf(fp, "dso: %s (", dso->short_name); + + if (dso->short_name != dso->long_name) + ret += fprintf(fp, "%s, ", dso->long_name); + ret += fprintf(fp, "%s, %sloaded, ", map_type__name[type], + dso__loaded(dso, type) ? "" : "NOT "); + ret += dso__fprintf_buildid(dso, fp); + ret += fprintf(fp, ")\n"); + for (nd = rb_first(&dso->symbols[type]); nd; nd = rb_next(nd)) { + struct symbol *pos = rb_entry(nd, struct symbol, rb_node); + ret += symbol__fprintf(pos, fp); + } + + return ret; +} diff --git a/tools/perf/util/dso.h b/tools/perf/util/dso.h new file mode 100644 index 00000000..450199ab --- /dev/null +++ b/tools/perf/util/dso.h @@ -0,0 +1,148 @@ +#ifndef __PERF_DSO +#define __PERF_DSO + +#include <linux/types.h> +#include <linux/rbtree.h> +#include "types.h" +#include "map.h" + +enum dso_binary_type { + DSO_BINARY_TYPE__KALLSYMS = 0, + DSO_BINARY_TYPE__GUEST_KALLSYMS, + DSO_BINARY_TYPE__VMLINUX, + DSO_BINARY_TYPE__GUEST_VMLINUX, + DSO_BINARY_TYPE__JAVA_JIT, + DSO_BINARY_TYPE__DEBUGLINK, + DSO_BINARY_TYPE__BUILD_ID_CACHE, + DSO_BINARY_TYPE__FEDORA_DEBUGINFO, + DSO_BINARY_TYPE__UBUNTU_DEBUGINFO, + DSO_BINARY_TYPE__BUILDID_DEBUGINFO, + DSO_BINARY_TYPE__SYSTEM_PATH_DSO, + DSO_BINARY_TYPE__GUEST_KMODULE, + DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE, + DSO_BINARY_TYPE__NOT_FOUND, +}; + +enum dso_kernel_type { + DSO_TYPE_USER = 0, + DSO_TYPE_KERNEL, + DSO_TYPE_GUEST_KERNEL +}; + +enum dso_swap_type { + DSO_SWAP__UNSET, + DSO_SWAP__NO, + DSO_SWAP__YES, +}; + +#define DSO__SWAP(dso, type, val) \ +({ \ + type ____r = val; \ + BUG_ON(dso->needs_swap == DSO_SWAP__UNSET); \ + if (dso->needs_swap == DSO_SWAP__YES) { \ + switch (sizeof(____r)) { \ + case 2: \ + ____r = bswap_16(val); \ + break; \ + case 4: \ + ____r = bswap_32(val); \ + break; \ + case 8: \ + ____r = bswap_64(val); \ + break; \ + default: \ + BUG_ON(1); \ + } \ + } \ + ____r; \ +}) + +#define DSO__DATA_CACHE_SIZE 4096 +#define DSO__DATA_CACHE_MASK ~(DSO__DATA_CACHE_SIZE - 1) + +struct dso_cache { + struct rb_node rb_node; + u64 offset; + u64 size; + char data[0]; +}; + +struct dso { + struct list_head node; + struct rb_root symbols[MAP__NR_TYPES]; + struct rb_root symbol_names[MAP__NR_TYPES]; + struct rb_root cache; + enum dso_kernel_type kernel; + enum dso_swap_type needs_swap; + enum dso_binary_type symtab_type; + enum dso_binary_type data_type; + u8 adjust_symbols:1; + u8 has_build_id:1; + u8 hit:1; + u8 annotate_warned:1; + u8 sname_alloc:1; + u8 lname_alloc:1; + u8 sorted_by_name; + u8 loaded; + u8 build_id[BUILD_ID_SIZE]; + const char *short_name; + char *long_name; + u16 long_name_len; + u16 short_name_len; + char name[0]; +}; + +static inline void dso__set_loaded(struct dso *dso, enum map_type type) +{ + dso->loaded |= (1 << type); +} + +struct dso *dso__new(const char *name); +void dso__delete(struct dso *dso); + +void dso__set_short_name(struct dso *dso, const char *name); +void dso__set_long_name(struct dso *dso, char *name); + +int dso__name_len(const struct dso *dso); + +bool dso__loaded(const struct dso *dso, enum map_type type); + +bool dso__sorted_by_name(const struct dso *dso, enum map_type type); +void dso__set_sorted_by_name(struct dso *dso, enum map_type type); +void dso__sort_by_name(struct dso *dso, enum map_type type); + +void dso__set_build_id(struct dso *dso, void *build_id); +bool dso__build_id_equal(const struct dso *dso, u8 *build_id); +void dso__read_running_kernel_build_id(struct dso *dso, + struct machine *machine); +int dso__kernel_module_get_build_id(struct dso *dso, const char *root_dir); + +char dso__symtab_origin(const struct dso *dso); +int dso__binary_type_file(struct dso *dso, enum dso_binary_type type, + char *root_dir, char *file, size_t size); + +int dso__data_fd(struct dso *dso, struct machine *machine); +ssize_t dso__data_read_offset(struct dso *dso, struct machine *machine, + u64 offset, u8 *data, ssize_t size); +ssize_t dso__data_read_addr(struct dso *dso, struct map *map, + struct machine *machine, u64 addr, + u8 *data, ssize_t size); + +struct map *dso__new_map(const char *name); +struct dso *dso__kernel_findnew(struct machine *machine, const char *name, + const char *short_name, int dso_type); + +void dsos__add(struct list_head *head, struct dso *dso); +struct dso *dsos__find(struct list_head *head, const char *name); +struct dso *__dsos__findnew(struct list_head *head, const char *name); +bool __dsos__read_build_ids(struct list_head *head, bool with_hits); + +size_t __dsos__fprintf_buildid(struct list_head *head, FILE *fp, + bool (skip)(struct dso *dso, int parm), int parm); +size_t __dsos__fprintf(struct list_head *head, FILE *fp); + +size_t dso__fprintf_buildid(struct dso *dso, FILE *fp); +size_t dso__fprintf_symbols_by_name(struct dso *dso, + enum map_type type, FILE *fp); +size_t dso__fprintf(struct dso *dso, enum map_type type, FILE *fp); +#endif /* __PERF_DSO */ diff --git a/tools/perf/util/dwarf-aux.c b/tools/perf/util/dwarf-aux.c new file mode 100644 index 00000000..3e5f5430 --- /dev/null +++ b/tools/perf/util/dwarf-aux.c @@ -0,0 +1,845 @@ +/* + * dwarf-aux.c : libdw auxiliary interfaces + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ + +#include <stdbool.h> +#include "util.h" +#include "debug.h" +#include "dwarf-aux.h" + +/** + * cu_find_realpath - Find the realpath of the target file + * @cu_die: A DIE(dwarf information entry) of CU(compilation Unit) + * @fname: The tail filename of the target file + * + * Find the real(long) path of @fname in @cu_die. + */ +const char *cu_find_realpath(Dwarf_Die *cu_die, const char *fname) +{ + Dwarf_Files *files; + size_t nfiles, i; + const char *src = NULL; + int ret; + + if (!fname) + return NULL; + + ret = dwarf_getsrcfiles(cu_die, &files, &nfiles); + if (ret != 0) + return NULL; + + for (i = 0; i < nfiles; i++) { + src = dwarf_filesrc(files, i, NULL, NULL); + if (strtailcmp(src, fname) == 0) + break; + } + if (i == nfiles) + return NULL; + return src; +} + +/** + * cu_get_comp_dir - Get the path of compilation directory + * @cu_die: a CU DIE + * + * Get the path of compilation directory of given @cu_die. + * Since this depends on DW_AT_comp_dir, older gcc will not + * embedded it. In that case, this returns NULL. + */ +const char *cu_get_comp_dir(Dwarf_Die *cu_die) +{ + Dwarf_Attribute attr; + if (dwarf_attr(cu_die, DW_AT_comp_dir, &attr) == NULL) + return NULL; + return dwarf_formstring(&attr); +} + +/** + * cu_find_lineinfo - Get a line number and file name for given address + * @cu_die: a CU DIE + * @addr: An address + * @fname: a pointer which returns the file name string + * @lineno: a pointer which returns the line number + * + * Find a line number and file name for @addr in @cu_die. + */ +int cu_find_lineinfo(Dwarf_Die *cu_die, unsigned long addr, + const char **fname, int *lineno) +{ + Dwarf_Line *line; + Dwarf_Addr laddr; + + line = dwarf_getsrc_die(cu_die, (Dwarf_Addr)addr); + if (line && dwarf_lineaddr(line, &laddr) == 0 && + addr == (unsigned long)laddr && dwarf_lineno(line, lineno) == 0) { + *fname = dwarf_linesrc(line, NULL, NULL); + if (!*fname) + /* line number is useless without filename */ + *lineno = 0; + } + + return *lineno ?: -ENOENT; +} + +static int __die_find_inline_cb(Dwarf_Die *die_mem, void *data); + +/** + * cu_walk_functions_at - Walk on function DIEs at given address + * @cu_die: A CU DIE + * @addr: An address + * @callback: A callback which called with found DIEs + * @data: A user data + * + * Walk on function DIEs at given @addr in @cu_die. Passed DIEs + * should be subprogram or inlined-subroutines. + */ +int cu_walk_functions_at(Dwarf_Die *cu_die, Dwarf_Addr addr, + int (*callback)(Dwarf_Die *, void *), void *data) +{ + Dwarf_Die die_mem; + Dwarf_Die *sc_die; + int ret = -ENOENT; + + /* Inlined function could be recursive. Trace it until fail */ + for (sc_die = die_find_realfunc(cu_die, addr, &die_mem); + sc_die != NULL; + sc_die = die_find_child(sc_die, __die_find_inline_cb, &addr, + &die_mem)) { + ret = callback(sc_die, data); + if (ret) + break; + } + + return ret; + +} + +/** + * die_compare_name - Compare diename and tname + * @dw_die: a DIE + * @tname: a string of target name + * + * Compare the name of @dw_die and @tname. Return false if @dw_die has no name. + */ +bool die_compare_name(Dwarf_Die *dw_die, const char *tname) +{ + const char *name; + name = dwarf_diename(dw_die); + return name ? (strcmp(tname, name) == 0) : false; +} + +/** + * die_get_call_lineno - Get callsite line number of inline-function instance + * @in_die: a DIE of an inlined function instance + * + * Get call-site line number of @in_die. This means from where the inline + * function is called. + */ +int die_get_call_lineno(Dwarf_Die *in_die) +{ + Dwarf_Attribute attr; + Dwarf_Word ret; + + if (!dwarf_attr(in_die, DW_AT_call_line, &attr)) + return -ENOENT; + + dwarf_formudata(&attr, &ret); + return (int)ret; +} + +/** + * die_get_type - Get type DIE + * @vr_die: a DIE of a variable + * @die_mem: where to store a type DIE + * + * Get a DIE of the type of given variable (@vr_die), and store + * it to die_mem. Return NULL if fails to get a type DIE. + */ +Dwarf_Die *die_get_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem) +{ + Dwarf_Attribute attr; + + if (dwarf_attr_integrate(vr_die, DW_AT_type, &attr) && + dwarf_formref_die(&attr, die_mem)) + return die_mem; + else + return NULL; +} + +/* Get a type die, but skip qualifiers */ +static Dwarf_Die *__die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem) +{ + int tag; + + do { + vr_die = die_get_type(vr_die, die_mem); + if (!vr_die) + break; + tag = dwarf_tag(vr_die); + } while (tag == DW_TAG_const_type || + tag == DW_TAG_restrict_type || + tag == DW_TAG_volatile_type || + tag == DW_TAG_shared_type); + + return vr_die; +} + +/** + * die_get_real_type - Get a type die, but skip qualifiers and typedef + * @vr_die: a DIE of a variable + * @die_mem: where to store a type DIE + * + * Get a DIE of the type of given variable (@vr_die), and store + * it to die_mem. Return NULL if fails to get a type DIE. + * If the type is qualifiers (e.g. const) or typedef, this skips it + * and tries to find real type (structure or basic types, e.g. int). + */ +Dwarf_Die *die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem) +{ + do { + vr_die = __die_get_real_type(vr_die, die_mem); + } while (vr_die && dwarf_tag(vr_die) == DW_TAG_typedef); + + return vr_die; +} + +/* Get attribute and translate it as a udata */ +static int die_get_attr_udata(Dwarf_Die *tp_die, unsigned int attr_name, + Dwarf_Word *result) +{ + Dwarf_Attribute attr; + + if (dwarf_attr(tp_die, attr_name, &attr) == NULL || + dwarf_formudata(&attr, result) != 0) + return -ENOENT; + + return 0; +} + +/* Get attribute and translate it as a sdata */ +static int die_get_attr_sdata(Dwarf_Die *tp_die, unsigned int attr_name, + Dwarf_Sword *result) +{ + Dwarf_Attribute attr; + + if (dwarf_attr(tp_die, attr_name, &attr) == NULL || + dwarf_formsdata(&attr, result) != 0) + return -ENOENT; + + return 0; +} + +/** + * die_is_signed_type - Check whether a type DIE is signed or not + * @tp_die: a DIE of a type + * + * Get the encoding of @tp_die and return true if the encoding + * is signed. + */ +bool die_is_signed_type(Dwarf_Die *tp_die) +{ + Dwarf_Word ret; + + if (die_get_attr_udata(tp_die, DW_AT_encoding, &ret)) + return false; + + return (ret == DW_ATE_signed_char || ret == DW_ATE_signed || + ret == DW_ATE_signed_fixed); +} + +/** + * die_get_data_member_location - Get the data-member offset + * @mb_die: a DIE of a member of a data structure + * @offs: The offset of the member in the data structure + * + * Get the offset of @mb_die in the data structure including @mb_die, and + * stores result offset to @offs. If any error occurs this returns errno. + */ +int die_get_data_member_location(Dwarf_Die *mb_die, Dwarf_Word *offs) +{ + Dwarf_Attribute attr; + Dwarf_Op *expr; + size_t nexpr; + int ret; + + if (dwarf_attr(mb_die, DW_AT_data_member_location, &attr) == NULL) + return -ENOENT; + + if (dwarf_formudata(&attr, offs) != 0) { + /* DW_AT_data_member_location should be DW_OP_plus_uconst */ + ret = dwarf_getlocation(&attr, &expr, &nexpr); + if (ret < 0 || nexpr == 0) + return -ENOENT; + + if (expr[0].atom != DW_OP_plus_uconst || nexpr != 1) { + pr_debug("Unable to get offset:Unexpected OP %x (%zd)\n", + expr[0].atom, nexpr); + return -ENOTSUP; + } + *offs = (Dwarf_Word)expr[0].number; + } + return 0; +} + +/* Get the call file index number in CU DIE */ +static int die_get_call_fileno(Dwarf_Die *in_die) +{ + Dwarf_Sword idx; + + if (die_get_attr_sdata(in_die, DW_AT_call_file, &idx) == 0) + return (int)idx; + else + return -ENOENT; +} + +/* Get the declared file index number in CU DIE */ +static int die_get_decl_fileno(Dwarf_Die *pdie) +{ + Dwarf_Sword idx; + + if (die_get_attr_sdata(pdie, DW_AT_decl_file, &idx) == 0) + return (int)idx; + else + return -ENOENT; +} + +/** + * die_get_call_file - Get callsite file name of inlined function instance + * @in_die: a DIE of an inlined function instance + * + * Get call-site file name of @in_die. This means from which file the inline + * function is called. + */ +const char *die_get_call_file(Dwarf_Die *in_die) +{ + Dwarf_Die cu_die; + Dwarf_Files *files; + int idx; + + idx = die_get_call_fileno(in_die); + if (idx < 0 || !dwarf_diecu(in_die, &cu_die, NULL, NULL) || + dwarf_getsrcfiles(&cu_die, &files, NULL) != 0) + return NULL; + + return dwarf_filesrc(files, idx, NULL, NULL); +} + + +/** + * die_find_child - Generic DIE search function in DIE tree + * @rt_die: a root DIE + * @callback: a callback function + * @data: a user data passed to the callback function + * @die_mem: a buffer for result DIE + * + * Trace DIE tree from @rt_die and call @callback for each child DIE. + * If @callback returns DIE_FIND_CB_END, this stores the DIE into + * @die_mem and returns it. If @callback returns DIE_FIND_CB_CONTINUE, + * this continues to trace the tree. Optionally, @callback can return + * DIE_FIND_CB_CHILD and DIE_FIND_CB_SIBLING, those means trace only + * the children and trace only the siblings respectively. + * Returns NULL if @callback can't find any appropriate DIE. + */ +Dwarf_Die *die_find_child(Dwarf_Die *rt_die, + int (*callback)(Dwarf_Die *, void *), + void *data, Dwarf_Die *die_mem) +{ + Dwarf_Die child_die; + int ret; + + ret = dwarf_child(rt_die, die_mem); + if (ret != 0) + return NULL; + + do { + ret = callback(die_mem, data); + if (ret == DIE_FIND_CB_END) + return die_mem; + + if ((ret & DIE_FIND_CB_CHILD) && + die_find_child(die_mem, callback, data, &child_die)) { + memcpy(die_mem, &child_die, sizeof(Dwarf_Die)); + return die_mem; + } + } while ((ret & DIE_FIND_CB_SIBLING) && + dwarf_siblingof(die_mem, die_mem) == 0); + + return NULL; +} + +struct __addr_die_search_param { + Dwarf_Addr addr; + Dwarf_Die *die_mem; +}; + +/* die_find callback for non-inlined function search */ +static int __die_search_func_cb(Dwarf_Die *fn_die, void *data) +{ + struct __addr_die_search_param *ad = data; + + if (dwarf_tag(fn_die) == DW_TAG_subprogram && + dwarf_haspc(fn_die, ad->addr)) { + memcpy(ad->die_mem, fn_die, sizeof(Dwarf_Die)); + return DWARF_CB_ABORT; + } + return DWARF_CB_OK; +} + +/** + * die_find_realfunc - Search a non-inlined function at given address + * @cu_die: a CU DIE which including @addr + * @addr: target address + * @die_mem: a buffer for result DIE + * + * Search a non-inlined function DIE which includes @addr. Stores the + * DIE to @die_mem and returns it if found. Returns NULl if failed. + */ +Dwarf_Die *die_find_realfunc(Dwarf_Die *cu_die, Dwarf_Addr addr, + Dwarf_Die *die_mem) +{ + struct __addr_die_search_param ad; + ad.addr = addr; + ad.die_mem = die_mem; + /* dwarf_getscopes can't find subprogram. */ + if (!dwarf_getfuncs(cu_die, __die_search_func_cb, &ad, 0)) + return NULL; + else + return die_mem; +} + +/* die_find callback for inline function search */ +static int __die_find_inline_cb(Dwarf_Die *die_mem, void *data) +{ + Dwarf_Addr *addr = data; + + if (dwarf_tag(die_mem) == DW_TAG_inlined_subroutine && + dwarf_haspc(die_mem, *addr)) + return DIE_FIND_CB_END; + + return DIE_FIND_CB_CONTINUE; +} + +/** + * die_find_inlinefunc - Search an inlined function at given address + * @cu_die: a CU DIE which including @addr + * @addr: target address + * @die_mem: a buffer for result DIE + * + * Search an inlined function DIE which includes @addr. Stores the + * DIE to @die_mem and returns it if found. Returns NULl if failed. + * If several inlined functions are expanded recursively, this trace + * it and returns deepest one. + */ +Dwarf_Die *die_find_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr, + Dwarf_Die *die_mem) +{ + Dwarf_Die tmp_die; + + sp_die = die_find_child(sp_die, __die_find_inline_cb, &addr, &tmp_die); + if (!sp_die) + return NULL; + + /* Inlined function could be recursive. Trace it until fail */ + while (sp_die) { + memcpy(die_mem, sp_die, sizeof(Dwarf_Die)); + sp_die = die_find_child(sp_die, __die_find_inline_cb, &addr, + &tmp_die); + } + + return die_mem; +} + +struct __instance_walk_param { + void *addr; + int (*callback)(Dwarf_Die *, void *); + void *data; + int retval; +}; + +static int __die_walk_instances_cb(Dwarf_Die *inst, void *data) +{ + struct __instance_walk_param *iwp = data; + Dwarf_Attribute attr_mem; + Dwarf_Die origin_mem; + Dwarf_Attribute *attr; + Dwarf_Die *origin; + int tmp; + + attr = dwarf_attr(inst, DW_AT_abstract_origin, &attr_mem); + if (attr == NULL) + return DIE_FIND_CB_CONTINUE; + + origin = dwarf_formref_die(attr, &origin_mem); + if (origin == NULL || origin->addr != iwp->addr) + return DIE_FIND_CB_CONTINUE; + + /* Ignore redundant instances */ + if (dwarf_tag(inst) == DW_TAG_inlined_subroutine) { + dwarf_decl_line(origin, &tmp); + if (die_get_call_lineno(inst) == tmp) { + tmp = die_get_decl_fileno(origin); + if (die_get_call_fileno(inst) == tmp) + return DIE_FIND_CB_CONTINUE; + } + } + + iwp->retval = iwp->callback(inst, iwp->data); + + return (iwp->retval) ? DIE_FIND_CB_END : DIE_FIND_CB_CONTINUE; +} + +/** + * die_walk_instances - Walk on instances of given DIE + * @or_die: an abstract original DIE + * @callback: a callback function which is called with instance DIE + * @data: user data + * + * Walk on the instances of give @in_die. @in_die must be an inlined function + * declartion. This returns the return value of @callback if it returns + * non-zero value, or -ENOENT if there is no instance. + */ +int die_walk_instances(Dwarf_Die *or_die, int (*callback)(Dwarf_Die *, void *), + void *data) +{ + Dwarf_Die cu_die; + Dwarf_Die die_mem; + struct __instance_walk_param iwp = { + .addr = or_die->addr, + .callback = callback, + .data = data, + .retval = -ENOENT, + }; + + if (dwarf_diecu(or_die, &cu_die, NULL, NULL) == NULL) + return -ENOENT; + + die_find_child(&cu_die, __die_walk_instances_cb, &iwp, &die_mem); + + return iwp.retval; +} + +/* Line walker internal parameters */ +struct __line_walk_param { + bool recursive; + line_walk_callback_t callback; + void *data; + int retval; +}; + +static int __die_walk_funclines_cb(Dwarf_Die *in_die, void *data) +{ + struct __line_walk_param *lw = data; + Dwarf_Addr addr = 0; + const char *fname; + int lineno; + + if (dwarf_tag(in_die) == DW_TAG_inlined_subroutine) { + fname = die_get_call_file(in_die); + lineno = die_get_call_lineno(in_die); + if (fname && lineno > 0 && dwarf_entrypc(in_die, &addr) == 0) { + lw->retval = lw->callback(fname, lineno, addr, lw->data); + if (lw->retval != 0) + return DIE_FIND_CB_END; + } + } + if (!lw->recursive) + /* Don't need to search recursively */ + return DIE_FIND_CB_SIBLING; + + if (addr) { + fname = dwarf_decl_file(in_die); + if (fname && dwarf_decl_line(in_die, &lineno) == 0) { + lw->retval = lw->callback(fname, lineno, addr, lw->data); + if (lw->retval != 0) + return DIE_FIND_CB_END; + } + } + + /* Continue to search nested inlined function call-sites */ + return DIE_FIND_CB_CONTINUE; +} + +/* Walk on lines of blocks included in given DIE */ +static int __die_walk_funclines(Dwarf_Die *sp_die, bool recursive, + line_walk_callback_t callback, void *data) +{ + struct __line_walk_param lw = { + .recursive = recursive, + .callback = callback, + .data = data, + .retval = 0, + }; + Dwarf_Die die_mem; + Dwarf_Addr addr; + const char *fname; + int lineno; + + /* Handle function declaration line */ + fname = dwarf_decl_file(sp_die); + if (fname && dwarf_decl_line(sp_die, &lineno) == 0 && + dwarf_entrypc(sp_die, &addr) == 0) { + lw.retval = callback(fname, lineno, addr, data); + if (lw.retval != 0) + goto done; + } + die_find_child(sp_die, __die_walk_funclines_cb, &lw, &die_mem); +done: + return lw.retval; +} + +static int __die_walk_culines_cb(Dwarf_Die *sp_die, void *data) +{ + struct __line_walk_param *lw = data; + + lw->retval = __die_walk_funclines(sp_die, true, lw->callback, lw->data); + if (lw->retval != 0) + return DWARF_CB_ABORT; + + return DWARF_CB_OK; +} + +/** + * die_walk_lines - Walk on lines inside given DIE + * @rt_die: a root DIE (CU, subprogram or inlined_subroutine) + * @callback: callback routine + * @data: user data + * + * Walk on all lines inside given @rt_die and call @callback on each line. + * If the @rt_die is a function, walk only on the lines inside the function, + * otherwise @rt_die must be a CU DIE. + * Note that this walks not only dwarf line list, but also function entries + * and inline call-site. + */ +int die_walk_lines(Dwarf_Die *rt_die, line_walk_callback_t callback, void *data) +{ + Dwarf_Lines *lines; + Dwarf_Line *line; + Dwarf_Addr addr; + const char *fname; + int lineno, ret = 0; + Dwarf_Die die_mem, *cu_die; + size_t nlines, i; + + /* Get the CU die */ + if (dwarf_tag(rt_die) != DW_TAG_compile_unit) + cu_die = dwarf_diecu(rt_die, &die_mem, NULL, NULL); + else + cu_die = rt_die; + if (!cu_die) { + pr_debug2("Failed to get CU from given DIE.\n"); + return -EINVAL; + } + + /* Get lines list in the CU */ + if (dwarf_getsrclines(cu_die, &lines, &nlines) != 0) { + pr_debug2("Failed to get source lines on this CU.\n"); + return -ENOENT; + } + pr_debug2("Get %zd lines from this CU\n", nlines); + + /* Walk on the lines on lines list */ + for (i = 0; i < nlines; i++) { + line = dwarf_onesrcline(lines, i); + if (line == NULL || + dwarf_lineno(line, &lineno) != 0 || + dwarf_lineaddr(line, &addr) != 0) { + pr_debug2("Failed to get line info. " + "Possible error in debuginfo.\n"); + continue; + } + /* Filter lines based on address */ + if (rt_die != cu_die) + /* + * Address filtering + * The line is included in given function, and + * no inline block includes it. + */ + if (!dwarf_haspc(rt_die, addr) || + die_find_inlinefunc(rt_die, addr, &die_mem)) + continue; + /* Get source line */ + fname = dwarf_linesrc(line, NULL, NULL); + + ret = callback(fname, lineno, addr, data); + if (ret != 0) + return ret; + } + + /* + * Dwarf lines doesn't include function declarations and inlined + * subroutines. We have to check functions list or given function. + */ + if (rt_die != cu_die) + /* + * Don't need walk functions recursively, because nested + * inlined functions don't have lines of the specified DIE. + */ + ret = __die_walk_funclines(rt_die, false, callback, data); + else { + struct __line_walk_param param = { + .callback = callback, + .data = data, + .retval = 0, + }; + dwarf_getfuncs(cu_die, __die_walk_culines_cb, ¶m, 0); + ret = param.retval; + } + + return ret; +} + +struct __find_variable_param { + const char *name; + Dwarf_Addr addr; +}; + +static int __die_find_variable_cb(Dwarf_Die *die_mem, void *data) +{ + struct __find_variable_param *fvp = data; + int tag; + + tag = dwarf_tag(die_mem); + if ((tag == DW_TAG_formal_parameter || + tag == DW_TAG_variable) && + die_compare_name(die_mem, fvp->name)) + return DIE_FIND_CB_END; + + if (dwarf_haspc(die_mem, fvp->addr)) + return DIE_FIND_CB_CONTINUE; + else + return DIE_FIND_CB_SIBLING; +} + +/** + * die_find_variable_at - Find a given name variable at given address + * @sp_die: a function DIE + * @name: variable name + * @addr: address + * @die_mem: a buffer for result DIE + * + * Find a variable DIE called @name at @addr in @sp_die. + */ +Dwarf_Die *die_find_variable_at(Dwarf_Die *sp_die, const char *name, + Dwarf_Addr addr, Dwarf_Die *die_mem) +{ + struct __find_variable_param fvp = { .name = name, .addr = addr}; + + return die_find_child(sp_die, __die_find_variable_cb, (void *)&fvp, + die_mem); +} + +static int __die_find_member_cb(Dwarf_Die *die_mem, void *data) +{ + const char *name = data; + + if ((dwarf_tag(die_mem) == DW_TAG_member) && + die_compare_name(die_mem, name)) + return DIE_FIND_CB_END; + + return DIE_FIND_CB_SIBLING; +} + +/** + * die_find_member - Find a given name member in a data structure + * @st_die: a data structure type DIE + * @name: member name + * @die_mem: a buffer for result DIE + * + * Find a member DIE called @name in @st_die. + */ +Dwarf_Die *die_find_member(Dwarf_Die *st_die, const char *name, + Dwarf_Die *die_mem) +{ + return die_find_child(st_die, __die_find_member_cb, (void *)name, + die_mem); +} + +/** + * die_get_typename - Get the name of given variable DIE + * @vr_die: a variable DIE + * @buf: a buffer for result type name + * @len: a max-length of @buf + * + * Get the name of @vr_die and stores it to @buf. Return the actual length + * of type name if succeeded. Return -E2BIG if @len is not enough long, and + * Return -ENOENT if failed to find type name. + * Note that the result will stores typedef name if possible, and stores + * "*(function_type)" if the type is a function pointer. + */ +int die_get_typename(Dwarf_Die *vr_die, char *buf, int len) +{ + Dwarf_Die type; + int tag, ret, ret2; + const char *tmp = ""; + + if (__die_get_real_type(vr_die, &type) == NULL) + return -ENOENT; + + tag = dwarf_tag(&type); + if (tag == DW_TAG_array_type || tag == DW_TAG_pointer_type) + tmp = "*"; + else if (tag == DW_TAG_subroutine_type) { + /* Function pointer */ + ret = snprintf(buf, len, "(function_type)"); + return (ret >= len) ? -E2BIG : ret; + } else { + if (!dwarf_diename(&type)) + return -ENOENT; + if (tag == DW_TAG_union_type) + tmp = "union "; + else if (tag == DW_TAG_structure_type) + tmp = "struct "; + else if (tag == DW_TAG_enumeration_type) + tmp = "enum "; + /* Write a base name */ + ret = snprintf(buf, len, "%s%s", tmp, dwarf_diename(&type)); + return (ret >= len) ? -E2BIG : ret; + } + ret = die_get_typename(&type, buf, len); + if (ret > 0) { + ret2 = snprintf(buf + ret, len - ret, "%s", tmp); + ret = (ret2 >= len - ret) ? -E2BIG : ret2 + ret; + } + return ret; +} + +/** + * die_get_varname - Get the name and type of given variable DIE + * @vr_die: a variable DIE + * @buf: a buffer for type and variable name + * @len: the max-length of @buf + * + * Get the name and type of @vr_die and stores it in @buf as "type\tname". + */ +int die_get_varname(Dwarf_Die *vr_die, char *buf, int len) +{ + int ret, ret2; + + ret = die_get_typename(vr_die, buf, len); + if (ret < 0) { + pr_debug("Failed to get type, make it unknown.\n"); + ret = snprintf(buf, len, "(unknown_type)"); + } + if (ret > 0) { + ret2 = snprintf(buf + ret, len - ret, "\t%s", + dwarf_diename(vr_die)); + ret = (ret2 >= len - ret) ? -E2BIG : ret2 + ret; + } + return ret; +} + diff --git a/tools/perf/util/dwarf-aux.h b/tools/perf/util/dwarf-aux.h new file mode 100644 index 00000000..6ce17177 --- /dev/null +++ b/tools/perf/util/dwarf-aux.h @@ -0,0 +1,111 @@ +#ifndef _DWARF_AUX_H +#define _DWARF_AUX_H +/* + * dwarf-aux.h : libdw auxiliary interfaces + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ + +#include <dwarf.h> +#include <elfutils/libdw.h> +#include <elfutils/libdwfl.h> +#include <elfutils/version.h> + +/* Find the realpath of the target file */ +extern const char *cu_find_realpath(Dwarf_Die *cu_die, const char *fname); + +/* Get DW_AT_comp_dir (should be NULL with older gcc) */ +extern const char *cu_get_comp_dir(Dwarf_Die *cu_die); + +/* Get a line number and file name for given address */ +extern int cu_find_lineinfo(Dwarf_Die *cudie, unsigned long addr, + const char **fname, int *lineno); + +/* Walk on funcitons at given address */ +extern int cu_walk_functions_at(Dwarf_Die *cu_die, Dwarf_Addr addr, + int (*callback)(Dwarf_Die *, void *), void *data); + +/* Compare diename and tname */ +extern bool die_compare_name(Dwarf_Die *dw_die, const char *tname); + +/* Get callsite line number of inline-function instance */ +extern int die_get_call_lineno(Dwarf_Die *in_die); + +/* Get callsite file name of inlined function instance */ +extern const char *die_get_call_file(Dwarf_Die *in_die); + +/* Get type die */ +extern Dwarf_Die *die_get_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem); + +/* Get a type die, but skip qualifiers and typedef */ +extern Dwarf_Die *die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem); + +/* Check whether the DIE is signed or not */ +extern bool die_is_signed_type(Dwarf_Die *tp_die); + +/* Get data_member_location offset */ +extern int die_get_data_member_location(Dwarf_Die *mb_die, Dwarf_Word *offs); + +/* Return values for die_find_child() callbacks */ +enum { + DIE_FIND_CB_END = 0, /* End of Search */ + DIE_FIND_CB_CHILD = 1, /* Search only children */ + DIE_FIND_CB_SIBLING = 2, /* Search only siblings */ + DIE_FIND_CB_CONTINUE = 3, /* Search children and siblings */ +}; + +/* Search child DIEs */ +extern Dwarf_Die *die_find_child(Dwarf_Die *rt_die, + int (*callback)(Dwarf_Die *, void *), + void *data, Dwarf_Die *die_mem); + +/* Search a non-inlined function including given address */ +extern Dwarf_Die *die_find_realfunc(Dwarf_Die *cu_die, Dwarf_Addr addr, + Dwarf_Die *die_mem); + +/* Search an inlined function including given address */ +extern Dwarf_Die *die_find_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr, + Dwarf_Die *die_mem); + +/* Walk on the instances of given DIE */ +extern int die_walk_instances(Dwarf_Die *in_die, + int (*callback)(Dwarf_Die *, void *), void *data); + +/* Walker on lines (Note: line number will not be sorted) */ +typedef int (* line_walk_callback_t) (const char *fname, int lineno, + Dwarf_Addr addr, void *data); + +/* + * Walk on lines inside given DIE. If the DIE is a subprogram, walk only on + * the lines inside the subprogram, otherwise the DIE must be a CU DIE. + */ +extern int die_walk_lines(Dwarf_Die *rt_die, line_walk_callback_t callback, + void *data); + +/* Find a variable called 'name' at given address */ +extern Dwarf_Die *die_find_variable_at(Dwarf_Die *sp_die, const char *name, + Dwarf_Addr addr, Dwarf_Die *die_mem); + +/* Find a member called 'name' */ +extern Dwarf_Die *die_find_member(Dwarf_Die *st_die, const char *name, + Dwarf_Die *die_mem); + +/* Get the name of given variable DIE */ +extern int die_get_typename(Dwarf_Die *vr_die, char *buf, int len); + +/* Get the name and type of given variable DIE, stored as "type\tname" */ +extern int die_get_varname(Dwarf_Die *vr_die, char *buf, int len); +#endif diff --git a/tools/perf/util/environment.c b/tools/perf/util/environment.c new file mode 100644 index 00000000..275b0ee3 --- /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/event.c b/tools/perf/util/event.c new file mode 100644 index 00000000..5cd13d76 --- /dev/null +++ b/tools/perf/util/event.c @@ -0,0 +1,733 @@ +#include <linux/types.h> +#include "event.h" +#include "debug.h" +#include "machine.h" +#include "sort.h" +#include "string.h" +#include "strlist.h" +#include "thread.h" +#include "thread_map.h" + +static const char *perf_event__names[] = { + [0] = "TOTAL", + [PERF_RECORD_MMAP] = "MMAP", + [PERF_RECORD_LOST] = "LOST", + [PERF_RECORD_COMM] = "COMM", + [PERF_RECORD_EXIT] = "EXIT", + [PERF_RECORD_THROTTLE] = "THROTTLE", + [PERF_RECORD_UNTHROTTLE] = "UNTHROTTLE", + [PERF_RECORD_FORK] = "FORK", + [PERF_RECORD_READ] = "READ", + [PERF_RECORD_SAMPLE] = "SAMPLE", + [PERF_RECORD_HEADER_ATTR] = "ATTR", + [PERF_RECORD_HEADER_EVENT_TYPE] = "EVENT_TYPE", + [PERF_RECORD_HEADER_TRACING_DATA] = "TRACING_DATA", + [PERF_RECORD_HEADER_BUILD_ID] = "BUILD_ID", + [PERF_RECORD_FINISHED_ROUND] = "FINISHED_ROUND", +}; + +const char *perf_event__name(unsigned int id) +{ + if (id >= ARRAY_SIZE(perf_event__names)) + return "INVALID"; + if (!perf_event__names[id]) + return "UNKNOWN"; + return perf_event__names[id]; +} + +static struct perf_sample synth_sample = { + .pid = -1, + .tid = -1, + .time = -1, + .stream_id = -1, + .cpu = -1, + .period = 1, +}; + +static pid_t perf_event__get_comm_tgid(pid_t pid, char *comm, size_t len) +{ + char filename[PATH_MAX]; + char bf[BUFSIZ]; + FILE *fp; + size_t size = 0; + pid_t tgid = -1; + + snprintf(filename, sizeof(filename), "/proc/%d/status", pid); + + fp = fopen(filename, "r"); + if (fp == NULL) { + pr_debug("couldn't open %s\n", filename); + return 0; + } + + while (!comm[0] || (tgid < 0)) { + if (fgets(bf, sizeof(bf), fp) == NULL) { + pr_warning("couldn't get COMM and pgid, malformed %s\n", + filename); + break; + } + + if (memcmp(bf, "Name:", 5) == 0) { + char *name = bf + 5; + while (*name && isspace(*name)) + ++name; + size = strlen(name) - 1; + if (size >= len) + size = len - 1; + memcpy(comm, name, size); + comm[size] = '\0'; + + } else if (memcmp(bf, "Tgid:", 5) == 0) { + char *tgids = bf + 5; + while (*tgids && isspace(*tgids)) + ++tgids; + tgid = atoi(tgids); + } + } + + fclose(fp); + + return tgid; +} + +static pid_t perf_event__synthesize_comm(struct perf_tool *tool, + union perf_event *event, pid_t pid, + int full, + perf_event__handler_t process, + struct machine *machine) +{ + char filename[PATH_MAX]; + size_t size; + DIR *tasks; + struct dirent dirent, *next; + pid_t tgid; + + memset(&event->comm, 0, sizeof(event->comm)); + + tgid = perf_event__get_comm_tgid(pid, event->comm.comm, + sizeof(event->comm.comm)); + if (tgid < 0) + goto out; + + event->comm.pid = tgid; + event->comm.header.type = PERF_RECORD_COMM; + + size = strlen(event->comm.comm) + 1; + size = PERF_ALIGN(size, sizeof(u64)); + memset(event->comm.comm + size, 0, machine->id_hdr_size); + event->comm.header.size = (sizeof(event->comm) - + (sizeof(event->comm.comm) - size) + + machine->id_hdr_size); + if (!full) { + event->comm.tid = pid; + + if (process(tool, event, &synth_sample, machine) != 0) + return -1; + + goto out; + } + + snprintf(filename, sizeof(filename), "/proc/%d/task", pid); + + tasks = opendir(filename); + if (tasks == NULL) { + pr_debug("couldn't open %s\n", filename); + return 0; + } + + while (!readdir_r(tasks, &dirent, &next) && next) { + char *end; + pid = strtol(dirent.d_name, &end, 10); + if (*end) + continue; + + /* already have tgid; jut want to update the comm */ + (void) perf_event__get_comm_tgid(pid, event->comm.comm, + sizeof(event->comm.comm)); + + size = strlen(event->comm.comm) + 1; + size = PERF_ALIGN(size, sizeof(u64)); + memset(event->comm.comm + size, 0, machine->id_hdr_size); + event->comm.header.size = (sizeof(event->comm) - + (sizeof(event->comm.comm) - size) + + machine->id_hdr_size); + + event->comm.tid = pid; + + if (process(tool, event, &synth_sample, machine) != 0) { + tgid = -1; + break; + } + } + + closedir(tasks); +out: + return tgid; +} + +static int perf_event__synthesize_mmap_events(struct perf_tool *tool, + union perf_event *event, + pid_t pid, pid_t tgid, + perf_event__handler_t process, + struct machine *machine) +{ + char filename[PATH_MAX]; + FILE *fp; + int rc = 0; + + snprintf(filename, sizeof(filename), "/proc/%d/maps", pid); + + fp = fopen(filename, "r"); + if (fp == NULL) { + /* + * We raced with a task exiting - just return: + */ + pr_debug("couldn't open %s\n", filename); + return -1; + } + + event->header.type = PERF_RECORD_MMAP; + /* + * Just like the kernel, see __perf_event_mmap in kernel/perf_event.c + */ + event->header.misc = PERF_RECORD_MISC_USER; + + while (1) { + char bf[BUFSIZ]; + char prot[5]; + char execname[PATH_MAX]; + char anonstr[] = "//anon"; + size_t size; + + if (fgets(bf, sizeof(bf), fp) == NULL) + break; + + /* ensure null termination since stack will be reused. */ + strcpy(execname, ""); + + /* 00400000-0040c000 r-xp 00000000 fd:01 41038 /bin/cat */ + sscanf(bf, "%"PRIx64"-%"PRIx64" %s %"PRIx64" %*x:%*x %*u %s\n", + &event->mmap.start, &event->mmap.len, prot, + &event->mmap.pgoff, execname); + + if (prot[2] != 'x') + continue; + + if (!strcmp(execname, "")) + strcpy(execname, anonstr); + + size = strlen(execname) + 1; + memcpy(event->mmap.filename, execname, size); + size = PERF_ALIGN(size, sizeof(u64)); + event->mmap.len -= event->mmap.start; + event->mmap.header.size = (sizeof(event->mmap) - + (sizeof(event->mmap.filename) - size)); + memset(event->mmap.filename + size, 0, machine->id_hdr_size); + event->mmap.header.size += machine->id_hdr_size; + event->mmap.pid = tgid; + event->mmap.tid = pid; + + if (process(tool, event, &synth_sample, machine) != 0) { + rc = -1; + break; + } + } + + fclose(fp); + return rc; +} + +int perf_event__synthesize_modules(struct perf_tool *tool, + perf_event__handler_t process, + struct machine *machine) +{ + int rc = 0; + struct rb_node *nd; + struct map_groups *kmaps = &machine->kmaps; + union perf_event *event = zalloc((sizeof(event->mmap) + + machine->id_hdr_size)); + if (event == NULL) { + pr_debug("Not enough memory synthesizing mmap event " + "for kernel modules\n"); + return -1; + } + + event->header.type = PERF_RECORD_MMAP; + + /* + * kernel uses 0 for user space maps, see kernel/perf_event.c + * __perf_event_mmap + */ + if (machine__is_host(machine)) + event->header.misc = PERF_RECORD_MISC_KERNEL; + else + event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL; + + for (nd = rb_first(&kmaps->maps[MAP__FUNCTION]); + nd; nd = rb_next(nd)) { + size_t size; + struct map *pos = rb_entry(nd, struct map, rb_node); + + if (pos->dso->kernel) + continue; + + size = PERF_ALIGN(pos->dso->long_name_len + 1, sizeof(u64)); + event->mmap.header.type = PERF_RECORD_MMAP; + event->mmap.header.size = (sizeof(event->mmap) - + (sizeof(event->mmap.filename) - size)); + memset(event->mmap.filename + size, 0, machine->id_hdr_size); + event->mmap.header.size += machine->id_hdr_size; + event->mmap.start = pos->start; + event->mmap.len = pos->end - pos->start; + event->mmap.pid = machine->pid; + + memcpy(event->mmap.filename, pos->dso->long_name, + pos->dso->long_name_len + 1); + if (process(tool, event, &synth_sample, machine) != 0) { + rc = -1; + break; + } + } + + free(event); + return rc; +} + +static int __event__synthesize_thread(union perf_event *comm_event, + union perf_event *mmap_event, + pid_t pid, int full, + perf_event__handler_t process, + struct perf_tool *tool, + struct machine *machine) +{ + pid_t tgid = perf_event__synthesize_comm(tool, comm_event, pid, full, + process, machine); + if (tgid == -1) + return -1; + return perf_event__synthesize_mmap_events(tool, mmap_event, pid, tgid, + process, machine); +} + +int perf_event__synthesize_thread_map(struct perf_tool *tool, + struct thread_map *threads, + perf_event__handler_t process, + struct machine *machine) +{ + union perf_event *comm_event, *mmap_event; + int err = -1, thread, j; + + comm_event = malloc(sizeof(comm_event->comm) + machine->id_hdr_size); + if (comm_event == NULL) + goto out; + + mmap_event = malloc(sizeof(mmap_event->mmap) + machine->id_hdr_size); + if (mmap_event == NULL) + goto out_free_comm; + + err = 0; + for (thread = 0; thread < threads->nr; ++thread) { + if (__event__synthesize_thread(comm_event, mmap_event, + threads->map[thread], 0, + process, tool, machine)) { + err = -1; + break; + } + + /* + * comm.pid is set to thread group id by + * perf_event__synthesize_comm + */ + if ((int) comm_event->comm.pid != threads->map[thread]) { + bool need_leader = true; + + /* is thread group leader in thread_map? */ + for (j = 0; j < threads->nr; ++j) { + if ((int) comm_event->comm.pid == threads->map[j]) { + need_leader = false; + break; + } + } + + /* if not, generate events for it */ + if (need_leader && + __event__synthesize_thread(comm_event, + mmap_event, + comm_event->comm.pid, 0, + process, tool, machine)) { + err = -1; + break; + } + } + } + free(mmap_event); +out_free_comm: + free(comm_event); +out: + return err; +} + +int perf_event__synthesize_threads(struct perf_tool *tool, + perf_event__handler_t process, + struct machine *machine) +{ + DIR *proc; + struct dirent dirent, *next; + union perf_event *comm_event, *mmap_event; + int err = -1; + + comm_event = malloc(sizeof(comm_event->comm) + machine->id_hdr_size); + if (comm_event == NULL) + goto out; + + mmap_event = malloc(sizeof(mmap_event->mmap) + machine->id_hdr_size); + if (mmap_event == NULL) + goto out_free_comm; + + proc = opendir("/proc"); + if (proc == NULL) + goto out_free_mmap; + + while (!readdir_r(proc, &dirent, &next) && next) { + char *end; + pid_t pid = strtol(dirent.d_name, &end, 10); + + if (*end) /* only interested in proper numerical dirents */ + continue; + /* + * We may race with exiting thread, so don't stop just because + * one thread couldn't be synthesized. + */ + __event__synthesize_thread(comm_event, mmap_event, pid, 1, + process, tool, machine); + } + + err = 0; + closedir(proc); +out_free_mmap: + free(mmap_event); +out_free_comm: + free(comm_event); +out: + return err; +} + +struct process_symbol_args { + const char *name; + u64 start; +}; + +static int find_symbol_cb(void *arg, const char *name, char type, + u64 start) +{ + struct process_symbol_args *args = arg; + + /* + * Must be a function or at least an alias, as in PARISC64, where "_text" is + * an 'A' to the same address as "_stext". + */ + if (!(symbol_type__is_a(type, MAP__FUNCTION) || + type == 'A') || strcmp(name, args->name)) + return 0; + + args->start = start; + return 1; +} + +int perf_event__synthesize_kernel_mmap(struct perf_tool *tool, + perf_event__handler_t process, + struct machine *machine, + const char *symbol_name) +{ + size_t size; + const char *filename, *mmap_name; + char path[PATH_MAX]; + char name_buff[PATH_MAX]; + struct map *map; + int err; + /* + * We should get this from /sys/kernel/sections/.text, but till that is + * available use this, and after it is use this as a fallback for older + * kernels. + */ + struct process_symbol_args args = { .name = symbol_name, }; + union perf_event *event = zalloc((sizeof(event->mmap) + + machine->id_hdr_size)); + if (event == NULL) { + pr_debug("Not enough memory synthesizing mmap event " + "for kernel modules\n"); + return -1; + } + + mmap_name = machine__mmap_name(machine, name_buff, sizeof(name_buff)); + if (machine__is_host(machine)) { + /* + * kernel uses PERF_RECORD_MISC_USER for user space maps, + * see kernel/perf_event.c __perf_event_mmap + */ + event->header.misc = PERF_RECORD_MISC_KERNEL; + filename = "/proc/kallsyms"; + } else { + event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL; + if (machine__is_default_guest(machine)) + filename = (char *) symbol_conf.default_guest_kallsyms; + else { + sprintf(path, "%s/proc/kallsyms", machine->root_dir); + filename = path; + } + } + + if (kallsyms__parse(filename, &args, find_symbol_cb) <= 0) { + free(event); + return -ENOENT; + } + + map = machine->vmlinux_maps[MAP__FUNCTION]; + size = snprintf(event->mmap.filename, sizeof(event->mmap.filename), + "%s%s", mmap_name, symbol_name) + 1; + size = PERF_ALIGN(size, sizeof(u64)); + event->mmap.header.type = PERF_RECORD_MMAP; + event->mmap.header.size = (sizeof(event->mmap) - + (sizeof(event->mmap.filename) - size) + machine->id_hdr_size); + event->mmap.pgoff = args.start; + event->mmap.start = map->start; + event->mmap.len = map->end - event->mmap.start; + event->mmap.pid = machine->pid; + + err = process(tool, event, &synth_sample, machine); + free(event); + + return err; +} + +size_t perf_event__fprintf_comm(union perf_event *event, FILE *fp) +{ + return fprintf(fp, ": %s:%d\n", event->comm.comm, event->comm.tid); +} + +int perf_event__process_comm(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine) +{ + return machine__process_comm_event(machine, event); +} + +int perf_event__process_lost(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine) +{ + return machine__process_lost_event(machine, event); +} + +size_t perf_event__fprintf_mmap(union perf_event *event, FILE *fp) +{ + return fprintf(fp, " %d/%d: [%#" PRIx64 "(%#" PRIx64 ") @ %#" PRIx64 "]: %s\n", + event->mmap.pid, event->mmap.tid, event->mmap.start, + event->mmap.len, event->mmap.pgoff, event->mmap.filename); +} + +int perf_event__process_mmap(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine) +{ + return machine__process_mmap_event(machine, event); +} + +size_t perf_event__fprintf_task(union perf_event *event, FILE *fp) +{ + return fprintf(fp, "(%d:%d):(%d:%d)\n", + event->fork.pid, event->fork.tid, + event->fork.ppid, event->fork.ptid); +} + +int perf_event__process_fork(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine) +{ + return machine__process_fork_event(machine, event); +} + +int perf_event__process_exit(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine) +{ + return machine__process_exit_event(machine, event); +} + +size_t perf_event__fprintf(union perf_event *event, FILE *fp) +{ + size_t ret = fprintf(fp, "PERF_RECORD_%s", + perf_event__name(event->header.type)); + + switch (event->header.type) { + case PERF_RECORD_COMM: + ret += perf_event__fprintf_comm(event, fp); + break; + case PERF_RECORD_FORK: + case PERF_RECORD_EXIT: + ret += perf_event__fprintf_task(event, fp); + break; + case PERF_RECORD_MMAP: + ret += perf_event__fprintf_mmap(event, fp); + break; + default: + ret += fprintf(fp, "\n"); + } + + return ret; +} + +int perf_event__process(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample __maybe_unused, + struct machine *machine) +{ + return machine__process_event(machine, event); +} + +void thread__find_addr_map(struct thread *self, + struct machine *machine, u8 cpumode, + enum map_type type, u64 addr, + struct addr_location *al) +{ + struct map_groups *mg = &self->mg; + + al->thread = self; + al->addr = addr; + al->cpumode = cpumode; + al->filtered = false; + + if (machine == NULL) { + al->map = NULL; + return; + } + + if (cpumode == PERF_RECORD_MISC_KERNEL && perf_host) { + al->level = 'k'; + mg = &machine->kmaps; + } else if (cpumode == PERF_RECORD_MISC_USER && perf_host) { + al->level = '.'; + } else if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL && perf_guest) { + al->level = 'g'; + mg = &machine->kmaps; + } else { + /* + * 'u' means guest os user space. + * TODO: We don't support guest user space. Might support late. + */ + if (cpumode == PERF_RECORD_MISC_GUEST_USER && perf_guest) + al->level = 'u'; + else + al->level = 'H'; + al->map = NULL; + + if ((cpumode == PERF_RECORD_MISC_GUEST_USER || + cpumode == PERF_RECORD_MISC_GUEST_KERNEL) && + !perf_guest) + al->filtered = true; + if ((cpumode == PERF_RECORD_MISC_USER || + cpumode == PERF_RECORD_MISC_KERNEL) && + !perf_host) + al->filtered = true; + + return; + } +try_again: + al->map = map_groups__find(mg, type, al->addr); + if (al->map == NULL) { + /* + * If this is outside of all known maps, and is a negative + * address, try to look it up in the kernel dso, as it might be + * a vsyscall or vdso (which executes in user-mode). + * + * XXX This is nasty, we should have a symbol list in the + * "[vdso]" dso, but for now lets use the old trick of looking + * in the whole kernel symbol list. + */ + if ((long long)al->addr < 0 && + cpumode == PERF_RECORD_MISC_USER && + machine && mg != &machine->kmaps) { + mg = &machine->kmaps; + goto try_again; + } + } else + al->addr = al->map->map_ip(al->map, al->addr); +} + +void thread__find_addr_location(struct thread *thread, struct machine *machine, + u8 cpumode, enum map_type type, u64 addr, + struct addr_location *al, + symbol_filter_t filter) +{ + thread__find_addr_map(thread, machine, cpumode, type, addr, al); + if (al->map != NULL) + al->sym = map__find_symbol(al->map, al->addr, filter); + else + al->sym = NULL; +} + +int perf_event__preprocess_sample(const union perf_event *event, + struct machine *machine, + struct addr_location *al, + struct perf_sample *sample, + symbol_filter_t filter) +{ + u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; + struct thread *thread = machine__findnew_thread(machine, event->ip.pid); + + if (thread == NULL) + return -1; + + if (symbol_conf.comm_list && + !strlist__has_entry(symbol_conf.comm_list, thread->comm)) + goto out_filtered; + + dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid); + /* + * Have we already created the kernel maps for this machine? + * + * This should have happened earlier, when we processed the kernel MMAP + * events, but for older perf.data files there was no such thing, so do + * it now. + */ + if (cpumode == PERF_RECORD_MISC_KERNEL && + machine->vmlinux_maps[MAP__FUNCTION] == NULL) + machine__create_kernel_maps(machine); + + thread__find_addr_map(thread, machine, cpumode, MAP__FUNCTION, + event->ip.ip, al); + dump_printf(" ...... dso: %s\n", + al->map ? al->map->dso->long_name : + al->level == 'H' ? "[hypervisor]" : "<not found>"); + al->sym = NULL; + al->cpu = sample->cpu; + + if (al->map) { + struct dso *dso = al->map->dso; + + if (symbol_conf.dso_list && + (!dso || !(strlist__has_entry(symbol_conf.dso_list, + dso->short_name) || + (dso->short_name != dso->long_name && + strlist__has_entry(symbol_conf.dso_list, + dso->long_name))))) + goto out_filtered; + + al->sym = map__find_symbol(al->map, al->addr, filter); + } + + if (symbol_conf.sym_list && + (!al->sym || !strlist__has_entry(symbol_conf.sym_list, + al->sym->name))) + goto out_filtered; + + return 0; + +out_filtered: + al->filtered = true; + return 0; +} diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h new file mode 100644 index 00000000..0d573ff4 --- /dev/null +++ b/tools/perf/util/event.h @@ -0,0 +1,224 @@ +#ifndef __PERF_RECORD_H +#define __PERF_RECORD_H + +#include <limits.h> +#include <stdio.h> + +#include "../perf.h" +#include "map.h" +#include "build-id.h" + +/* + * PERF_SAMPLE_IP | PERF_SAMPLE_TID | * + */ +struct ip_event { + struct perf_event_header header; + u64 ip; + u32 pid, tid; + unsigned char __more_data[]; +}; + +struct mmap_event { + struct perf_event_header header; + u32 pid, tid; + u64 start; + u64 len; + u64 pgoff; + char filename[PATH_MAX]; +}; + +struct comm_event { + struct perf_event_header header; + u32 pid, tid; + char comm[16]; +}; + +struct fork_event { + struct perf_event_header header; + u32 pid, ppid; + u32 tid, ptid; + u64 time; +}; + +struct lost_event { + struct perf_event_header header; + u64 id; + u64 lost; +}; + +/* + * PERF_FORMAT_ENABLED | PERF_FORMAT_RUNNING | PERF_FORMAT_ID + */ +struct read_event { + struct perf_event_header header; + u32 pid, tid; + u64 value; + u64 time_enabled; + u64 time_running; + u64 id; +}; + + +#define PERF_SAMPLE_MASK \ + (PERF_SAMPLE_IP | PERF_SAMPLE_TID | \ + PERF_SAMPLE_TIME | PERF_SAMPLE_ADDR | \ + PERF_SAMPLE_ID | PERF_SAMPLE_STREAM_ID | \ + PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD) + +struct sample_event { + struct perf_event_header header; + u64 array[]; +}; + +struct regs_dump { + u64 *regs; +}; + +struct stack_dump { + u16 offset; + u64 size; + char *data; +}; + +struct perf_sample { + u64 ip; + u32 pid, tid; + u64 time; + u64 addr; + u64 id; + u64 stream_id; + u64 period; + u32 cpu; + u32 raw_size; + void *raw_data; + struct ip_callchain *callchain; + struct branch_stack *branch_stack; + struct regs_dump user_regs; + struct stack_dump user_stack; +}; + +struct build_id_event { + struct perf_event_header header; + pid_t pid; + u8 build_id[PERF_ALIGN(BUILD_ID_SIZE, sizeof(u64))]; + char filename[]; +}; + +enum perf_user_event_type { /* above any possible kernel type */ + PERF_RECORD_USER_TYPE_START = 64, + PERF_RECORD_HEADER_ATTR = 64, + PERF_RECORD_HEADER_EVENT_TYPE = 65, + PERF_RECORD_HEADER_TRACING_DATA = 66, + PERF_RECORD_HEADER_BUILD_ID = 67, + PERF_RECORD_FINISHED_ROUND = 68, + PERF_RECORD_HEADER_MAX +}; + +struct attr_event { + struct perf_event_header header; + struct perf_event_attr attr; + u64 id[]; +}; + +#define MAX_EVENT_NAME 64 + +struct perf_trace_event_type { + u64 event_id; + char name[MAX_EVENT_NAME]; +}; + +struct event_type_event { + struct perf_event_header header; + struct perf_trace_event_type event_type; +}; + +struct tracing_data_event { + struct perf_event_header header; + u32 size; +}; + +union perf_event { + struct perf_event_header header; + struct ip_event ip; + struct mmap_event mmap; + struct comm_event comm; + struct fork_event fork; + struct lost_event lost; + struct read_event read; + struct sample_event sample; + struct attr_event attr; + struct event_type_event event_type; + struct tracing_data_event tracing_data; + struct build_id_event build_id; +}; + +void perf_event__print_totals(void); + +struct perf_tool; +struct thread_map; + +typedef int (*perf_event__handler_t)(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct machine *machine); + +int perf_event__synthesize_thread_map(struct perf_tool *tool, + struct thread_map *threads, + perf_event__handler_t process, + struct machine *machine); +int perf_event__synthesize_threads(struct perf_tool *tool, + perf_event__handler_t process, + struct machine *machine); +int perf_event__synthesize_kernel_mmap(struct perf_tool *tool, + perf_event__handler_t process, + struct machine *machine, + const char *symbol_name); + +int perf_event__synthesize_modules(struct perf_tool *tool, + perf_event__handler_t process, + struct machine *machine); + +int perf_event__process_comm(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct machine *machine); +int perf_event__process_lost(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct machine *machine); +int perf_event__process_mmap(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct machine *machine); +int perf_event__process_fork(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct machine *machine); +int perf_event__process_exit(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct machine *machine); +int perf_event__process(struct perf_tool *tool, + union perf_event *event, + struct perf_sample *sample, + struct machine *machine); + +struct addr_location; +int perf_event__preprocess_sample(const union perf_event *self, + struct machine *machine, + struct addr_location *al, + struct perf_sample *sample, + symbol_filter_t filter); + +const char *perf_event__name(unsigned int id); + +int perf_event__synthesize_sample(union perf_event *event, u64 type, + const struct perf_sample *sample, + bool swapped); + +size_t perf_event__fprintf_comm(union perf_event *event, FILE *fp); +size_t perf_event__fprintf_mmap(union perf_event *event, FILE *fp); +size_t perf_event__fprintf_task(union perf_event *event, FILE *fp); +size_t perf_event__fprintf(union perf_event *event, FILE *fp); + +#endif /* __PERF_RECORD_H */ diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c new file mode 100644 index 00000000..c8be0fbc --- /dev/null +++ b/tools/perf/util/evlist.c @@ -0,0 +1,850 @@ +/* + * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com> + * + * Parts came from builtin-{top,stat,record}.c, see those files for further + * copyright notes. + * + * Released under the GPL v2. (and only v2, not any later version) + */ +#include "util.h" +#include "debugfs.h" +#include <poll.h> +#include "cpumap.h" +#include "thread_map.h" +#include "target.h" +#include "evlist.h" +#include "evsel.h" +#include <unistd.h> + +#include "parse-events.h" + +#include <sys/mman.h> + +#include <linux/bitops.h> +#include <linux/hash.h> + +#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y)) +#define SID(e, x, y) xyarray__entry(e->sample_id, x, y) + +void perf_evlist__init(struct perf_evlist *evlist, struct cpu_map *cpus, + struct thread_map *threads) +{ + int i; + + for (i = 0; i < PERF_EVLIST__HLIST_SIZE; ++i) + INIT_HLIST_HEAD(&evlist->heads[i]); + INIT_LIST_HEAD(&evlist->entries); + perf_evlist__set_maps(evlist, cpus, threads); + evlist->workload.pid = -1; +} + +struct perf_evlist *perf_evlist__new(struct cpu_map *cpus, + struct thread_map *threads) +{ + struct perf_evlist *evlist = zalloc(sizeof(*evlist)); + + if (evlist != NULL) + perf_evlist__init(evlist, cpus, threads); + + return evlist; +} + +void perf_evlist__config(struct perf_evlist *evlist, + struct perf_record_opts *opts) +{ + struct perf_evsel *evsel; + /* + * Set the evsel leader links before we configure attributes, + * since some might depend on this info. + */ + if (opts->group) + perf_evlist__set_leader(evlist); + + if (evlist->cpus->map[0] < 0) + opts->no_inherit = true; + + list_for_each_entry(evsel, &evlist->entries, node) { + perf_evsel__config(evsel, opts); + + if (evlist->nr_entries > 1) + perf_evsel__set_sample_id(evsel); + } +} + +static void perf_evlist__purge(struct perf_evlist *evlist) +{ + struct perf_evsel *pos, *n; + + list_for_each_entry_safe(pos, n, &evlist->entries, node) { + list_del_init(&pos->node); + perf_evsel__delete(pos); + } + + evlist->nr_entries = 0; +} + +void perf_evlist__exit(struct perf_evlist *evlist) +{ + free(evlist->mmap); + free(evlist->pollfd); + evlist->mmap = NULL; + evlist->pollfd = NULL; +} + +void perf_evlist__delete(struct perf_evlist *evlist) +{ + perf_evlist__purge(evlist); + perf_evlist__exit(evlist); + free(evlist); +} + +void perf_evlist__add(struct perf_evlist *evlist, struct perf_evsel *entry) +{ + list_add_tail(&entry->node, &evlist->entries); + ++evlist->nr_entries; +} + +void perf_evlist__splice_list_tail(struct perf_evlist *evlist, + struct list_head *list, + int nr_entries) +{ + list_splice_tail(list, &evlist->entries); + evlist->nr_entries += nr_entries; +} + +void __perf_evlist__set_leader(struct list_head *list) +{ + struct perf_evsel *evsel, *leader; + + leader = list_entry(list->next, struct perf_evsel, node); + evsel = list_entry(list->prev, struct perf_evsel, node); + + leader->nr_members = evsel->idx - leader->idx + 1; + + list_for_each_entry(evsel, list, node) { + evsel->leader = leader; + } +} + +void perf_evlist__set_leader(struct perf_evlist *evlist) +{ + if (evlist->nr_entries) { + evlist->nr_groups = evlist->nr_entries > 1 ? 1 : 0; + __perf_evlist__set_leader(&evlist->entries); + } +} + +int perf_evlist__add_default(struct perf_evlist *evlist) +{ + struct perf_event_attr attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES, + }; + struct perf_evsel *evsel; + + event_attr_init(&attr); + + evsel = perf_evsel__new(&attr, 0); + if (evsel == NULL) + goto error; + + /* use strdup() because free(evsel) assumes name is allocated */ + evsel->name = strdup("cycles"); + if (!evsel->name) + goto error_free; + + perf_evlist__add(evlist, evsel); + return 0; +error_free: + perf_evsel__delete(evsel); +error: + return -ENOMEM; +} + +static int perf_evlist__add_attrs(struct perf_evlist *evlist, + struct perf_event_attr *attrs, size_t nr_attrs) +{ + struct perf_evsel *evsel, *n; + LIST_HEAD(head); + size_t i; + + for (i = 0; i < nr_attrs; i++) { + evsel = perf_evsel__new(attrs + i, evlist->nr_entries + i); + if (evsel == NULL) + goto out_delete_partial_list; + list_add_tail(&evsel->node, &head); + } + + perf_evlist__splice_list_tail(evlist, &head, nr_attrs); + + return 0; + +out_delete_partial_list: + list_for_each_entry_safe(evsel, n, &head, node) + perf_evsel__delete(evsel); + return -1; +} + +int __perf_evlist__add_default_attrs(struct perf_evlist *evlist, + struct perf_event_attr *attrs, size_t nr_attrs) +{ + size_t i; + + for (i = 0; i < nr_attrs; i++) + event_attr_init(attrs + i); + + return perf_evlist__add_attrs(evlist, attrs, nr_attrs); +} + +struct perf_evsel * +perf_evlist__find_tracepoint_by_id(struct perf_evlist *evlist, int id) +{ + struct perf_evsel *evsel; + + list_for_each_entry(evsel, &evlist->entries, node) { + if (evsel->attr.type == PERF_TYPE_TRACEPOINT && + (int)evsel->attr.config == id) + return evsel; + } + + return NULL; +} + +int perf_evlist__add_newtp(struct perf_evlist *evlist, + const char *sys, const char *name, void *handler) +{ + struct perf_evsel *evsel; + + evsel = perf_evsel__newtp(sys, name, evlist->nr_entries); + if (evsel == NULL) + return -1; + + evsel->handler.func = handler; + perf_evlist__add(evlist, evsel); + return 0; +} + +void perf_evlist__disable(struct perf_evlist *evlist) +{ + int cpu, thread; + struct perf_evsel *pos; + + for (cpu = 0; cpu < evlist->cpus->nr; cpu++) { + list_for_each_entry(pos, &evlist->entries, node) { + if (!perf_evsel__is_group_leader(pos)) + continue; + for (thread = 0; thread < evlist->threads->nr; thread++) + ioctl(FD(pos, cpu, thread), + PERF_EVENT_IOC_DISABLE, 0); + } + } +} + +void perf_evlist__enable(struct perf_evlist *evlist) +{ + int cpu, thread; + struct perf_evsel *pos; + + for (cpu = 0; cpu < cpu_map__nr(evlist->cpus); cpu++) { + list_for_each_entry(pos, &evlist->entries, node) { + if (!perf_evsel__is_group_leader(pos)) + continue; + for (thread = 0; thread < evlist->threads->nr; thread++) + ioctl(FD(pos, cpu, thread), + PERF_EVENT_IOC_ENABLE, 0); + } + } +} + +static int perf_evlist__alloc_pollfd(struct perf_evlist *evlist) +{ + int nfds = cpu_map__nr(evlist->cpus) * evlist->threads->nr * evlist->nr_entries; + evlist->pollfd = malloc(sizeof(struct pollfd) * nfds); + return evlist->pollfd != NULL ? 0 : -ENOMEM; +} + +void perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd) +{ + fcntl(fd, F_SETFL, O_NONBLOCK); + evlist->pollfd[evlist->nr_fds].fd = fd; + evlist->pollfd[evlist->nr_fds].events = POLLIN; + evlist->nr_fds++; +} + +static void perf_evlist__id_hash(struct perf_evlist *evlist, + struct perf_evsel *evsel, + int cpu, int thread, u64 id) +{ + int hash; + struct perf_sample_id *sid = SID(evsel, cpu, thread); + + sid->id = id; + sid->evsel = evsel; + hash = hash_64(sid->id, PERF_EVLIST__HLIST_BITS); + hlist_add_head(&sid->node, &evlist->heads[hash]); +} + +void perf_evlist__id_add(struct perf_evlist *evlist, struct perf_evsel *evsel, + int cpu, int thread, u64 id) +{ + perf_evlist__id_hash(evlist, evsel, cpu, thread, id); + evsel->id[evsel->ids++] = id; +} + +static int perf_evlist__id_add_fd(struct perf_evlist *evlist, + struct perf_evsel *evsel, + int cpu, int thread, int fd) +{ + u64 read_data[4] = { 0, }; + int id_idx = 1; /* The first entry is the counter value */ + + if (!(evsel->attr.read_format & PERF_FORMAT_ID) || + read(fd, &read_data, sizeof(read_data)) == -1) + return -1; + + if (evsel->attr.read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) + ++id_idx; + if (evsel->attr.read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) + ++id_idx; + + perf_evlist__id_add(evlist, evsel, cpu, thread, read_data[id_idx]); + return 0; +} + +struct perf_evsel *perf_evlist__id2evsel(struct perf_evlist *evlist, u64 id) +{ + struct hlist_head *head; + struct perf_sample_id *sid; + int hash; + + if (evlist->nr_entries == 1) + return perf_evlist__first(evlist); + + hash = hash_64(id, PERF_EVLIST__HLIST_BITS); + head = &evlist->heads[hash]; + + hlist_for_each_entry(sid, head, node) + if (sid->id == id) + return sid->evsel; + + if (!perf_evlist__sample_id_all(evlist)) + return perf_evlist__first(evlist); + + return NULL; +} + +union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx) +{ + struct perf_mmap *md = &evlist->mmap[idx]; + unsigned int head = perf_mmap__read_head(md); + unsigned int old = md->prev; + unsigned char *data = md->base + page_size; + union perf_event *event = NULL; + + if (evlist->overwrite) { + /* + * 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. + */ + int diff = head - old; + if (diff > md->mask / 2 || diff < 0) { + fprintf(stderr, "WARNING: failed to keep up with mmap data.\n"); + + /* + * head points to a known good entry, start there. + */ + old = head; + } + } + + if (old != head) { + size_t size; + + event = (union perf_event *)&data[old & md->mask]; + 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 = &md->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 = &md->event_copy; + } + + old += size; + } + + md->prev = old; + + if (!evlist->overwrite) + perf_mmap__write_tail(md, old); + + return event; +} + +void perf_evlist__munmap(struct perf_evlist *evlist) +{ + int i; + + for (i = 0; i < evlist->nr_mmaps; i++) { + if (evlist->mmap[i].base != NULL) { + munmap(evlist->mmap[i].base, evlist->mmap_len); + evlist->mmap[i].base = NULL; + } + } + + free(evlist->mmap); + evlist->mmap = NULL; +} + +static int perf_evlist__alloc_mmap(struct perf_evlist *evlist) +{ + evlist->nr_mmaps = cpu_map__nr(evlist->cpus); + if (cpu_map__all(evlist->cpus)) + evlist->nr_mmaps = evlist->threads->nr; + evlist->mmap = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap)); + return evlist->mmap != NULL ? 0 : -ENOMEM; +} + +static int __perf_evlist__mmap(struct perf_evlist *evlist, + int idx, int prot, int mask, int fd) +{ + evlist->mmap[idx].prev = 0; + evlist->mmap[idx].mask = mask; + evlist->mmap[idx].base = mmap(NULL, evlist->mmap_len, prot, + MAP_SHARED, fd, 0); + if (evlist->mmap[idx].base == MAP_FAILED) { + evlist->mmap[idx].base = NULL; + return -1; + } + + perf_evlist__add_pollfd(evlist, fd); + return 0; +} + +static int perf_evlist__mmap_per_cpu(struct perf_evlist *evlist, int prot, int mask) +{ + struct perf_evsel *evsel; + int cpu, thread; + + for (cpu = 0; cpu < evlist->cpus->nr; cpu++) { + int output = -1; + + for (thread = 0; thread < evlist->threads->nr; thread++) { + list_for_each_entry(evsel, &evlist->entries, node) { + int fd = FD(evsel, cpu, thread); + + if (output == -1) { + output = fd; + if (__perf_evlist__mmap(evlist, cpu, + prot, mask, output) < 0) + goto out_unmap; + } else { + if (ioctl(fd, PERF_EVENT_IOC_SET_OUTPUT, output) != 0) + goto out_unmap; + } + + if ((evsel->attr.read_format & PERF_FORMAT_ID) && + perf_evlist__id_add_fd(evlist, evsel, cpu, thread, fd) < 0) + goto out_unmap; + } + } + } + + return 0; + +out_unmap: + for (cpu = 0; cpu < evlist->cpus->nr; cpu++) { + if (evlist->mmap[cpu].base != NULL) { + munmap(evlist->mmap[cpu].base, evlist->mmap_len); + evlist->mmap[cpu].base = NULL; + } + } + return -1; +} + +static int perf_evlist__mmap_per_thread(struct perf_evlist *evlist, int prot, int mask) +{ + struct perf_evsel *evsel; + int thread; + + for (thread = 0; thread < evlist->threads->nr; thread++) { + int output = -1; + + list_for_each_entry(evsel, &evlist->entries, node) { + int fd = FD(evsel, 0, thread); + + if (output == -1) { + output = fd; + if (__perf_evlist__mmap(evlist, thread, + prot, mask, output) < 0) + goto out_unmap; + } else { + if (ioctl(fd, PERF_EVENT_IOC_SET_OUTPUT, output) != 0) + goto out_unmap; + } + + if ((evsel->attr.read_format & PERF_FORMAT_ID) && + perf_evlist__id_add_fd(evlist, evsel, 0, thread, fd) < 0) + goto out_unmap; + } + } + + return 0; + +out_unmap: + for (thread = 0; thread < evlist->threads->nr; thread++) { + if (evlist->mmap[thread].base != NULL) { + munmap(evlist->mmap[thread].base, evlist->mmap_len); + evlist->mmap[thread].base = NULL; + } + } + return -1; +} + +/** perf_evlist__mmap - Create per cpu maps to receive events + * + * @evlist - list of events + * @pages - map length in pages + * @overwrite - overwrite older events? + * + * If overwrite is false the user needs to signal event consuption using: + * + * struct perf_mmap *m = &evlist->mmap[cpu]; + * unsigned int head = perf_mmap__read_head(m); + * + * perf_mmap__write_tail(m, head) + * + * Using perf_evlist__read_on_cpu does this automatically. + */ +int perf_evlist__mmap(struct perf_evlist *evlist, unsigned int pages, + bool overwrite) +{ + struct perf_evsel *evsel; + const struct cpu_map *cpus = evlist->cpus; + const struct thread_map *threads = evlist->threads; + int prot = PROT_READ | (overwrite ? 0 : PROT_WRITE), mask; + + /* 512 kiB: default amount of unprivileged mlocked memory */ + if (pages == UINT_MAX) + pages = (512 * 1024) / page_size; + else if (!is_power_of_2(pages)) + return -EINVAL; + + mask = pages * page_size - 1; + + if (evlist->mmap == NULL && perf_evlist__alloc_mmap(evlist) < 0) + return -ENOMEM; + + if (evlist->pollfd == NULL && perf_evlist__alloc_pollfd(evlist) < 0) + return -ENOMEM; + + evlist->overwrite = overwrite; + evlist->mmap_len = (pages + 1) * page_size; + + list_for_each_entry(evsel, &evlist->entries, node) { + if ((evsel->attr.read_format & PERF_FORMAT_ID) && + evsel->sample_id == NULL && + perf_evsel__alloc_id(evsel, cpu_map__nr(cpus), threads->nr) < 0) + return -ENOMEM; + } + + if (cpu_map__all(cpus)) + return perf_evlist__mmap_per_thread(evlist, prot, mask); + + return perf_evlist__mmap_per_cpu(evlist, prot, mask); +} + +int perf_evlist__create_maps(struct perf_evlist *evlist, + struct perf_target *target) +{ + evlist->threads = thread_map__new_str(target->pid, target->tid, + target->uid); + + if (evlist->threads == NULL) + return -1; + + if (perf_target__has_task(target)) + evlist->cpus = cpu_map__dummy_new(); + else if (!perf_target__has_cpu(target) && !target->uses_mmap) + evlist->cpus = cpu_map__dummy_new(); + else + evlist->cpus = cpu_map__new(target->cpu_list); + + if (evlist->cpus == NULL) + goto out_delete_threads; + + return 0; + +out_delete_threads: + thread_map__delete(evlist->threads); + return -1; +} + +void perf_evlist__delete_maps(struct perf_evlist *evlist) +{ + cpu_map__delete(evlist->cpus); + thread_map__delete(evlist->threads); + evlist->cpus = NULL; + evlist->threads = NULL; +} + +int perf_evlist__apply_filters(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel; + int err = 0; + const int ncpus = cpu_map__nr(evlist->cpus), + nthreads = evlist->threads->nr; + + list_for_each_entry(evsel, &evlist->entries, node) { + if (evsel->filter == NULL) + continue; + + err = perf_evsel__set_filter(evsel, ncpus, nthreads, evsel->filter); + if (err) + break; + } + + return err; +} + +int perf_evlist__set_filter(struct perf_evlist *evlist, const char *filter) +{ + struct perf_evsel *evsel; + int err = 0; + const int ncpus = cpu_map__nr(evlist->cpus), + nthreads = evlist->threads->nr; + + list_for_each_entry(evsel, &evlist->entries, node) { + err = perf_evsel__set_filter(evsel, ncpus, nthreads, filter); + if (err) + break; + } + + return err; +} + +bool perf_evlist__valid_sample_type(struct perf_evlist *evlist) +{ + struct perf_evsel *first = perf_evlist__first(evlist), *pos = first; + + list_for_each_entry_continue(pos, &evlist->entries, node) { + if (first->attr.sample_type != pos->attr.sample_type) + return false; + } + + return true; +} + +u64 perf_evlist__sample_type(struct perf_evlist *evlist) +{ + struct perf_evsel *first = perf_evlist__first(evlist); + return first->attr.sample_type; +} + +u16 perf_evlist__id_hdr_size(struct perf_evlist *evlist) +{ + struct perf_evsel *first = perf_evlist__first(evlist); + struct perf_sample *data; + u64 sample_type; + u16 size = 0; + + if (!first->attr.sample_id_all) + goto out; + + sample_type = first->attr.sample_type; + + if (sample_type & PERF_SAMPLE_TID) + size += sizeof(data->tid) * 2; + + if (sample_type & PERF_SAMPLE_TIME) + size += sizeof(data->time); + + if (sample_type & PERF_SAMPLE_ID) + size += sizeof(data->id); + + if (sample_type & PERF_SAMPLE_STREAM_ID) + size += sizeof(data->stream_id); + + if (sample_type & PERF_SAMPLE_CPU) + size += sizeof(data->cpu) * 2; +out: + return size; +} + +bool perf_evlist__valid_sample_id_all(struct perf_evlist *evlist) +{ + struct perf_evsel *first = perf_evlist__first(evlist), *pos = first; + + list_for_each_entry_continue(pos, &evlist->entries, node) { + if (first->attr.sample_id_all != pos->attr.sample_id_all) + return false; + } + + return true; +} + +bool perf_evlist__sample_id_all(struct perf_evlist *evlist) +{ + struct perf_evsel *first = perf_evlist__first(evlist); + return first->attr.sample_id_all; +} + +void perf_evlist__set_selected(struct perf_evlist *evlist, + struct perf_evsel *evsel) +{ + evlist->selected = evsel; +} + +int perf_evlist__open(struct perf_evlist *evlist) +{ + struct perf_evsel *evsel; + int err, ncpus, nthreads; + + list_for_each_entry(evsel, &evlist->entries, node) { + err = perf_evsel__open(evsel, evlist->cpus, evlist->threads); + if (err < 0) + goto out_err; + } + + return 0; +out_err: + ncpus = evlist->cpus ? evlist->cpus->nr : 1; + nthreads = evlist->threads ? evlist->threads->nr : 1; + + list_for_each_entry_reverse(evsel, &evlist->entries, node) + perf_evsel__close(evsel, ncpus, nthreads); + + errno = -err; + return err; +} + +int perf_evlist__prepare_workload(struct perf_evlist *evlist, + struct perf_record_opts *opts, + const char *argv[]) +{ + int child_ready_pipe[2], go_pipe[2]; + char bf; + + if (pipe(child_ready_pipe) < 0) { + perror("failed to create 'ready' pipe"); + return -1; + } + + if (pipe(go_pipe) < 0) { + perror("failed to create 'go' pipe"); + goto out_close_ready_pipe; + } + + evlist->workload.pid = fork(); + if (evlist->workload.pid < 0) { + perror("failed to fork"); + goto out_close_pipes; + } + + if (!evlist->workload.pid) { + if (opts->pipe_output) + dup2(2, 1); + + close(child_ready_pipe[0]); + close(go_pipe[1]); + fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC); + + /* + * Do a dummy execvp to get the PLT entry resolved, + * so we avoid the resolver overhead on the real + * execvp call. + */ + execvp("", (char **)argv); + + /* + * Tell the parent we're ready to go + */ + close(child_ready_pipe[1]); + + /* + * Wait until the parent tells us to go. + */ + if (read(go_pipe[0], &bf, 1) == -1) + perror("unable to read pipe"); + + execvp(argv[0], (char **)argv); + + perror(argv[0]); + kill(getppid(), SIGUSR1); + exit(-1); + } + + if (perf_target__none(&opts->target)) + evlist->threads->map[0] = evlist->workload.pid; + + close(child_ready_pipe[1]); + close(go_pipe[0]); + /* + * wait for child to settle + */ + if (read(child_ready_pipe[0], &bf, 1) == -1) { + perror("unable to read pipe"); + goto out_close_pipes; + } + + evlist->workload.cork_fd = go_pipe[1]; + close(child_ready_pipe[0]); + return 0; + +out_close_pipes: + close(go_pipe[0]); + close(go_pipe[1]); +out_close_ready_pipe: + close(child_ready_pipe[0]); + close(child_ready_pipe[1]); + return -1; +} + +int perf_evlist__start_workload(struct perf_evlist *evlist) +{ + if (evlist->workload.cork_fd > 0) { + /* + * Remove the cork, let it rip! + */ + return close(evlist->workload.cork_fd); + } + + return 0; +} + +int perf_evlist__parse_sample(struct perf_evlist *evlist, union perf_event *event, + struct perf_sample *sample) +{ + struct perf_evsel *evsel = perf_evlist__first(evlist); + return perf_evsel__parse_sample(evsel, event, sample); +} + +size_t perf_evlist__fprintf(struct perf_evlist *evlist, FILE *fp) +{ + struct perf_evsel *evsel; + size_t printed = 0; + + list_for_each_entry(evsel, &evlist->entries, node) { + printed += fprintf(fp, "%s%s", evsel->idx ? ", " : "", + perf_evsel__name(evsel)); + } + + return printed + fprintf(fp, "\n");; +} diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h new file mode 100644 index 00000000..2dd07bd6 --- /dev/null +++ b/tools/perf/util/evlist.h @@ -0,0 +1,166 @@ +#ifndef __PERF_EVLIST_H +#define __PERF_EVLIST_H 1 + +#include <linux/list.h> +#include <stdio.h> +#include "../perf.h" +#include "event.h" +#include "evsel.h" +#include "util.h" +#include <unistd.h> + +struct pollfd; +struct thread_map; +struct cpu_map; +struct perf_record_opts; + +#define PERF_EVLIST__HLIST_BITS 8 +#define PERF_EVLIST__HLIST_SIZE (1 << PERF_EVLIST__HLIST_BITS) + +struct perf_mmap { + void *base; + int mask; + unsigned int prev; + union perf_event event_copy; +}; + +struct perf_evlist { + struct list_head entries; + struct hlist_head heads[PERF_EVLIST__HLIST_SIZE]; + int nr_entries; + int nr_groups; + int nr_fds; + int nr_mmaps; + int mmap_len; + struct { + int cork_fd; + pid_t pid; + } workload; + bool overwrite; + struct perf_mmap *mmap; + struct pollfd *pollfd; + struct thread_map *threads; + struct cpu_map *cpus; + struct perf_evsel *selected; +}; + +struct perf_evsel_str_handler { + const char *name; + void *handler; +}; + +struct perf_evlist *perf_evlist__new(struct cpu_map *cpus, + struct thread_map *threads); +void perf_evlist__init(struct perf_evlist *evlist, struct cpu_map *cpus, + struct thread_map *threads); +void perf_evlist__exit(struct perf_evlist *evlist); +void perf_evlist__delete(struct perf_evlist *evlist); + +void perf_evlist__add(struct perf_evlist *evlist, struct perf_evsel *entry); +int perf_evlist__add_default(struct perf_evlist *evlist); +int __perf_evlist__add_default_attrs(struct perf_evlist *evlist, + struct perf_event_attr *attrs, size_t nr_attrs); + +#define perf_evlist__add_default_attrs(evlist, array) \ + __perf_evlist__add_default_attrs(evlist, array, ARRAY_SIZE(array)) + +int perf_evlist__add_newtp(struct perf_evlist *evlist, + const char *sys, const char *name, void *handler); + +int perf_evlist__set_filter(struct perf_evlist *evlist, const char *filter); + +struct perf_evsel * +perf_evlist__find_tracepoint_by_id(struct perf_evlist *evlist, int id); + +void perf_evlist__id_add(struct perf_evlist *evlist, struct perf_evsel *evsel, + int cpu, int thread, u64 id); + +void perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd); + +struct perf_evsel *perf_evlist__id2evsel(struct perf_evlist *evlist, u64 id); + +union perf_event *perf_evlist__mmap_read(struct perf_evlist *self, int idx); + +int perf_evlist__open(struct perf_evlist *evlist); + +void perf_evlist__config(struct perf_evlist *evlist, + struct perf_record_opts *opts); + +int perf_evlist__prepare_workload(struct perf_evlist *evlist, + struct perf_record_opts *opts, + const char *argv[]); +int perf_evlist__start_workload(struct perf_evlist *evlist); + +int perf_evlist__mmap(struct perf_evlist *evlist, unsigned int pages, + bool overwrite); +void perf_evlist__munmap(struct perf_evlist *evlist); + +void perf_evlist__disable(struct perf_evlist *evlist); +void perf_evlist__enable(struct perf_evlist *evlist); + +void perf_evlist__set_selected(struct perf_evlist *evlist, + struct perf_evsel *evsel); + +static inline void perf_evlist__set_maps(struct perf_evlist *evlist, + struct cpu_map *cpus, + struct thread_map *threads) +{ + evlist->cpus = cpus; + evlist->threads = threads; +} + +int perf_evlist__create_maps(struct perf_evlist *evlist, + struct perf_target *target); +void perf_evlist__delete_maps(struct perf_evlist *evlist); +int perf_evlist__apply_filters(struct perf_evlist *evlist); + +void __perf_evlist__set_leader(struct list_head *list); +void perf_evlist__set_leader(struct perf_evlist *evlist); + +u64 perf_evlist__sample_type(struct perf_evlist *evlist); +bool perf_evlist__sample_id_all(struct perf_evlist *evlist); +u16 perf_evlist__id_hdr_size(struct perf_evlist *evlist); + +int perf_evlist__parse_sample(struct perf_evlist *evlist, union perf_event *event, + struct perf_sample *sample); + +bool perf_evlist__valid_sample_type(struct perf_evlist *evlist); +bool perf_evlist__valid_sample_id_all(struct perf_evlist *evlist); + +void perf_evlist__splice_list_tail(struct perf_evlist *evlist, + struct list_head *list, + int nr_entries); + +static inline struct perf_evsel *perf_evlist__first(struct perf_evlist *evlist) +{ + return list_entry(evlist->entries.next, struct perf_evsel, node); +} + +static inline struct perf_evsel *perf_evlist__last(struct perf_evlist *evlist) +{ + return list_entry(evlist->entries.prev, struct perf_evsel, node); +} + +size_t perf_evlist__fprintf(struct perf_evlist *evlist, FILE *fp); + +static inline unsigned int perf_mmap__read_head(struct perf_mmap *mm) +{ + struct perf_event_mmap_page *pc = mm->base; + int head = pc->data_head; + rmb(); + return head; +} + +static inline void perf_mmap__write_tail(struct perf_mmap *md, + unsigned long tail) +{ + struct perf_event_mmap_page *pc = md->base; + + /* + * ensure all reads are done before we write the tail out. + */ + /* mb(); */ + pc->data_tail = tail; +} + +#endif /* __PERF_EVLIST_H */ diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c new file mode 100644 index 00000000..9c82f98f --- /dev/null +++ b/tools/perf/util/evsel.c @@ -0,0 +1,1533 @@ +/* + * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com> + * + * Parts came from builtin-{top,stat,record}.c, see those files for further + * copyright notes. + * + * Released under the GPL v2. (and only v2, not any later version) + */ + +#include <byteswap.h> +#include <linux/bitops.h> +#include "asm/bug.h" +#include "debugfs.h" +#include "event-parse.h" +#include "evsel.h" +#include "evlist.h" +#include "util.h" +#include "cpumap.h" +#include "thread_map.h" +#include "target.h" +#include <linux/hw_breakpoint.h> +#include <linux/perf_event.h> +#include "perf_regs.h" + +static struct { + bool sample_id_all; + bool exclude_guest; +} perf_missing_features; + +#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y)) + +static int __perf_evsel__sample_size(u64 sample_type) +{ + u64 mask = sample_type & PERF_SAMPLE_MASK; + int size = 0; + int i; + + for (i = 0; i < 64; i++) { + if (mask & (1ULL << i)) + size++; + } + + size *= sizeof(u64); + + return size; +} + +void hists__init(struct hists *hists) +{ + memset(hists, 0, sizeof(*hists)); + hists->entries_in_array[0] = hists->entries_in_array[1] = RB_ROOT; + hists->entries_in = &hists->entries_in_array[0]; + hists->entries_collapsed = RB_ROOT; + hists->entries = RB_ROOT; + pthread_mutex_init(&hists->lock, NULL); +} + +void __perf_evsel__set_sample_bit(struct perf_evsel *evsel, + enum perf_event_sample_format bit) +{ + if (!(evsel->attr.sample_type & bit)) { + evsel->attr.sample_type |= bit; + evsel->sample_size += sizeof(u64); + } +} + +void __perf_evsel__reset_sample_bit(struct perf_evsel *evsel, + enum perf_event_sample_format bit) +{ + if (evsel->attr.sample_type & bit) { + evsel->attr.sample_type &= ~bit; + evsel->sample_size -= sizeof(u64); + } +} + +void perf_evsel__set_sample_id(struct perf_evsel *evsel) +{ + perf_evsel__set_sample_bit(evsel, ID); + evsel->attr.read_format |= PERF_FORMAT_ID; +} + +void perf_evsel__init(struct perf_evsel *evsel, + struct perf_event_attr *attr, int idx) +{ + evsel->idx = idx; + evsel->attr = *attr; + evsel->leader = evsel; + INIT_LIST_HEAD(&evsel->node); + hists__init(&evsel->hists); + evsel->sample_size = __perf_evsel__sample_size(attr->sample_type); +} + +struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx) +{ + struct perf_evsel *evsel = zalloc(sizeof(*evsel)); + + if (evsel != NULL) + perf_evsel__init(evsel, attr, idx); + + return evsel; +} + +struct event_format *event_format__new(const char *sys, const char *name) +{ + int fd, n; + char *filename; + void *bf = NULL, *nbf; + size_t size = 0, alloc_size = 0; + struct event_format *format = NULL; + + if (asprintf(&filename, "%s/%s/%s/format", tracing_events_path, sys, name) < 0) + goto out; + + fd = open(filename, O_RDONLY); + if (fd < 0) + goto out_free_filename; + + do { + if (size == alloc_size) { + alloc_size += BUFSIZ; + nbf = realloc(bf, alloc_size); + if (nbf == NULL) + goto out_free_bf; + bf = nbf; + } + + n = read(fd, bf + size, BUFSIZ); + if (n < 0) + goto out_free_bf; + size += n; + } while (n > 0); + + pevent_parse_format(&format, bf, size, sys); + +out_free_bf: + free(bf); + close(fd); +out_free_filename: + free(filename); +out: + return format; +} + +struct perf_evsel *perf_evsel__newtp(const char *sys, const char *name, int idx) +{ + struct perf_evsel *evsel = zalloc(sizeof(*evsel)); + + if (evsel != NULL) { + struct perf_event_attr attr = { + .type = PERF_TYPE_TRACEPOINT, + .sample_type = (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME | + PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD), + }; + + if (asprintf(&evsel->name, "%s:%s", sys, name) < 0) + goto out_free; + + evsel->tp_format = event_format__new(sys, name); + if (evsel->tp_format == NULL) + goto out_free; + + event_attr_init(&attr); + attr.config = evsel->tp_format->id; + attr.sample_period = 1; + perf_evsel__init(evsel, &attr, idx); + } + + return evsel; + +out_free: + free(evsel->name); + free(evsel); + return NULL; +} + +const char *perf_evsel__hw_names[PERF_COUNT_HW_MAX] = { + "cycles", + "instructions", + "cache-references", + "cache-misses", + "branches", + "branch-misses", + "bus-cycles", + "stalled-cycles-frontend", + "stalled-cycles-backend", + "ref-cycles", +}; + +static const char *__perf_evsel__hw_name(u64 config) +{ + if (config < PERF_COUNT_HW_MAX && perf_evsel__hw_names[config]) + return perf_evsel__hw_names[config]; + + return "unknown-hardware"; +} + +static int perf_evsel__add_modifiers(struct perf_evsel *evsel, char *bf, size_t size) +{ + int colon = 0, r = 0; + struct perf_event_attr *attr = &evsel->attr; + bool exclude_guest_default = false; + +#define MOD_PRINT(context, mod) do { \ + if (!attr->exclude_##context) { \ + if (!colon) colon = ++r; \ + r += scnprintf(bf + r, size - r, "%c", mod); \ + } } while(0) + + if (attr->exclude_kernel || attr->exclude_user || attr->exclude_hv) { + MOD_PRINT(kernel, 'k'); + MOD_PRINT(user, 'u'); + MOD_PRINT(hv, 'h'); + exclude_guest_default = true; + } + + if (attr->precise_ip) { + if (!colon) + colon = ++r; + r += scnprintf(bf + r, size - r, "%.*s", attr->precise_ip, "ppp"); + exclude_guest_default = true; + } + + if (attr->exclude_host || attr->exclude_guest == exclude_guest_default) { + MOD_PRINT(host, 'H'); + MOD_PRINT(guest, 'G'); + } +#undef MOD_PRINT + if (colon) + bf[colon - 1] = ':'; + return r; +} + +static int perf_evsel__hw_name(struct perf_evsel *evsel, char *bf, size_t size) +{ + int r = scnprintf(bf, size, "%s", __perf_evsel__hw_name(evsel->attr.config)); + return r + perf_evsel__add_modifiers(evsel, bf + r, size - r); +} + +const char *perf_evsel__sw_names[PERF_COUNT_SW_MAX] = { + "cpu-clock", + "task-clock", + "page-faults", + "context-switches", + "cpu-migrations", + "minor-faults", + "major-faults", + "alignment-faults", + "emulation-faults", +}; + +static const char *__perf_evsel__sw_name(u64 config) +{ + if (config < PERF_COUNT_SW_MAX && perf_evsel__sw_names[config]) + return perf_evsel__sw_names[config]; + return "unknown-software"; +} + +static int perf_evsel__sw_name(struct perf_evsel *evsel, char *bf, size_t size) +{ + int r = scnprintf(bf, size, "%s", __perf_evsel__sw_name(evsel->attr.config)); + return r + perf_evsel__add_modifiers(evsel, bf + r, size - r); +} + +static int __perf_evsel__bp_name(char *bf, size_t size, u64 addr, u64 type) +{ + int r; + + r = scnprintf(bf, size, "mem:0x%" PRIx64 ":", addr); + + if (type & HW_BREAKPOINT_R) + r += scnprintf(bf + r, size - r, "r"); + + if (type & HW_BREAKPOINT_W) + r += scnprintf(bf + r, size - r, "w"); + + if (type & HW_BREAKPOINT_X) + r += scnprintf(bf + r, size - r, "x"); + + return r; +} + +static int perf_evsel__bp_name(struct perf_evsel *evsel, char *bf, size_t size) +{ + struct perf_event_attr *attr = &evsel->attr; + int r = __perf_evsel__bp_name(bf, size, attr->bp_addr, attr->bp_type); + return r + perf_evsel__add_modifiers(evsel, bf + r, size - r); +} + +const char *perf_evsel__hw_cache[PERF_COUNT_HW_CACHE_MAX] + [PERF_EVSEL__MAX_ALIASES] = { + { "L1-dcache", "l1-d", "l1d", "L1-data", }, + { "L1-icache", "l1-i", "l1i", "L1-instruction", }, + { "LLC", "L2", }, + { "dTLB", "d-tlb", "Data-TLB", }, + { "iTLB", "i-tlb", "Instruction-TLB", }, + { "branch", "branches", "bpu", "btb", "bpc", }, + { "node", }, +}; + +const char *perf_evsel__hw_cache_op[PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_EVSEL__MAX_ALIASES] = { + { "load", "loads", "read", }, + { "store", "stores", "write", }, + { "prefetch", "prefetches", "speculative-read", "speculative-load", }, +}; + +const char *perf_evsel__hw_cache_result[PERF_COUNT_HW_CACHE_RESULT_MAX] + [PERF_EVSEL__MAX_ALIASES] = { + { "refs", "Reference", "ops", "access", }, + { "misses", "miss", }, +}; + +#define C(x) PERF_COUNT_HW_CACHE_##x +#define CACHE_READ (1 << C(OP_READ)) +#define CACHE_WRITE (1 << C(OP_WRITE)) +#define CACHE_PREFETCH (1 << C(OP_PREFETCH)) +#define COP(x) (1 << x) + +/* + * cache operartion stat + * L1I : Read and prefetch only + * ITLB and BPU : Read-only + */ +static unsigned long perf_evsel__hw_cache_stat[C(MAX)] = { + [C(L1D)] = (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH), + [C(L1I)] = (CACHE_READ | CACHE_PREFETCH), + [C(LL)] = (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH), + [C(DTLB)] = (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH), + [C(ITLB)] = (CACHE_READ), + [C(BPU)] = (CACHE_READ), + [C(NODE)] = (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH), +}; + +bool perf_evsel__is_cache_op_valid(u8 type, u8 op) +{ + if (perf_evsel__hw_cache_stat[type] & COP(op)) + return true; /* valid */ + else + return false; /* invalid */ +} + +int __perf_evsel__hw_cache_type_op_res_name(u8 type, u8 op, u8 result, + char *bf, size_t size) +{ + if (result) { + return scnprintf(bf, size, "%s-%s-%s", perf_evsel__hw_cache[type][0], + perf_evsel__hw_cache_op[op][0], + perf_evsel__hw_cache_result[result][0]); + } + + return scnprintf(bf, size, "%s-%s", perf_evsel__hw_cache[type][0], + perf_evsel__hw_cache_op[op][1]); +} + +static int __perf_evsel__hw_cache_name(u64 config, char *bf, size_t size) +{ + u8 op, result, type = (config >> 0) & 0xff; + const char *err = "unknown-ext-hardware-cache-type"; + + if (type > PERF_COUNT_HW_CACHE_MAX) + goto out_err; + + op = (config >> 8) & 0xff; + err = "unknown-ext-hardware-cache-op"; + if (op > PERF_COUNT_HW_CACHE_OP_MAX) + goto out_err; + + result = (config >> 16) & 0xff; + err = "unknown-ext-hardware-cache-result"; + if (result > PERF_COUNT_HW_CACHE_RESULT_MAX) + goto out_err; + + err = "invalid-cache"; + if (!perf_evsel__is_cache_op_valid(type, op)) + goto out_err; + + return __perf_evsel__hw_cache_type_op_res_name(type, op, result, bf, size); +out_err: + return scnprintf(bf, size, "%s", err); +} + +static int perf_evsel__hw_cache_name(struct perf_evsel *evsel, char *bf, size_t size) +{ + int ret = __perf_evsel__hw_cache_name(evsel->attr.config, bf, size); + return ret + perf_evsel__add_modifiers(evsel, bf + ret, size - ret); +} + +static int perf_evsel__raw_name(struct perf_evsel *evsel, char *bf, size_t size) +{ + int ret = scnprintf(bf, size, "raw 0x%" PRIx64, evsel->attr.config); + return ret + perf_evsel__add_modifiers(evsel, bf + ret, size - ret); +} + +const char *perf_evsel__name(struct perf_evsel *evsel) +{ + char bf[128]; + + if (evsel->name) + return evsel->name; + + switch (evsel->attr.type) { + case PERF_TYPE_RAW: + perf_evsel__raw_name(evsel, bf, sizeof(bf)); + break; + + case PERF_TYPE_HARDWARE: + perf_evsel__hw_name(evsel, bf, sizeof(bf)); + break; + + case PERF_TYPE_HW_CACHE: + perf_evsel__hw_cache_name(evsel, bf, sizeof(bf)); + break; + + case PERF_TYPE_SOFTWARE: + perf_evsel__sw_name(evsel, bf, sizeof(bf)); + break; + + case PERF_TYPE_TRACEPOINT: + scnprintf(bf, sizeof(bf), "%s", "unknown tracepoint"); + break; + + case PERF_TYPE_BREAKPOINT: + perf_evsel__bp_name(evsel, bf, sizeof(bf)); + break; + + default: + scnprintf(bf, sizeof(bf), "unknown attr type: %d", + evsel->attr.type); + break; + } + + evsel->name = strdup(bf); + + return evsel->name ?: "unknown"; +} + +const char *perf_evsel__group_name(struct perf_evsel *evsel) +{ + return evsel->group_name ?: "anon group"; +} + +int perf_evsel__group_desc(struct perf_evsel *evsel, char *buf, size_t size) +{ + int ret; + struct perf_evsel *pos; + const char *group_name = perf_evsel__group_name(evsel); + + ret = scnprintf(buf, size, "%s", group_name); + + ret += scnprintf(buf + ret, size - ret, " { %s", + perf_evsel__name(evsel)); + + for_each_group_member(pos, evsel) + ret += scnprintf(buf + ret, size - ret, ", %s", + perf_evsel__name(pos)); + + ret += scnprintf(buf + ret, size - ret, " }"); + + return ret; +} + +/* + * The enable_on_exec/disabled value strategy: + * + * 1) For any type of traced program: + * - all independent events and group leaders are disabled + * - all group members are enabled + * + * Group members are ruled by group leaders. They need to + * be enabled, because the group scheduling relies on that. + * + * 2) For traced programs executed by perf: + * - all independent events and group leaders have + * enable_on_exec set + * - we don't specifically enable or disable any event during + * the record command + * + * Independent events and group leaders are initially disabled + * and get enabled by exec. Group members are ruled by group + * leaders as stated in 1). + * + * 3) For traced programs attached by perf (pid/tid): + * - we specifically enable or disable all events during + * the record command + * + * When attaching events to already running traced we + * enable/disable events specifically, as there's no + * initial traced exec call. + */ +void perf_evsel__config(struct perf_evsel *evsel, + struct perf_record_opts *opts) +{ + struct perf_event_attr *attr = &evsel->attr; + int track = !evsel->idx; /* only the first counter needs these */ + + attr->sample_id_all = perf_missing_features.sample_id_all ? 0 : 1; + attr->inherit = !opts->no_inherit; + + perf_evsel__set_sample_bit(evsel, IP); + perf_evsel__set_sample_bit(evsel, TID); + + /* + * We default some events to a 1 default interval. But keep + * it a weak assumption overridable by the user. + */ + if (!attr->sample_period || (opts->user_freq != UINT_MAX && + opts->user_interval != ULLONG_MAX)) { + if (opts->freq) { + perf_evsel__set_sample_bit(evsel, PERIOD); + attr->freq = 1; + attr->sample_freq = opts->freq; + } else { + attr->sample_period = opts->default_interval; + } + } + + if (opts->no_samples) + attr->sample_freq = 0; + + if (opts->inherit_stat) + attr->inherit_stat = 1; + + if (opts->sample_address) { + perf_evsel__set_sample_bit(evsel, ADDR); + attr->mmap_data = track; + } + + if (opts->call_graph) { + perf_evsel__set_sample_bit(evsel, CALLCHAIN); + + if (opts->call_graph == CALLCHAIN_DWARF) { + perf_evsel__set_sample_bit(evsel, REGS_USER); + perf_evsel__set_sample_bit(evsel, STACK_USER); + attr->sample_regs_user = PERF_REGS_MASK; + attr->sample_stack_user = opts->stack_dump_size; + attr->exclude_callchain_user = 1; + } + } + + if (perf_target__has_cpu(&opts->target)) + perf_evsel__set_sample_bit(evsel, CPU); + + if (opts->period) + perf_evsel__set_sample_bit(evsel, PERIOD); + + if (!perf_missing_features.sample_id_all && + (opts->sample_time || !opts->no_inherit || + perf_target__has_cpu(&opts->target))) + perf_evsel__set_sample_bit(evsel, TIME); + + if (opts->raw_samples) { + perf_evsel__set_sample_bit(evsel, TIME); + perf_evsel__set_sample_bit(evsel, RAW); + perf_evsel__set_sample_bit(evsel, CPU); + } + + if (opts->no_delay) { + attr->watermark = 0; + attr->wakeup_events = 1; + } + if (opts->branch_stack) { + perf_evsel__set_sample_bit(evsel, BRANCH_STACK); + attr->branch_sample_type = opts->branch_stack; + } + + attr->mmap = track; + attr->comm = track; + + /* + * XXX see the function comment above + * + * Disabling only independent events or group leaders, + * keeping group members enabled. + */ + if (perf_evsel__is_group_leader(evsel)) + attr->disabled = 1; + + /* + * Setting enable_on_exec for independent events and + * group leaders for traced executed by perf. + */ + if (perf_target__none(&opts->target) && perf_evsel__is_group_leader(evsel)) + attr->enable_on_exec = 1; +} + +int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads) +{ + int cpu, thread; + evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int)); + + if (evsel->fd) { + for (cpu = 0; cpu < ncpus; cpu++) { + for (thread = 0; thread < nthreads; thread++) { + FD(evsel, cpu, thread) = -1; + } + } + } + + return evsel->fd != NULL ? 0 : -ENOMEM; +} + +int perf_evsel__set_filter(struct perf_evsel *evsel, int ncpus, int nthreads, + const char *filter) +{ + int cpu, thread; + + for (cpu = 0; cpu < ncpus; cpu++) { + for (thread = 0; thread < nthreads; thread++) { + int fd = FD(evsel, cpu, thread), + err = ioctl(fd, PERF_EVENT_IOC_SET_FILTER, filter); + + if (err) + return err; + } + } + + return 0; +} + +int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads) +{ + evsel->sample_id = xyarray__new(ncpus, nthreads, sizeof(struct perf_sample_id)); + if (evsel->sample_id == NULL) + return -ENOMEM; + + evsel->id = zalloc(ncpus * nthreads * sizeof(u64)); + if (evsel->id == NULL) { + xyarray__delete(evsel->sample_id); + evsel->sample_id = NULL; + return -ENOMEM; + } + + return 0; +} + +int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus) +{ + evsel->counts = zalloc((sizeof(*evsel->counts) + + (ncpus * sizeof(struct perf_counts_values)))); + return evsel->counts != NULL ? 0 : -ENOMEM; +} + +void perf_evsel__free_fd(struct perf_evsel *evsel) +{ + xyarray__delete(evsel->fd); + evsel->fd = NULL; +} + +void perf_evsel__free_id(struct perf_evsel *evsel) +{ + xyarray__delete(evsel->sample_id); + evsel->sample_id = NULL; + free(evsel->id); + evsel->id = NULL; +} + +void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads) +{ + int cpu, thread; + + for (cpu = 0; cpu < ncpus; cpu++) + for (thread = 0; thread < nthreads; ++thread) { + close(FD(evsel, cpu, thread)); + FD(evsel, cpu, thread) = -1; + } +} + +void perf_evsel__free_counts(struct perf_evsel *evsel) +{ + free(evsel->counts); +} + +void perf_evsel__exit(struct perf_evsel *evsel) +{ + assert(list_empty(&evsel->node)); + xyarray__delete(evsel->fd); + xyarray__delete(evsel->sample_id); + free(evsel->id); +} + +void perf_evsel__delete(struct perf_evsel *evsel) +{ + perf_evsel__exit(evsel); + close_cgroup(evsel->cgrp); + free(evsel->group_name); + if (evsel->tp_format) + pevent_free_format(evsel->tp_format); + free(evsel->name); + free(evsel); +} + +static inline void compute_deltas(struct perf_evsel *evsel, + int cpu, + struct perf_counts_values *count) +{ + struct perf_counts_values tmp; + + if (!evsel->prev_raw_counts) + return; + + if (cpu == -1) { + tmp = evsel->prev_raw_counts->aggr; + evsel->prev_raw_counts->aggr = *count; + } else { + tmp = evsel->prev_raw_counts->cpu[cpu]; + evsel->prev_raw_counts->cpu[cpu] = *count; + } + + count->val = count->val - tmp.val; + count->ena = count->ena - tmp.ena; + count->run = count->run - tmp.run; +} + +int __perf_evsel__read_on_cpu(struct perf_evsel *evsel, + int cpu, int thread, bool scale) +{ + struct perf_counts_values count; + size_t nv = scale ? 3 : 1; + + if (FD(evsel, cpu, thread) < 0) + return -EINVAL; + + if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0) + return -ENOMEM; + + if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0) + return -errno; + + compute_deltas(evsel, cpu, &count); + + if (scale) { + if (count.run == 0) + count.val = 0; + else if (count.run < count.ena) + count.val = (u64)((double)count.val * count.ena / count.run + 0.5); + } else + count.ena = count.run = 0; + + evsel->counts->cpu[cpu] = count; + return 0; +} + +int __perf_evsel__read(struct perf_evsel *evsel, + int ncpus, int nthreads, bool scale) +{ + size_t nv = scale ? 3 : 1; + int cpu, thread; + struct perf_counts_values *aggr = &evsel->counts->aggr, count; + + aggr->val = aggr->ena = aggr->run = 0; + + for (cpu = 0; cpu < ncpus; cpu++) { + for (thread = 0; thread < nthreads; thread++) { + if (FD(evsel, cpu, thread) < 0) + continue; + + if (readn(FD(evsel, cpu, thread), + &count, nv * sizeof(u64)) < 0) + return -errno; + + aggr->val += count.val; + if (scale) { + aggr->ena += count.ena; + aggr->run += count.run; + } + } + } + + compute_deltas(evsel, -1, aggr); + + evsel->counts->scaled = 0; + if (scale) { + if (aggr->run == 0) { + evsel->counts->scaled = -1; + aggr->val = 0; + return 0; + } + + if (aggr->run < aggr->ena) { + evsel->counts->scaled = 1; + aggr->val = (u64)((double)aggr->val * aggr->ena / aggr->run + 0.5); + } + } else + aggr->ena = aggr->run = 0; + + return 0; +} + +static int get_group_fd(struct perf_evsel *evsel, int cpu, int thread) +{ + struct perf_evsel *leader = evsel->leader; + int fd; + + if (perf_evsel__is_group_leader(evsel)) + return -1; + + /* + * Leader must be already processed/open, + * if not it's a bug. + */ + BUG_ON(!leader->fd); + + fd = FD(leader, cpu, thread); + BUG_ON(fd == -1); + + return fd; +} + +static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus, + struct thread_map *threads) +{ + int cpu, thread; + unsigned long flags = 0; + int pid = -1, err; + + if (evsel->fd == NULL && + perf_evsel__alloc_fd(evsel, cpus->nr, threads->nr) < 0) + return -ENOMEM; + + if (evsel->cgrp) { + flags = PERF_FLAG_PID_CGROUP; + pid = evsel->cgrp->fd; + } + +fallback_missing_features: + if (perf_missing_features.exclude_guest) + evsel->attr.exclude_guest = evsel->attr.exclude_host = 0; +retry_sample_id: + if (perf_missing_features.sample_id_all) + evsel->attr.sample_id_all = 0; + + for (cpu = 0; cpu < cpus->nr; cpu++) { + + for (thread = 0; thread < threads->nr; thread++) { + int group_fd; + + if (!evsel->cgrp) + pid = threads->map[thread]; + + group_fd = get_group_fd(evsel, cpu, thread); + + FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr, + pid, + cpus->map[cpu], + group_fd, flags); + if (FD(evsel, cpu, thread) < 0) { + err = -errno; + goto try_fallback; + } + } + } + + return 0; + +try_fallback: + if (err != -EINVAL || cpu > 0 || thread > 0) + goto out_close; + + if (!perf_missing_features.exclude_guest && + (evsel->attr.exclude_guest || evsel->attr.exclude_host)) { + perf_missing_features.exclude_guest = true; + goto fallback_missing_features; + } else if (!perf_missing_features.sample_id_all) { + perf_missing_features.sample_id_all = true; + goto retry_sample_id; + } + +out_close: + do { + while (--thread >= 0) { + close(FD(evsel, cpu, thread)); + FD(evsel, cpu, thread) = -1; + } + thread = threads->nr; + } while (--cpu >= 0); + return err; +} + +void perf_evsel__close(struct perf_evsel *evsel, int ncpus, int nthreads) +{ + if (evsel->fd == NULL) + return; + + perf_evsel__close_fd(evsel, ncpus, nthreads); + perf_evsel__free_fd(evsel); + evsel->fd = NULL; +} + +static struct { + struct cpu_map map; + int cpus[1]; +} empty_cpu_map = { + .map.nr = 1, + .cpus = { -1, }, +}; + +static struct { + struct thread_map map; + int threads[1]; +} empty_thread_map = { + .map.nr = 1, + .threads = { -1, }, +}; + +int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus, + struct thread_map *threads) +{ + if (cpus == NULL) { + /* Work around old compiler warnings about strict aliasing */ + cpus = &empty_cpu_map.map; + } + + if (threads == NULL) + threads = &empty_thread_map.map; + + return __perf_evsel__open(evsel, cpus, threads); +} + +int perf_evsel__open_per_cpu(struct perf_evsel *evsel, + struct cpu_map *cpus) +{ + return __perf_evsel__open(evsel, cpus, &empty_thread_map.map); +} + +int perf_evsel__open_per_thread(struct perf_evsel *evsel, + struct thread_map *threads) +{ + return __perf_evsel__open(evsel, &empty_cpu_map.map, threads); +} + +static int perf_evsel__parse_id_sample(const struct perf_evsel *evsel, + const union perf_event *event, + struct perf_sample *sample) +{ + u64 type = evsel->attr.sample_type; + const u64 *array = event->sample.array; + bool swapped = evsel->needs_swap; + union u64_swap u; + + array += ((event->header.size - + sizeof(event->header)) / sizeof(u64)) - 1; + + if (type & PERF_SAMPLE_CPU) { + u.val64 = *array; + if (swapped) { + /* undo swap of u64, then swap on individual u32s */ + u.val64 = bswap_64(u.val64); + u.val32[0] = bswap_32(u.val32[0]); + } + + sample->cpu = u.val32[0]; + array--; + } + + if (type & PERF_SAMPLE_STREAM_ID) { + sample->stream_id = *array; + array--; + } + + if (type & PERF_SAMPLE_ID) { + sample->id = *array; + array--; + } + + if (type & PERF_SAMPLE_TIME) { + sample->time = *array; + array--; + } + + if (type & PERF_SAMPLE_TID) { + u.val64 = *array; + if (swapped) { + /* undo swap of u64, then swap on individual u32s */ + u.val64 = bswap_64(u.val64); + u.val32[0] = bswap_32(u.val32[0]); + u.val32[1] = bswap_32(u.val32[1]); + } + + sample->pid = u.val32[0]; + sample->tid = u.val32[1]; + } + + return 0; +} + +static bool sample_overlap(const union perf_event *event, + const void *offset, u64 size) +{ + const void *base = event; + + if (offset + size > base + event->header.size) + return true; + + return false; +} + +int perf_evsel__parse_sample(struct perf_evsel *evsel, union perf_event *event, + struct perf_sample *data) +{ + u64 type = evsel->attr.sample_type; + u64 regs_user = evsel->attr.sample_regs_user; + bool swapped = evsel->needs_swap; + const u64 *array; + + /* + * used for cross-endian analysis. See git commit 65014ab3 + * for why this goofiness is needed. + */ + union u64_swap u; + + memset(data, 0, sizeof(*data)); + data->cpu = data->pid = data->tid = -1; + data->stream_id = data->id = data->time = -1ULL; + data->period = 1; + + if (event->header.type != PERF_RECORD_SAMPLE) { + if (!evsel->attr.sample_id_all) + return 0; + return perf_evsel__parse_id_sample(evsel, event, data); + } + + array = event->sample.array; + + if (evsel->sample_size + sizeof(event->header) > event->header.size) + return -EFAULT; + + if (type & PERF_SAMPLE_IP) { + data->ip = event->ip.ip; + array++; + } + + if (type & PERF_SAMPLE_TID) { + u.val64 = *array; + if (swapped) { + /* undo swap of u64, then swap on individual u32s */ + u.val64 = bswap_64(u.val64); + u.val32[0] = bswap_32(u.val32[0]); + u.val32[1] = bswap_32(u.val32[1]); + } + + data->pid = u.val32[0]; + data->tid = u.val32[1]; + array++; + } + + if (type & PERF_SAMPLE_TIME) { + data->time = *array; + array++; + } + + data->addr = 0; + if (type & PERF_SAMPLE_ADDR) { + data->addr = *array; + array++; + } + + data->id = -1ULL; + if (type & PERF_SAMPLE_ID) { + data->id = *array; + array++; + } + + if (type & PERF_SAMPLE_STREAM_ID) { + data->stream_id = *array; + array++; + } + + if (type & PERF_SAMPLE_CPU) { + + u.val64 = *array; + if (swapped) { + /* undo swap of u64, then swap on individual u32s */ + u.val64 = bswap_64(u.val64); + u.val32[0] = bswap_32(u.val32[0]); + } + + data->cpu = u.val32[0]; + array++; + } + + if (type & PERF_SAMPLE_PERIOD) { + data->period = *array; + array++; + } + + if (type & PERF_SAMPLE_READ) { + fprintf(stderr, "PERF_SAMPLE_READ is unsupported for now\n"); + return -1; + } + + if (type & PERF_SAMPLE_CALLCHAIN) { + if (sample_overlap(event, array, sizeof(data->callchain->nr))) + return -EFAULT; + + data->callchain = (struct ip_callchain *)array; + + if (sample_overlap(event, array, data->callchain->nr)) + return -EFAULT; + + array += 1 + data->callchain->nr; + } + + if (type & PERF_SAMPLE_RAW) { + const u64 *pdata; + + u.val64 = *array; + if (WARN_ONCE(swapped, + "Endianness of raw data not corrected!\n")) { + /* undo swap of u64, then swap on individual u32s */ + u.val64 = bswap_64(u.val64); + u.val32[0] = bswap_32(u.val32[0]); + u.val32[1] = bswap_32(u.val32[1]); + } + + if (sample_overlap(event, array, sizeof(u32))) + return -EFAULT; + + data->raw_size = u.val32[0]; + pdata = (void *) array + sizeof(u32); + + if (sample_overlap(event, pdata, data->raw_size)) + return -EFAULT; + + data->raw_data = (void *) pdata; + + array = (void *)array + data->raw_size + sizeof(u32); + } + + if (type & PERF_SAMPLE_BRANCH_STACK) { + u64 sz; + + data->branch_stack = (struct branch_stack *)array; + array++; /* nr */ + + sz = data->branch_stack->nr * sizeof(struct branch_entry); + sz /= sizeof(u64); + array += sz; + } + + if (type & PERF_SAMPLE_REGS_USER) { + /* First u64 tells us if we have any regs in sample. */ + u64 avail = *array++; + + if (avail) { + data->user_regs.regs = (u64 *)array; + array += hweight_long(regs_user); + } + } + + if (type & PERF_SAMPLE_STACK_USER) { + u64 size = *array++; + + data->user_stack.offset = ((char *)(array - 1) + - (char *) event); + + if (!size) { + data->user_stack.size = 0; + } else { + data->user_stack.data = (char *)array; + array += size / sizeof(*array); + data->user_stack.size = *array; + } + } + + return 0; +} + +int perf_event__synthesize_sample(union perf_event *event, u64 type, + const struct perf_sample *sample, + bool swapped) +{ + u64 *array; + + /* + * used for cross-endian analysis. See git commit 65014ab3 + * for why this goofiness is needed. + */ + union u64_swap u; + + array = event->sample.array; + + if (type & PERF_SAMPLE_IP) { + event->ip.ip = sample->ip; + array++; + } + + if (type & PERF_SAMPLE_TID) { + u.val32[0] = sample->pid; + u.val32[1] = sample->tid; + if (swapped) { + /* + * Inverse of what is done in perf_evsel__parse_sample + */ + u.val32[0] = bswap_32(u.val32[0]); + u.val32[1] = bswap_32(u.val32[1]); + u.val64 = bswap_64(u.val64); + } + + *array = u.val64; + array++; + } + + if (type & PERF_SAMPLE_TIME) { + *array = sample->time; + array++; + } + + if (type & PERF_SAMPLE_ADDR) { + *array = sample->addr; + array++; + } + + if (type & PERF_SAMPLE_ID) { + *array = sample->id; + array++; + } + + if (type & PERF_SAMPLE_STREAM_ID) { + *array = sample->stream_id; + array++; + } + + if (type & PERF_SAMPLE_CPU) { + u.val32[0] = sample->cpu; + if (swapped) { + /* + * Inverse of what is done in perf_evsel__parse_sample + */ + u.val32[0] = bswap_32(u.val32[0]); + u.val64 = bswap_64(u.val64); + } + *array = u.val64; + array++; + } + + if (type & PERF_SAMPLE_PERIOD) { + *array = sample->period; + array++; + } + + return 0; +} + +struct format_field *perf_evsel__field(struct perf_evsel *evsel, const char *name) +{ + return pevent_find_field(evsel->tp_format, name); +} + +void *perf_evsel__rawptr(struct perf_evsel *evsel, struct perf_sample *sample, + const char *name) +{ + struct format_field *field = perf_evsel__field(evsel, name); + int offset; + + if (!field) + return NULL; + + offset = field->offset; + + if (field->flags & FIELD_IS_DYNAMIC) { + offset = *(int *)(sample->raw_data + field->offset); + offset &= 0xffff; + } + + return sample->raw_data + offset; +} + +u64 perf_evsel__intval(struct perf_evsel *evsel, struct perf_sample *sample, + const char *name) +{ + struct format_field *field = perf_evsel__field(evsel, name); + void *ptr; + u64 value; + + if (!field) + return 0; + + ptr = sample->raw_data + field->offset; + + switch (field->size) { + case 1: + return *(u8 *)ptr; + case 2: + value = *(u16 *)ptr; + break; + case 4: + value = *(u32 *)ptr; + break; + case 8: + value = *(u64 *)ptr; + break; + default: + return 0; + } + + if (!evsel->needs_swap) + return value; + + switch (field->size) { + case 2: + return bswap_16(value); + case 4: + return bswap_32(value); + case 8: + return bswap_64(value); + default: + return 0; + } + + return 0; +} + +static int comma_fprintf(FILE *fp, bool *first, const char *fmt, ...) +{ + va_list args; + int ret = 0; + + if (!*first) { + ret += fprintf(fp, ","); + } else { + ret += fprintf(fp, ":"); + *first = false; + } + + va_start(args, fmt); + ret += vfprintf(fp, fmt, args); + va_end(args); + return ret; +} + +static int __if_fprintf(FILE *fp, bool *first, const char *field, u64 value) +{ + if (value == 0) + return 0; + + return comma_fprintf(fp, first, " %s: %" PRIu64, field, value); +} + +#define if_print(field) printed += __if_fprintf(fp, &first, #field, evsel->attr.field) + +struct bit_names { + int bit; + const char *name; +}; + +static int bits__fprintf(FILE *fp, const char *field, u64 value, + struct bit_names *bits, bool *first) +{ + int i = 0, printed = comma_fprintf(fp, first, " %s: ", field); + bool first_bit = true; + + do { + if (value & bits[i].bit) { + printed += fprintf(fp, "%s%s", first_bit ? "" : "|", bits[i].name); + first_bit = false; + } + } while (bits[++i].name != NULL); + + return printed; +} + +static int sample_type__fprintf(FILE *fp, bool *first, u64 value) +{ +#define bit_name(n) { PERF_SAMPLE_##n, #n } + struct bit_names bits[] = { + bit_name(IP), bit_name(TID), bit_name(TIME), bit_name(ADDR), + bit_name(READ), bit_name(CALLCHAIN), bit_name(ID), bit_name(CPU), + bit_name(PERIOD), bit_name(STREAM_ID), bit_name(RAW), + bit_name(BRANCH_STACK), bit_name(REGS_USER), bit_name(STACK_USER), + { .name = NULL, } + }; +#undef bit_name + return bits__fprintf(fp, "sample_type", value, bits, first); +} + +static int read_format__fprintf(FILE *fp, bool *first, u64 value) +{ +#define bit_name(n) { PERF_FORMAT_##n, #n } + struct bit_names bits[] = { + bit_name(TOTAL_TIME_ENABLED), bit_name(TOTAL_TIME_RUNNING), + bit_name(ID), bit_name(GROUP), + { .name = NULL, } + }; +#undef bit_name + return bits__fprintf(fp, "read_format", value, bits, first); +} + +int perf_evsel__fprintf(struct perf_evsel *evsel, + struct perf_attr_details *details, FILE *fp) +{ + bool first = true; + int printed = 0; + + if (details->event_group) { + struct perf_evsel *pos; + + if (!perf_evsel__is_group_leader(evsel)) + return 0; + + if (evsel->nr_members > 1) + printed += fprintf(fp, "%s{", evsel->group_name ?: ""); + + printed += fprintf(fp, "%s", perf_evsel__name(evsel)); + for_each_group_member(pos, evsel) + printed += fprintf(fp, ",%s", perf_evsel__name(pos)); + + if (evsel->nr_members > 1) + printed += fprintf(fp, "}"); + goto out; + } + + printed += fprintf(fp, "%s", perf_evsel__name(evsel)); + + if (details->verbose || details->freq) { + printed += comma_fprintf(fp, &first, " sample_freq=%" PRIu64, + (u64)evsel->attr.sample_freq); + } + + if (details->verbose) { + if_print(type); + if_print(config); + if_print(config1); + if_print(config2); + if_print(size); + printed += sample_type__fprintf(fp, &first, evsel->attr.sample_type); + if (evsel->attr.read_format) + printed += read_format__fprintf(fp, &first, evsel->attr.read_format); + if_print(disabled); + if_print(inherit); + if_print(pinned); + if_print(exclusive); + if_print(exclude_user); + if_print(exclude_kernel); + if_print(exclude_hv); + if_print(exclude_idle); + if_print(mmap); + if_print(comm); + if_print(freq); + if_print(inherit_stat); + if_print(enable_on_exec); + if_print(task); + if_print(watermark); + if_print(precise_ip); + if_print(mmap_data); + if_print(sample_id_all); + if_print(exclude_host); + if_print(exclude_guest); + if_print(__reserved_1); + if_print(wakeup_events); + if_print(bp_type); + if_print(branch_sample_type); + } +out: + fputc('\n', fp); + return ++printed; +} + +bool perf_evsel__fallback(struct perf_evsel *evsel, int err, + char *msg, size_t msgsize) +{ + if ((err == ENOENT || err == ENXIO) && + evsel->attr.type == PERF_TYPE_HARDWARE && + evsel->attr.config == PERF_COUNT_HW_CPU_CYCLES) { + /* + * If it's cycles then fall back to hrtimer based + * cpu-clock-tick sw counter, which is always available even if + * no PMU support. + * + * PPC returns ENXIO until 2.6.37 (behavior changed with commit + * b0a873e). + */ + scnprintf(msg, msgsize, "%s", +"The cycles event is not supported, trying to fall back to cpu-clock-ticks"); + + evsel->attr.type = PERF_TYPE_SOFTWARE; + evsel->attr.config = PERF_COUNT_SW_CPU_CLOCK; + + free(evsel->name); + evsel->name = NULL; + return true; + } + + return false; +} + +int perf_evsel__open_strerror(struct perf_evsel *evsel, + struct perf_target *target, + int err, char *msg, size_t size) +{ + switch (err) { + case EPERM: + case EACCES: + return scnprintf(msg, size, "%s", + "You may not have permission to collect %sstats.\n" + "Consider tweaking /proc/sys/kernel/perf_event_paranoid:\n" + " -1 - Not paranoid at all\n" + " 0 - Disallow raw tracepoint access for unpriv\n" + " 1 - Disallow cpu events for unpriv\n" + " 2 - Disallow kernel profiling for unpriv", + target->system_wide ? "system-wide " : ""); + case ENOENT: + return scnprintf(msg, size, "The %s event is not supported.", + perf_evsel__name(evsel)); + case EMFILE: + return scnprintf(msg, size, "%s", + "Too many events are opened.\n" + "Try again after reducing the number of events."); + case ENODEV: + if (target->cpu_list) + return scnprintf(msg, size, "%s", + "No such device - did you specify an out-of-range profile CPU?\n"); + break; + case EOPNOTSUPP: + if (evsel->attr.precise_ip) + return scnprintf(msg, size, "%s", + "\'precise\' request may not be supported. Try removing 'p' modifier."); +#if defined(__i386__) || defined(__x86_64__) + if (evsel->attr.type == PERF_TYPE_HARDWARE) + return scnprintf(msg, size, "%s", + "No hardware sampling interrupt available.\n" + "No APIC? If so then you can boot the kernel with the \"lapic\" boot parameter to force-enable it."); +#endif + break; + default: + break; + } + + return scnprintf(msg, size, + "The sys_perf_event_open() syscall returned with %d (%s) for event (%s). \n" + "/bin/dmesg may provide additional information.\n" + "No CONFIG_PERF_EVENTS=y kernel support configured?\n", + err, strerror(err), perf_evsel__name(evsel)); +} diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h new file mode 100644 index 00000000..52021c30 --- /dev/null +++ b/tools/perf/util/evsel.h @@ -0,0 +1,279 @@ +#ifndef __PERF_EVSEL_H +#define __PERF_EVSEL_H 1 + +#include <linux/list.h> +#include <stdbool.h> +#include <stddef.h> +#include <linux/perf_event.h> +#include "types.h" +#include "xyarray.h" +#include "cgroup.h" +#include "hist.h" + +struct perf_counts_values { + union { + struct { + u64 val; + u64 ena; + u64 run; + }; + u64 values[3]; + }; +}; + +struct perf_counts { + s8 scaled; + struct perf_counts_values aggr; + struct perf_counts_values cpu[]; +}; + +struct perf_evsel; + +/* + * Per fd, to map back from PERF_SAMPLE_ID to evsel, only used when there are + * more than one entry in the evlist. + */ +struct perf_sample_id { + struct hlist_node node; + u64 id; + struct perf_evsel *evsel; +}; + +/** struct perf_evsel - event selector + * + * @name - Can be set to retain the original event name passed by the user, + * so that when showing results in tools such as 'perf stat', we + * show the name used, not some alias. + */ +struct perf_evsel { + struct list_head node; + struct perf_event_attr attr; + char *filter; + struct xyarray *fd; + struct xyarray *sample_id; + u64 *id; + struct perf_counts *counts; + struct perf_counts *prev_raw_counts; + int idx; + u32 ids; + struct hists hists; + char *name; + struct event_format *tp_format; + union { + void *priv; + off_t id_offset; + }; + struct cgroup_sel *cgrp; + struct { + void *func; + void *data; + } handler; + struct cpu_map *cpus; + unsigned int sample_size; + bool supported; + bool needs_swap; + /* parse modifier helper */ + int exclude_GH; + int nr_members; + struct perf_evsel *leader; + char *group_name; +}; + +#define hists_to_evsel(h) container_of(h, struct perf_evsel, hists) + +struct cpu_map; +struct thread_map; +struct perf_evlist; +struct perf_record_opts; + +struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx); +struct perf_evsel *perf_evsel__newtp(const char *sys, const char *name, int idx); + +struct event_format *event_format__new(const char *sys, const char *name); + +void perf_evsel__init(struct perf_evsel *evsel, + struct perf_event_attr *attr, int idx); +void perf_evsel__exit(struct perf_evsel *evsel); +void perf_evsel__delete(struct perf_evsel *evsel); + +void perf_evsel__config(struct perf_evsel *evsel, + struct perf_record_opts *opts); + +bool perf_evsel__is_cache_op_valid(u8 type, u8 op); + +#define PERF_EVSEL__MAX_ALIASES 8 + +extern const char *perf_evsel__hw_cache[PERF_COUNT_HW_CACHE_MAX] + [PERF_EVSEL__MAX_ALIASES]; +extern const char *perf_evsel__hw_cache_op[PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_EVSEL__MAX_ALIASES]; +extern const char *perf_evsel__hw_cache_result[PERF_COUNT_HW_CACHE_RESULT_MAX] + [PERF_EVSEL__MAX_ALIASES]; +extern const char *perf_evsel__hw_names[PERF_COUNT_HW_MAX]; +extern const char *perf_evsel__sw_names[PERF_COUNT_SW_MAX]; +int __perf_evsel__hw_cache_type_op_res_name(u8 type, u8 op, u8 result, + char *bf, size_t size); +const char *perf_evsel__name(struct perf_evsel *evsel); +const char *perf_evsel__group_name(struct perf_evsel *evsel); +int perf_evsel__group_desc(struct perf_evsel *evsel, char *buf, size_t size); + +int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads); +int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads); +int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus); +void perf_evsel__free_fd(struct perf_evsel *evsel); +void perf_evsel__free_id(struct perf_evsel *evsel); +void perf_evsel__free_counts(struct perf_evsel *evsel); +void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads); + +void __perf_evsel__set_sample_bit(struct perf_evsel *evsel, + enum perf_event_sample_format bit); +void __perf_evsel__reset_sample_bit(struct perf_evsel *evsel, + enum perf_event_sample_format bit); + +#define perf_evsel__set_sample_bit(evsel, bit) \ + __perf_evsel__set_sample_bit(evsel, PERF_SAMPLE_##bit) + +#define perf_evsel__reset_sample_bit(evsel, bit) \ + __perf_evsel__reset_sample_bit(evsel, PERF_SAMPLE_##bit) + +void perf_evsel__set_sample_id(struct perf_evsel *evsel); + +int perf_evsel__set_filter(struct perf_evsel *evsel, int ncpus, int nthreads, + const char *filter); + +int perf_evsel__open_per_cpu(struct perf_evsel *evsel, + struct cpu_map *cpus); +int perf_evsel__open_per_thread(struct perf_evsel *evsel, + struct thread_map *threads); +int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus, + struct thread_map *threads); +void perf_evsel__close(struct perf_evsel *evsel, int ncpus, int nthreads); + +struct perf_sample; + +void *perf_evsel__rawptr(struct perf_evsel *evsel, struct perf_sample *sample, + const char *name); +u64 perf_evsel__intval(struct perf_evsel *evsel, struct perf_sample *sample, + const char *name); + +static inline char *perf_evsel__strval(struct perf_evsel *evsel, + struct perf_sample *sample, + const char *name) +{ + return perf_evsel__rawptr(evsel, sample, name); +} + +struct format_field; + +struct format_field *perf_evsel__field(struct perf_evsel *evsel, const char *name); + +#define perf_evsel__match(evsel, t, c) \ + (evsel->attr.type == PERF_TYPE_##t && \ + evsel->attr.config == PERF_COUNT_##c) + +static inline bool perf_evsel__match2(struct perf_evsel *e1, + struct perf_evsel *e2) +{ + return (e1->attr.type == e2->attr.type) && + (e1->attr.config == e2->attr.config); +} + +int __perf_evsel__read_on_cpu(struct perf_evsel *evsel, + int cpu, int thread, bool scale); + +/** + * perf_evsel__read_on_cpu - Read out the results on a CPU and thread + * + * @evsel - event selector to read value + * @cpu - CPU of interest + * @thread - thread of interest + */ +static inline int perf_evsel__read_on_cpu(struct perf_evsel *evsel, + int cpu, int thread) +{ + return __perf_evsel__read_on_cpu(evsel, cpu, thread, false); +} + +/** + * perf_evsel__read_on_cpu_scaled - Read out the results on a CPU and thread, scaled + * + * @evsel - event selector to read value + * @cpu - CPU of interest + * @thread - thread of interest + */ +static inline int perf_evsel__read_on_cpu_scaled(struct perf_evsel *evsel, + int cpu, int thread) +{ + return __perf_evsel__read_on_cpu(evsel, cpu, thread, true); +} + +int __perf_evsel__read(struct perf_evsel *evsel, int ncpus, int nthreads, + bool scale); + +/** + * perf_evsel__read - Read the aggregate results on all CPUs + * + * @evsel - event selector to read value + * @ncpus - Number of cpus affected, from zero + * @nthreads - Number of threads affected, from zero + */ +static inline int perf_evsel__read(struct perf_evsel *evsel, + int ncpus, int nthreads) +{ + return __perf_evsel__read(evsel, ncpus, nthreads, false); +} + +/** + * perf_evsel__read_scaled - Read the aggregate results on all CPUs, scaled + * + * @evsel - event selector to read value + * @ncpus - Number of cpus affected, from zero + * @nthreads - Number of threads affected, from zero + */ +static inline int perf_evsel__read_scaled(struct perf_evsel *evsel, + int ncpus, int nthreads) +{ + return __perf_evsel__read(evsel, ncpus, nthreads, true); +} + +void hists__init(struct hists *hists); + +int perf_evsel__parse_sample(struct perf_evsel *evsel, union perf_event *event, + struct perf_sample *sample); + +static inline struct perf_evsel *perf_evsel__next(struct perf_evsel *evsel) +{ + return list_entry(evsel->node.next, struct perf_evsel, node); +} + +static inline bool perf_evsel__is_group_leader(const struct perf_evsel *evsel) +{ + return evsel->leader == evsel; +} + +struct perf_attr_details { + bool freq; + bool verbose; + bool event_group; +}; + +int perf_evsel__fprintf(struct perf_evsel *evsel, + struct perf_attr_details *details, FILE *fp); + +bool perf_evsel__fallback(struct perf_evsel *evsel, int err, + char *msg, size_t msgsize); +int perf_evsel__open_strerror(struct perf_evsel *evsel, + struct perf_target *target, + int err, char *msg, size_t size); + +static inline int perf_evsel__group_idx(struct perf_evsel *evsel) +{ + return evsel->idx - evsel->leader->idx; +} + +#define for_each_group_member(_evsel, _leader) \ +for ((_evsel) = list_entry((_leader)->node.next, struct perf_evsel, node); \ + (_evsel) && (_evsel)->leader == (_leader); \ + (_evsel) = list_entry((_evsel)->node.next, struct perf_evsel, node)) + +#endif /* __PERF_EVSEL_H */ diff --git a/tools/perf/util/exec_cmd.c b/tools/perf/util/exec_cmd.c new file mode 100644 index 00000000..7adf4ad1 --- /dev/null +++ b/tools/perf/util/exec_cmd.c @@ -0,0 +1,148 @@ +#include "cache.h" +#include "exec_cmd.h" +#include "quote.h" + +#include <string.h> + +#define MAX_ARGS 32 + +static const char *argv_exec_path; +static const char *argv0_path; + +const char *system_path(const char *path) +{ + static const char *prefix = PREFIX; + struct strbuf d = STRBUF_INIT; + + if (is_absolute_path(path)) + return path; + + 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 argv0_path ? slash + 1 : NULL; + } + + 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); +} + +static 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 00000000..bc4b9159 --- /dev/null +++ b/tools/perf/util/exec_cmd.h @@ -0,0 +1,12 @@ +#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 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 00000000..3ac38031 --- /dev/null +++ b/tools/perf/util/generate-cmdlist.sh @@ -0,0 +1,39 @@ +#!/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 "#ifdef LIBELF_SUPPORT" +sed -n -e 's/^perf-\([^ ]*\)[ ].* full.*/\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 "#endif /* LIBELF_SUPPORT */" +echo "};" diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c new file mode 100644 index 00000000..f4bfd79e --- /dev/null +++ b/tools/perf/util/header.c @@ -0,0 +1,3155 @@ +#define _FILE_OFFSET_BITS 64 + +#include "util.h" +#include <sys/types.h> +#include <byteswap.h> +#include <unistd.h> +#include <stdio.h> +#include <stdlib.h> +#include <linux/list.h> +#include <linux/kernel.h> +#include <linux/bitops.h> +#include <sys/utsname.h> + +#include "evlist.h" +#include "evsel.h" +#include "header.h" +#include "../perf.h" +#include "trace-event.h" +#include "session.h" +#include "symbol.h" +#include "debug.h" +#include "cpumap.h" +#include "pmu.h" +#include "vdso.h" +#include "strbuf.h" +#include "build-id.h" + +static bool no_buildid_cache = false; + +static int trace_event_count; +static struct perf_trace_event_type *trace_events; + +static u32 header_argc; +static const char **header_argv; + +int perf_header__push_event(u64 id, const char *name) +{ + struct perf_trace_event_type *nevents; + + if (strlen(name) > MAX_EVENT_NAME) + pr_warning("Event %s will be truncated\n", name); + + nevents = realloc(trace_events, (trace_event_count + 1) * sizeof(*trace_events)); + if (nevents == NULL) + return -ENOMEM; + trace_events = nevents; + + memset(&trace_events[trace_event_count], 0, sizeof(struct perf_trace_event_type)); + trace_events[trace_event_count].event_id = id; + strncpy(trace_events[trace_event_count].name, name, MAX_EVENT_NAME - 1); + trace_event_count++; + return 0; +} + +char *perf_header__find_event(u64 id) +{ + int i; + for (i = 0 ; i < trace_event_count; i++) { + if (trace_events[i].event_id == id) + return trace_events[i].name; + } + return NULL; +} + +/* + * magic2 = "PERFILE2" + * must be a numerical value to let the endianness + * determine the memory layout. That way we are able + * to detect endianness when reading the perf.data file + * back. + * + * we check for legacy (PERFFILE) format. + */ +static const char *__perf_magic1 = "PERFFILE"; +static const u64 __perf_magic2 = 0x32454c4946524550ULL; +static const u64 __perf_magic2_sw = 0x50455246494c4532ULL; + +#define PERF_MAGIC __perf_magic2 + +struct perf_file_attr { + struct perf_event_attr attr; + struct perf_file_section ids; +}; + +void perf_header__set_feat(struct perf_header *header, int feat) +{ + set_bit(feat, header->adds_features); +} + +void perf_header__clear_feat(struct perf_header *header, int feat) +{ + clear_bit(feat, header->adds_features); +} + +bool perf_header__has_feat(const struct perf_header *header, int feat) +{ + return test_bit(feat, header->adds_features); +} + +static int do_write(int fd, const void *buf, size_t size) +{ + while (size) { + int ret = write(fd, buf, size); + + if (ret < 0) + return -errno; + + size -= ret; + buf += ret; + } + + return 0; +} + +#define NAME_ALIGN 64 + +static int write_padded(int fd, const void *bf, size_t count, + size_t count_aligned) +{ + static const char zero_buf[NAME_ALIGN]; + int err = do_write(fd, bf, count); + + if (!err) + err = do_write(fd, zero_buf, count_aligned - count); + + return err; +} + +static int do_write_string(int fd, const char *str) +{ + u32 len, olen; + int ret; + + olen = strlen(str) + 1; + len = PERF_ALIGN(olen, NAME_ALIGN); + + /* write len, incl. \0 */ + ret = do_write(fd, &len, sizeof(len)); + if (ret < 0) + return ret; + + return write_padded(fd, str, olen, len); +} + +static char *do_read_string(int fd, struct perf_header *ph) +{ + ssize_t sz, ret; + u32 len; + char *buf; + + sz = readn(fd, &len, sizeof(len)); + if (sz < (ssize_t)sizeof(len)) + return NULL; + + if (ph->needs_swap) + len = bswap_32(len); + + buf = malloc(len); + if (!buf) + return NULL; + + ret = readn(fd, buf, len); + if (ret == (ssize_t)len) { + /* + * strings are padded by zeroes + * thus the actual strlen of buf + * may be less than len + */ + return buf; + } + + free(buf); + return NULL; +} + +int +perf_header__set_cmdline(int argc, const char **argv) +{ + int i; + + /* + * If header_argv has already been set, do not override it. + * This allows a command to set the cmdline, parse args and + * then call another builtin function that implements a + * command -- e.g, cmd_kvm calling cmd_record. + */ + if (header_argv) + return 0; + + header_argc = (u32)argc; + + /* do not include NULL termination */ + header_argv = calloc(argc, sizeof(char *)); + if (!header_argv) + return -ENOMEM; + + /* + * must copy argv contents because it gets moved + * around during option parsing + */ + for (i = 0; i < argc ; i++) + header_argv[i] = argv[i]; + + return 0; +} + +#define dsos__for_each_with_build_id(pos, head) \ + list_for_each_entry(pos, head, node) \ + if (!pos->has_build_id) \ + continue; \ + else + +static int write_buildid(char *name, size_t name_len, u8 *build_id, + pid_t pid, u16 misc, int fd) +{ + int err; + struct build_id_event b; + size_t len; + + len = name_len + 1; + len = PERF_ALIGN(len, NAME_ALIGN); + + memset(&b, 0, sizeof(b)); + memcpy(&b.build_id, build_id, BUILD_ID_SIZE); + b.pid = pid; + b.header.misc = misc; + b.header.size = sizeof(b) + len; + + err = do_write(fd, &b, sizeof(b)); + if (err < 0) + return err; + + return write_padded(fd, name, name_len + 1, len); +} + +static int __dsos__write_buildid_table(struct list_head *head, pid_t pid, + u16 misc, int fd) +{ + struct dso *pos; + + dsos__for_each_with_build_id(pos, head) { + int err; + char *name; + size_t name_len; + + if (!pos->hit) + continue; + + if (is_vdso_map(pos->short_name)) { + name = (char *) VDSO__MAP_NAME; + name_len = sizeof(VDSO__MAP_NAME) + 1; + } else { + name = pos->long_name; + name_len = pos->long_name_len + 1; + } + + err = write_buildid(name, name_len, pos->build_id, + pid, misc, fd); + if (err) + return err; + } + + return 0; +} + +static int machine__write_buildid_table(struct machine *machine, int fd) +{ + int err; + u16 kmisc = PERF_RECORD_MISC_KERNEL, + umisc = PERF_RECORD_MISC_USER; + + if (!machine__is_host(machine)) { + kmisc = PERF_RECORD_MISC_GUEST_KERNEL; + umisc = PERF_RECORD_MISC_GUEST_USER; + } + + err = __dsos__write_buildid_table(&machine->kernel_dsos, machine->pid, + kmisc, fd); + if (err == 0) + err = __dsos__write_buildid_table(&machine->user_dsos, + machine->pid, umisc, fd); + return err; +} + +static int dsos__write_buildid_table(struct perf_header *header, int fd) +{ + struct perf_session *session = container_of(header, + struct perf_session, header); + struct rb_node *nd; + int err = machine__write_buildid_table(&session->machines.host, fd); + + if (err) + return err; + + for (nd = rb_first(&session->machines.guests); nd; nd = rb_next(nd)) { + struct machine *pos = rb_entry(nd, struct machine, rb_node); + err = machine__write_buildid_table(pos, fd); + if (err) + break; + } + return err; +} + +int build_id_cache__add_s(const char *sbuild_id, const char *debugdir, + const char *name, bool is_kallsyms, bool is_vdso) +{ + const size_t size = PATH_MAX; + char *realname, *filename = zalloc(size), + *linkname = zalloc(size), *targetname; + int len, err = -1; + bool slash = is_kallsyms || is_vdso; + + if (is_kallsyms) { + if (symbol_conf.kptr_restrict) { + pr_debug("Not caching a kptr_restrict'ed /proc/kallsyms\n"); + err = 0; + goto out_free; + } + realname = (char *) name; + } else + realname = realpath(name, NULL); + + if (realname == NULL || filename == NULL || linkname == NULL) + goto out_free; + + len = scnprintf(filename, size, "%s%s%s", + debugdir, slash ? "/" : "", + is_vdso ? VDSO__MAP_NAME : realname); + if (mkdir_p(filename, 0755)) + goto out_free; + + snprintf(filename + len, size - len, "/%s", sbuild_id); + + if (access(filename, F_OK)) { + if (is_kallsyms) { + if (copyfile("/proc/kallsyms", filename)) + goto out_free; + } else if (link(realname, filename) && copyfile(name, filename)) + goto out_free; + } + + len = scnprintf(linkname, size, "%s/.build-id/%.2s", + debugdir, sbuild_id); + + if (access(linkname, X_OK) && mkdir_p(linkname, 0755)) + goto out_free; + + snprintf(linkname + len, size - len, "/%s", sbuild_id + 2); + targetname = filename + strlen(debugdir) - 5; + memcpy(targetname, "../..", 5); + + if (symlink(targetname, linkname) == 0) + err = 0; +out_free: + if (!is_kallsyms) + free(realname); + free(filename); + free(linkname); + return err; +} + +static int build_id_cache__add_b(const u8 *build_id, size_t build_id_size, + const char *name, const char *debugdir, + bool is_kallsyms, bool is_vdso) +{ + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + build_id__sprintf(build_id, build_id_size, sbuild_id); + + return build_id_cache__add_s(sbuild_id, debugdir, name, + is_kallsyms, is_vdso); +} + +int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir) +{ + const size_t size = PATH_MAX; + char *filename = zalloc(size), + *linkname = zalloc(size); + int err = -1; + + if (filename == NULL || linkname == NULL) + goto out_free; + + snprintf(linkname, size, "%s/.build-id/%.2s/%s", + debugdir, sbuild_id, sbuild_id + 2); + + if (access(linkname, F_OK)) + goto out_free; + + if (readlink(linkname, filename, size - 1) < 0) + goto out_free; + + if (unlink(linkname)) + goto out_free; + + /* + * Since the link is relative, we must make it absolute: + */ + snprintf(linkname, size, "%s/.build-id/%.2s/%s", + debugdir, sbuild_id, filename); + + if (unlink(linkname)) + goto out_free; + + err = 0; +out_free: + free(filename); + free(linkname); + return err; +} + +static int dso__cache_build_id(struct dso *dso, const char *debugdir) +{ + bool is_kallsyms = dso->kernel && dso->long_name[0] != '/'; + bool is_vdso = is_vdso_map(dso->short_name); + + return build_id_cache__add_b(dso->build_id, sizeof(dso->build_id), + dso->long_name, debugdir, + is_kallsyms, is_vdso); +} + +static int __dsos__cache_build_ids(struct list_head *head, const char *debugdir) +{ + struct dso *pos; + int err = 0; + + dsos__for_each_with_build_id(pos, head) + if (dso__cache_build_id(pos, debugdir)) + err = -1; + + return err; +} + +static int machine__cache_build_ids(struct machine *machine, const char *debugdir) +{ + int ret = __dsos__cache_build_ids(&machine->kernel_dsos, debugdir); + ret |= __dsos__cache_build_ids(&machine->user_dsos, debugdir); + return ret; +} + +static int perf_session__cache_build_ids(struct perf_session *session) +{ + struct rb_node *nd; + int ret; + char debugdir[PATH_MAX]; + + snprintf(debugdir, sizeof(debugdir), "%s", buildid_dir); + + if (mkdir(debugdir, 0755) != 0 && errno != EEXIST) + return -1; + + ret = machine__cache_build_ids(&session->machines.host, debugdir); + + for (nd = rb_first(&session->machines.guests); nd; nd = rb_next(nd)) { + struct machine *pos = rb_entry(nd, struct machine, rb_node); + ret |= machine__cache_build_ids(pos, debugdir); + } + return ret ? -1 : 0; +} + +static bool machine__read_build_ids(struct machine *machine, bool with_hits) +{ + bool ret = __dsos__read_build_ids(&machine->kernel_dsos, with_hits); + ret |= __dsos__read_build_ids(&machine->user_dsos, with_hits); + return ret; +} + +static bool perf_session__read_build_ids(struct perf_session *session, bool with_hits) +{ + struct rb_node *nd; + bool ret = machine__read_build_ids(&session->machines.host, with_hits); + + for (nd = rb_first(&session->machines.guests); nd; nd = rb_next(nd)) { + struct machine *pos = rb_entry(nd, struct machine, rb_node); + ret |= machine__read_build_ids(pos, with_hits); + } + + return ret; +} + +static int write_tracing_data(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist) +{ + return read_tracing_data(fd, &evlist->entries); +} + + +static int write_build_id(int fd, struct perf_header *h, + struct perf_evlist *evlist __maybe_unused) +{ + struct perf_session *session; + int err; + + session = container_of(h, struct perf_session, header); + + if (!perf_session__read_build_ids(session, true)) + return -1; + + err = dsos__write_buildid_table(h, fd); + if (err < 0) { + pr_debug("failed to write buildid table\n"); + return err; + } + if (!no_buildid_cache) + perf_session__cache_build_ids(session); + + return 0; +} + +static int write_hostname(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + struct utsname uts; + int ret; + + ret = uname(&uts); + if (ret < 0) + return -1; + + return do_write_string(fd, uts.nodename); +} + +static int write_osrelease(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + struct utsname uts; + int ret; + + ret = uname(&uts); + if (ret < 0) + return -1; + + return do_write_string(fd, uts.release); +} + +static int write_arch(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + struct utsname uts; + int ret; + + ret = uname(&uts); + if (ret < 0) + return -1; + + return do_write_string(fd, uts.machine); +} + +static int write_version(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + return do_write_string(fd, perf_version_string); +} + +static int write_cpudesc(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ +#ifndef CPUINFO_PROC +#define CPUINFO_PROC NULL +#endif + FILE *file; + char *buf = NULL; + char *s, *p; + const char *search = CPUINFO_PROC; + size_t len = 0; + int ret = -1; + + if (!search) + return -1; + + file = fopen("/proc/cpuinfo", "r"); + if (!file) + return -1; + + while (getline(&buf, &len, file) > 0) { + ret = strncmp(buf, search, strlen(search)); + if (!ret) + break; + } + + if (ret) + goto done; + + s = buf; + + p = strchr(buf, ':'); + if (p && *(p+1) == ' ' && *(p+2)) + s = p + 2; + p = strchr(s, '\n'); + if (p) + *p = '\0'; + + /* squash extra space characters (branding string) */ + p = s; + while (*p) { + if (isspace(*p)) { + char *r = p + 1; + char *q = r; + *p = ' '; + while (*q && isspace(*q)) + q++; + if (q != (p+1)) + while ((*r++ = *q++)); + } + p++; + } + ret = do_write_string(fd, s); +done: + free(buf); + fclose(file); + return ret; +} + +static int write_nrcpus(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + long nr; + u32 nrc, nra; + int ret; + + nr = sysconf(_SC_NPROCESSORS_CONF); + if (nr < 0) + return -1; + + nrc = (u32)(nr & UINT_MAX); + + nr = sysconf(_SC_NPROCESSORS_ONLN); + if (nr < 0) + return -1; + + nra = (u32)(nr & UINT_MAX); + + ret = do_write(fd, &nrc, sizeof(nrc)); + if (ret < 0) + return ret; + + return do_write(fd, &nra, sizeof(nra)); +} + +static int write_event_desc(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist) +{ + struct perf_evsel *evsel; + u32 nre, nri, sz; + int ret; + + nre = evlist->nr_entries; + + /* + * write number of events + */ + ret = do_write(fd, &nre, sizeof(nre)); + if (ret < 0) + return ret; + + /* + * size of perf_event_attr struct + */ + sz = (u32)sizeof(evsel->attr); + ret = do_write(fd, &sz, sizeof(sz)); + if (ret < 0) + return ret; + + list_for_each_entry(evsel, &evlist->entries, node) { + + ret = do_write(fd, &evsel->attr, sz); + if (ret < 0) + return ret; + /* + * write number of unique id per event + * there is one id per instance of an event + * + * copy into an nri to be independent of the + * type of ids, + */ + nri = evsel->ids; + ret = do_write(fd, &nri, sizeof(nri)); + if (ret < 0) + return ret; + + /* + * write event string as passed on cmdline + */ + ret = do_write_string(fd, perf_evsel__name(evsel)); + if (ret < 0) + return ret; + /* + * write unique ids for this event + */ + ret = do_write(fd, evsel->id, evsel->ids * sizeof(u64)); + if (ret < 0) + return ret; + } + return 0; +} + +static int write_cmdline(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + char buf[MAXPATHLEN]; + char proc[32]; + u32 i, n; + int ret; + + /* + * actual atual path to perf binary + */ + sprintf(proc, "/proc/%d/exe", getpid()); + ret = readlink(proc, buf, sizeof(buf)); + if (ret <= 0) + return -1; + + /* readlink() does not add null termination */ + buf[ret] = '\0'; + + /* account for binary path */ + n = header_argc + 1; + + ret = do_write(fd, &n, sizeof(n)); + if (ret < 0) + return ret; + + ret = do_write_string(fd, buf); + if (ret < 0) + return ret; + + for (i = 0 ; i < header_argc; i++) { + ret = do_write_string(fd, header_argv[i]); + if (ret < 0) + return ret; + } + return 0; +} + +#define CORE_SIB_FMT \ + "/sys/devices/system/cpu/cpu%d/topology/core_siblings_list" +#define THRD_SIB_FMT \ + "/sys/devices/system/cpu/cpu%d/topology/thread_siblings_list" + +struct cpu_topo { + u32 core_sib; + u32 thread_sib; + char **core_siblings; + char **thread_siblings; +}; + +static int build_cpu_topo(struct cpu_topo *tp, int cpu) +{ + FILE *fp; + char filename[MAXPATHLEN]; + char *buf = NULL, *p; + size_t len = 0; + u32 i = 0; + int ret = -1; + + sprintf(filename, CORE_SIB_FMT, cpu); + fp = fopen(filename, "r"); + if (!fp) + return -1; + + if (getline(&buf, &len, fp) <= 0) + goto done; + + fclose(fp); + + p = strchr(buf, '\n'); + if (p) + *p = '\0'; + + for (i = 0; i < tp->core_sib; i++) { + if (!strcmp(buf, tp->core_siblings[i])) + break; + } + if (i == tp->core_sib) { + tp->core_siblings[i] = buf; + tp->core_sib++; + buf = NULL; + len = 0; + } + + sprintf(filename, THRD_SIB_FMT, cpu); + fp = fopen(filename, "r"); + if (!fp) + goto done; + + if (getline(&buf, &len, fp) <= 0) + goto done; + + p = strchr(buf, '\n'); + if (p) + *p = '\0'; + + for (i = 0; i < tp->thread_sib; i++) { + if (!strcmp(buf, tp->thread_siblings[i])) + break; + } + if (i == tp->thread_sib) { + tp->thread_siblings[i] = buf; + tp->thread_sib++; + buf = NULL; + } + ret = 0; +done: + if(fp) + fclose(fp); + free(buf); + return ret; +} + +static void free_cpu_topo(struct cpu_topo *tp) +{ + u32 i; + + if (!tp) + return; + + for (i = 0 ; i < tp->core_sib; i++) + free(tp->core_siblings[i]); + + for (i = 0 ; i < tp->thread_sib; i++) + free(tp->thread_siblings[i]); + + free(tp); +} + +static struct cpu_topo *build_cpu_topology(void) +{ + struct cpu_topo *tp; + void *addr; + u32 nr, i; + size_t sz; + long ncpus; + int ret = -1; + + ncpus = sysconf(_SC_NPROCESSORS_CONF); + if (ncpus < 0) + return NULL; + + nr = (u32)(ncpus & UINT_MAX); + + sz = nr * sizeof(char *); + + addr = calloc(1, sizeof(*tp) + 2 * sz); + if (!addr) + return NULL; + + tp = addr; + + addr += sizeof(*tp); + tp->core_siblings = addr; + addr += sz; + tp->thread_siblings = addr; + + for (i = 0; i < nr; i++) { + ret = build_cpu_topo(tp, i); + if (ret < 0) + break; + } + if (ret) { + free_cpu_topo(tp); + tp = NULL; + } + return tp; +} + +static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + struct cpu_topo *tp; + u32 i; + int ret; + + tp = build_cpu_topology(); + if (!tp) + return -1; + + ret = do_write(fd, &tp->core_sib, sizeof(tp->core_sib)); + if (ret < 0) + goto done; + + for (i = 0; i < tp->core_sib; i++) { + ret = do_write_string(fd, tp->core_siblings[i]); + if (ret < 0) + goto done; + } + ret = do_write(fd, &tp->thread_sib, sizeof(tp->thread_sib)); + if (ret < 0) + goto done; + + for (i = 0; i < tp->thread_sib; i++) { + ret = do_write_string(fd, tp->thread_siblings[i]); + if (ret < 0) + break; + } +done: + free_cpu_topo(tp); + return ret; +} + + + +static int write_total_mem(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + char *buf = NULL; + FILE *fp; + size_t len = 0; + int ret = -1, n; + uint64_t mem; + + fp = fopen("/proc/meminfo", "r"); + if (!fp) + return -1; + + while (getline(&buf, &len, fp) > 0) { + ret = strncmp(buf, "MemTotal:", 9); + if (!ret) + break; + } + if (!ret) { + n = sscanf(buf, "%*s %"PRIu64, &mem); + if (n == 1) + ret = do_write(fd, &mem, sizeof(mem)); + } + free(buf); + fclose(fp); + return ret; +} + +static int write_topo_node(int fd, int node) +{ + char str[MAXPATHLEN]; + char field[32]; + char *buf = NULL, *p; + size_t len = 0; + FILE *fp; + u64 mem_total, mem_free, mem; + int ret = -1; + + sprintf(str, "/sys/devices/system/node/node%d/meminfo", node); + fp = fopen(str, "r"); + if (!fp) + return -1; + + while (getline(&buf, &len, fp) > 0) { + /* skip over invalid lines */ + if (!strchr(buf, ':')) + continue; + if (sscanf(buf, "%*s %*d %s %"PRIu64, field, &mem) != 2) + goto done; + if (!strcmp(field, "MemTotal:")) + mem_total = mem; + if (!strcmp(field, "MemFree:")) + mem_free = mem; + } + + fclose(fp); + fp = NULL; + + ret = do_write(fd, &mem_total, sizeof(u64)); + if (ret) + goto done; + + ret = do_write(fd, &mem_free, sizeof(u64)); + if (ret) + goto done; + + ret = -1; + sprintf(str, "/sys/devices/system/node/node%d/cpulist", node); + + fp = fopen(str, "r"); + if (!fp) + goto done; + + if (getline(&buf, &len, fp) <= 0) + goto done; + + p = strchr(buf, '\n'); + if (p) + *p = '\0'; + + ret = do_write_string(fd, buf); +done: + free(buf); + if (fp) + fclose(fp); + return ret; +} + +static int write_numa_topology(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + char *buf = NULL; + size_t len = 0; + FILE *fp; + struct cpu_map *node_map = NULL; + char *c; + u32 nr, i, j; + int ret = -1; + + fp = fopen("/sys/devices/system/node/online", "r"); + if (!fp) + return -1; + + if (getline(&buf, &len, fp) <= 0) + goto done; + + c = strchr(buf, '\n'); + if (c) + *c = '\0'; + + node_map = cpu_map__new(buf); + if (!node_map) + goto done; + + nr = (u32)node_map->nr; + + ret = do_write(fd, &nr, sizeof(nr)); + if (ret < 0) + goto done; + + for (i = 0; i < nr; i++) { + j = (u32)node_map->map[i]; + ret = do_write(fd, &j, sizeof(j)); + if (ret < 0) + break; + + ret = write_topo_node(fd, i); + if (ret < 0) + break; + } +done: + free(buf); + fclose(fp); + free(node_map); + return ret; +} + +/* + * File format: + * + * struct pmu_mappings { + * u32 pmu_num; + * struct pmu_map { + * u32 type; + * char name[]; + * }[pmu_num]; + * }; + */ + +static int write_pmu_mappings(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + struct perf_pmu *pmu = NULL; + off_t offset = lseek(fd, 0, SEEK_CUR); + __u32 pmu_num = 0; + int ret; + + /* write real pmu_num later */ + ret = do_write(fd, &pmu_num, sizeof(pmu_num)); + if (ret < 0) + return ret; + + while ((pmu = perf_pmu__scan(pmu))) { + if (!pmu->name) + continue; + pmu_num++; + + ret = do_write(fd, &pmu->type, sizeof(pmu->type)); + if (ret < 0) + return ret; + + ret = do_write_string(fd, pmu->name); + if (ret < 0) + return ret; + } + + if (pwrite(fd, &pmu_num, sizeof(pmu_num), offset) != sizeof(pmu_num)) { + /* discard all */ + lseek(fd, offset, SEEK_SET); + return -1; + } + + return 0; +} + +/* + * File format: + * + * struct group_descs { + * u32 nr_groups; + * struct group_desc { + * char name[]; + * u32 leader_idx; + * u32 nr_members; + * }[nr_groups]; + * }; + */ +static int write_group_desc(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist) +{ + u32 nr_groups = evlist->nr_groups; + struct perf_evsel *evsel; + int ret; + + ret = do_write(fd, &nr_groups, sizeof(nr_groups)); + if (ret < 0) + return ret; + + list_for_each_entry(evsel, &evlist->entries, node) { + if (perf_evsel__is_group_leader(evsel) && + evsel->nr_members > 1) { + const char *name = evsel->group_name ?: "{anon_group}"; + u32 leader_idx = evsel->idx; + u32 nr_members = evsel->nr_members; + + ret = do_write_string(fd, name); + if (ret < 0) + return ret; + + ret = do_write(fd, &leader_idx, sizeof(leader_idx)); + if (ret < 0) + return ret; + + ret = do_write(fd, &nr_members, sizeof(nr_members)); + if (ret < 0) + return ret; + } + } + return 0; +} + +/* + * default get_cpuid(): nothing gets recorded + * actual implementation must be in arch/$(ARCH)/util/header.c + */ +int __attribute__ ((weak)) get_cpuid(char *buffer __maybe_unused, + size_t sz __maybe_unused) +{ + return -1; +} + +static int write_cpuid(int fd, struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + char buffer[64]; + int ret; + + ret = get_cpuid(buffer, sizeof(buffer)); + if (!ret) + goto write_it; + + return -1; +write_it: + return do_write_string(fd, buffer); +} + +static int write_branch_stack(int fd __maybe_unused, + struct perf_header *h __maybe_unused, + struct perf_evlist *evlist __maybe_unused) +{ + return 0; +} + +static void print_hostname(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + fprintf(fp, "# hostname : %s\n", ph->env.hostname); +} + +static void print_osrelease(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + fprintf(fp, "# os release : %s\n", ph->env.os_release); +} + +static void print_arch(struct perf_header *ph, int fd __maybe_unused, FILE *fp) +{ + fprintf(fp, "# arch : %s\n", ph->env.arch); +} + +static void print_cpudesc(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + fprintf(fp, "# cpudesc : %s\n", ph->env.cpu_desc); +} + +static void print_nrcpus(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + fprintf(fp, "# nrcpus online : %u\n", ph->env.nr_cpus_online); + fprintf(fp, "# nrcpus avail : %u\n", ph->env.nr_cpus_avail); +} + +static void print_version(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + fprintf(fp, "# perf version : %s\n", ph->env.version); +} + +static void print_cmdline(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + int nr, i; + char *str; + + nr = ph->env.nr_cmdline; + str = ph->env.cmdline; + + fprintf(fp, "# cmdline : "); + + for (i = 0; i < nr; i++) { + fprintf(fp, "%s ", str); + str += strlen(str) + 1; + } + fputc('\n', fp); +} + +static void print_cpu_topology(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + int nr, i; + char *str; + + nr = ph->env.nr_sibling_cores; + str = ph->env.sibling_cores; + + for (i = 0; i < nr; i++) { + fprintf(fp, "# sibling cores : %s\n", str); + str += strlen(str) + 1; + } + + nr = ph->env.nr_sibling_threads; + str = ph->env.sibling_threads; + + for (i = 0; i < nr; i++) { + fprintf(fp, "# sibling threads : %s\n", str); + str += strlen(str) + 1; + } +} + +static void free_event_desc(struct perf_evsel *events) +{ + struct perf_evsel *evsel; + + if (!events) + return; + + for (evsel = events; evsel->attr.size; evsel++) { + if (evsel->name) + free(evsel->name); + if (evsel->id) + free(evsel->id); + } + + free(events); +} + +static struct perf_evsel * +read_event_desc(struct perf_header *ph, int fd) +{ + struct perf_evsel *evsel, *events = NULL; + u64 *id; + void *buf = NULL; + u32 nre, sz, nr, i, j; + ssize_t ret; + size_t msz; + + /* number of events */ + ret = readn(fd, &nre, sizeof(nre)); + if (ret != (ssize_t)sizeof(nre)) + goto error; + + if (ph->needs_swap) + nre = bswap_32(nre); + + ret = readn(fd, &sz, sizeof(sz)); + if (ret != (ssize_t)sizeof(sz)) + goto error; + + if (ph->needs_swap) + sz = bswap_32(sz); + + /* buffer to hold on file attr struct */ + buf = malloc(sz); + if (!buf) + goto error; + + /* the last event terminates with evsel->attr.size == 0: */ + events = calloc(nre + 1, sizeof(*events)); + if (!events) + goto error; + + msz = sizeof(evsel->attr); + if (sz < msz) + msz = sz; + + for (i = 0, evsel = events; i < nre; evsel++, i++) { + evsel->idx = i; + + /* + * must read entire on-file attr struct to + * sync up with layout. + */ + ret = readn(fd, buf, sz); + if (ret != (ssize_t)sz) + goto error; + + if (ph->needs_swap) + perf_event__attr_swap(buf); + + memcpy(&evsel->attr, buf, msz); + + ret = readn(fd, &nr, sizeof(nr)); + if (ret != (ssize_t)sizeof(nr)) + goto error; + + if (ph->needs_swap) { + nr = bswap_32(nr); + evsel->needs_swap = true; + } + + evsel->name = do_read_string(fd, ph); + + if (!nr) + continue; + + id = calloc(nr, sizeof(*id)); + if (!id) + goto error; + evsel->ids = nr; + evsel->id = id; + + for (j = 0 ; j < nr; j++) { + ret = readn(fd, id, sizeof(*id)); + if (ret != (ssize_t)sizeof(*id)) + goto error; + if (ph->needs_swap) + *id = bswap_64(*id); + id++; + } + } +out: + if (buf) + free(buf); + return events; +error: + if (events) + free_event_desc(events); + events = NULL; + goto out; +} + +static void print_event_desc(struct perf_header *ph, int fd, FILE *fp) +{ + struct perf_evsel *evsel, *events = read_event_desc(ph, fd); + u32 j; + u64 *id; + + if (!events) { + fprintf(fp, "# event desc: not available or unable to read\n"); + return; + } + + for (evsel = events; evsel->attr.size; evsel++) { + fprintf(fp, "# event : name = %s, ", evsel->name); + + fprintf(fp, "type = %d, config = 0x%"PRIx64 + ", config1 = 0x%"PRIx64", config2 = 0x%"PRIx64, + evsel->attr.type, + (u64)evsel->attr.config, + (u64)evsel->attr.config1, + (u64)evsel->attr.config2); + + fprintf(fp, ", excl_usr = %d, excl_kern = %d", + evsel->attr.exclude_user, + evsel->attr.exclude_kernel); + + fprintf(fp, ", excl_host = %d, excl_guest = %d", + evsel->attr.exclude_host, + evsel->attr.exclude_guest); + + fprintf(fp, ", precise_ip = %d", evsel->attr.precise_ip); + + if (evsel->ids) { + fprintf(fp, ", id = {"); + for (j = 0, id = evsel->id; j < evsel->ids; j++, id++) { + if (j) + fputc(',', fp); + fprintf(fp, " %"PRIu64, *id); + } + fprintf(fp, " }"); + } + + fputc('\n', fp); + } + + free_event_desc(events); +} + +static void print_total_mem(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + fprintf(fp, "# total memory : %Lu kB\n", ph->env.total_mem); +} + +static void print_numa_topology(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + u32 nr, c, i; + char *str, *tmp; + uint64_t mem_total, mem_free; + + /* nr nodes */ + nr = ph->env.nr_numa_nodes; + str = ph->env.numa_nodes; + + for (i = 0; i < nr; i++) { + /* node number */ + c = strtoul(str, &tmp, 0); + if (*tmp != ':') + goto error; + + str = tmp + 1; + mem_total = strtoull(str, &tmp, 0); + if (*tmp != ':') + goto error; + + str = tmp + 1; + mem_free = strtoull(str, &tmp, 0); + if (*tmp != ':') + goto error; + + fprintf(fp, "# node%u meminfo : total = %"PRIu64" kB," + " free = %"PRIu64" kB\n", + c, mem_total, mem_free); + + str = tmp + 1; + fprintf(fp, "# node%u cpu list : %s\n", c, str); + + str += strlen(str) + 1; + } + return; +error: + fprintf(fp, "# numa topology : not available\n"); +} + +static void print_cpuid(struct perf_header *ph, int fd __maybe_unused, FILE *fp) +{ + fprintf(fp, "# cpuid : %s\n", ph->env.cpuid); +} + +static void print_branch_stack(struct perf_header *ph __maybe_unused, + int fd __maybe_unused, FILE *fp) +{ + fprintf(fp, "# contains samples with branch stack\n"); +} + +static void print_pmu_mappings(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + const char *delimiter = "# pmu mappings: "; + char *str, *tmp; + u32 pmu_num; + u32 type; + + pmu_num = ph->env.nr_pmu_mappings; + if (!pmu_num) { + fprintf(fp, "# pmu mappings: not available\n"); + return; + } + + str = ph->env.pmu_mappings; + + while (pmu_num) { + type = strtoul(str, &tmp, 0); + if (*tmp != ':') + goto error; + + str = tmp + 1; + fprintf(fp, "%s%s = %" PRIu32, delimiter, str, type); + + delimiter = ", "; + str += strlen(str) + 1; + pmu_num--; + } + + fprintf(fp, "\n"); + + if (!pmu_num) + return; +error: + fprintf(fp, "# pmu mappings: unable to read\n"); +} + +static void print_group_desc(struct perf_header *ph, int fd __maybe_unused, + FILE *fp) +{ + struct perf_session *session; + struct perf_evsel *evsel; + u32 nr = 0; + + session = container_of(ph, struct perf_session, header); + + list_for_each_entry(evsel, &session->evlist->entries, node) { + if (perf_evsel__is_group_leader(evsel) && + evsel->nr_members > 1) { + fprintf(fp, "# group: %s{%s", evsel->group_name ?: "", + perf_evsel__name(evsel)); + + nr = evsel->nr_members - 1; + } else if (nr) { + fprintf(fp, ",%s", perf_evsel__name(evsel)); + + if (--nr == 0) + fprintf(fp, "}\n"); + } + } +} + +static int __event_process_build_id(struct build_id_event *bev, + char *filename, + struct perf_session *session) +{ + int err = -1; + struct list_head *head; + struct machine *machine; + u16 misc; + struct dso *dso; + enum dso_kernel_type dso_type; + + machine = perf_session__findnew_machine(session, bev->pid); + if (!machine) + goto out; + + misc = bev->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; + + switch (misc) { + case PERF_RECORD_MISC_KERNEL: + dso_type = DSO_TYPE_KERNEL; + head = &machine->kernel_dsos; + break; + case PERF_RECORD_MISC_GUEST_KERNEL: + dso_type = DSO_TYPE_GUEST_KERNEL; + head = &machine->kernel_dsos; + break; + case PERF_RECORD_MISC_USER: + case PERF_RECORD_MISC_GUEST_USER: + dso_type = DSO_TYPE_USER; + head = &machine->user_dsos; + break; + default: + goto out; + } + + dso = __dsos__findnew(head, filename); + if (dso != NULL) { + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + dso__set_build_id(dso, &bev->build_id); + + if (filename[0] == '[') + dso->kernel = dso_type; + + build_id__sprintf(dso->build_id, sizeof(dso->build_id), + sbuild_id); + pr_debug("build id event received for %s: %s\n", + dso->long_name, sbuild_id); + } + + err = 0; +out: + return err; +} + +static int perf_header__read_build_ids_abi_quirk(struct perf_header *header, + int input, u64 offset, u64 size) +{ + struct perf_session *session = container_of(header, struct perf_session, header); + struct { + struct perf_event_header header; + u8 build_id[PERF_ALIGN(BUILD_ID_SIZE, sizeof(u64))]; + char filename[0]; + } old_bev; + struct build_id_event bev; + char filename[PATH_MAX]; + u64 limit = offset + size; + + while (offset < limit) { + ssize_t len; + + if (readn(input, &old_bev, sizeof(old_bev)) != sizeof(old_bev)) + return -1; + + if (header->needs_swap) + perf_event_header__bswap(&old_bev.header); + + len = old_bev.header.size - sizeof(old_bev); + if (readn(input, filename, len) != len) + return -1; + + bev.header = old_bev.header; + + /* + * As the pid is the missing value, we need to fill + * it properly. The header.misc value give us nice hint. + */ + bev.pid = HOST_KERNEL_ID; + if (bev.header.misc == PERF_RECORD_MISC_GUEST_USER || + bev.header.misc == PERF_RECORD_MISC_GUEST_KERNEL) + bev.pid = DEFAULT_GUEST_KERNEL_ID; + + memcpy(bev.build_id, old_bev.build_id, sizeof(bev.build_id)); + __event_process_build_id(&bev, filename, session); + + offset += bev.header.size; + } + + return 0; +} + +static int perf_header__read_build_ids(struct perf_header *header, + int input, u64 offset, u64 size) +{ + struct perf_session *session = container_of(header, struct perf_session, header); + struct build_id_event bev; + char filename[PATH_MAX]; + u64 limit = offset + size, orig_offset = offset; + int err = -1; + + while (offset < limit) { + ssize_t len; + + if (readn(input, &bev, sizeof(bev)) != sizeof(bev)) + goto out; + + if (header->needs_swap) + perf_event_header__bswap(&bev.header); + + len = bev.header.size - sizeof(bev); + if (readn(input, filename, len) != len) + goto out; + /* + * The a1645ce1 changeset: + * + * "perf: 'perf kvm' tool for monitoring guest performance from host" + * + * Added a field to struct build_id_event that broke the file + * format. + * + * Since the kernel build-id is the first entry, process the + * table using the old format if the well known + * '[kernel.kallsyms]' string for the kernel build-id has the + * first 4 characters chopped off (where the pid_t sits). + */ + if (memcmp(filename, "nel.kallsyms]", 13) == 0) { + if (lseek(input, orig_offset, SEEK_SET) == (off_t)-1) + return -1; + return perf_header__read_build_ids_abi_quirk(header, input, offset, size); + } + + __event_process_build_id(&bev, filename, session); + + offset += bev.header.size; + } + err = 0; +out: + return err; +} + +static int process_tracing_data(struct perf_file_section *section __maybe_unused, + struct perf_header *ph __maybe_unused, + int fd, void *data) +{ + trace_report(fd, data, false); + return 0; +} + +static int process_build_id(struct perf_file_section *section, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + if (perf_header__read_build_ids(ph, fd, section->offset, section->size)) + pr_debug("Failed to read buildids, continuing...\n"); + return 0; +} + +static int process_hostname(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + ph->env.hostname = do_read_string(fd, ph); + return ph->env.hostname ? 0 : -ENOMEM; +} + +static int process_osrelease(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + ph->env.os_release = do_read_string(fd, ph); + return ph->env.os_release ? 0 : -ENOMEM; +} + +static int process_version(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + ph->env.version = do_read_string(fd, ph); + return ph->env.version ? 0 : -ENOMEM; +} + +static int process_arch(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + ph->env.arch = do_read_string(fd, ph); + return ph->env.arch ? 0 : -ENOMEM; +} + +static int process_nrcpus(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + size_t ret; + u32 nr; + + ret = readn(fd, &nr, sizeof(nr)); + if (ret != sizeof(nr)) + return -1; + + if (ph->needs_swap) + nr = bswap_32(nr); + + ph->env.nr_cpus_online = nr; + + ret = readn(fd, &nr, sizeof(nr)); + if (ret != sizeof(nr)) + return -1; + + if (ph->needs_swap) + nr = bswap_32(nr); + + ph->env.nr_cpus_avail = nr; + return 0; +} + +static int process_cpudesc(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + ph->env.cpu_desc = do_read_string(fd, ph); + return ph->env.cpu_desc ? 0 : -ENOMEM; +} + +static int process_cpuid(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + ph->env.cpuid = do_read_string(fd, ph); + return ph->env.cpuid ? 0 : -ENOMEM; +} + +static int process_total_mem(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + uint64_t mem; + size_t ret; + + ret = readn(fd, &mem, sizeof(mem)); + if (ret != sizeof(mem)) + return -1; + + if (ph->needs_swap) + mem = bswap_64(mem); + + ph->env.total_mem = mem; + return 0; +} + +static struct perf_evsel * +perf_evlist__find_by_index(struct perf_evlist *evlist, int idx) +{ + struct perf_evsel *evsel; + + list_for_each_entry(evsel, &evlist->entries, node) { + if (evsel->idx == idx) + return evsel; + } + + return NULL; +} + +static void +perf_evlist__set_event_name(struct perf_evlist *evlist, + struct perf_evsel *event) +{ + struct perf_evsel *evsel; + + if (!event->name) + return; + + evsel = perf_evlist__find_by_index(evlist, event->idx); + if (!evsel) + return; + + if (evsel->name) + return; + + evsel->name = strdup(event->name); +} + +static int +process_event_desc(struct perf_file_section *section __maybe_unused, + struct perf_header *header, int fd, + void *data __maybe_unused) +{ + struct perf_session *session; + struct perf_evsel *evsel, *events = read_event_desc(header, fd); + + if (!events) + return 0; + + session = container_of(header, struct perf_session, header); + for (evsel = events; evsel->attr.size; evsel++) + perf_evlist__set_event_name(session->evlist, evsel); + + free_event_desc(events); + + return 0; +} + +static int process_cmdline(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + size_t ret; + char *str; + u32 nr, i; + struct strbuf sb; + + ret = readn(fd, &nr, sizeof(nr)); + if (ret != sizeof(nr)) + return -1; + + if (ph->needs_swap) + nr = bswap_32(nr); + + ph->env.nr_cmdline = nr; + strbuf_init(&sb, 128); + + for (i = 0; i < nr; i++) { + str = do_read_string(fd, ph); + if (!str) + goto error; + + /* include a NULL character at the end */ + strbuf_add(&sb, str, strlen(str) + 1); + free(str); + } + ph->env.cmdline = strbuf_detach(&sb, NULL); + return 0; + +error: + strbuf_release(&sb); + return -1; +} + +static int process_cpu_topology(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + size_t ret; + u32 nr, i; + char *str; + struct strbuf sb; + + ret = readn(fd, &nr, sizeof(nr)); + if (ret != sizeof(nr)) + return -1; + + if (ph->needs_swap) + nr = bswap_32(nr); + + ph->env.nr_sibling_cores = nr; + strbuf_init(&sb, 128); + + for (i = 0; i < nr; i++) { + str = do_read_string(fd, ph); + if (!str) + goto error; + + /* include a NULL character at the end */ + strbuf_add(&sb, str, strlen(str) + 1); + free(str); + } + ph->env.sibling_cores = strbuf_detach(&sb, NULL); + + ret = readn(fd, &nr, sizeof(nr)); + if (ret != sizeof(nr)) + return -1; + + if (ph->needs_swap) + nr = bswap_32(nr); + + ph->env.nr_sibling_threads = nr; + + for (i = 0; i < nr; i++) { + str = do_read_string(fd, ph); + if (!str) + goto error; + + /* include a NULL character at the end */ + strbuf_add(&sb, str, strlen(str) + 1); + free(str); + } + ph->env.sibling_threads = strbuf_detach(&sb, NULL); + return 0; + +error: + strbuf_release(&sb); + return -1; +} + +static int process_numa_topology(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + size_t ret; + u32 nr, node, i; + char *str; + uint64_t mem_total, mem_free; + struct strbuf sb; + + /* nr nodes */ + ret = readn(fd, &nr, sizeof(nr)); + if (ret != sizeof(nr)) + goto error; + + if (ph->needs_swap) + nr = bswap_32(nr); + + ph->env.nr_numa_nodes = nr; + strbuf_init(&sb, 256); + + for (i = 0; i < nr; i++) { + /* node number */ + ret = readn(fd, &node, sizeof(node)); + if (ret != sizeof(node)) + goto error; + + ret = readn(fd, &mem_total, sizeof(u64)); + if (ret != sizeof(u64)) + goto error; + + ret = readn(fd, &mem_free, sizeof(u64)); + if (ret != sizeof(u64)) + goto error; + + if (ph->needs_swap) { + node = bswap_32(node); + mem_total = bswap_64(mem_total); + mem_free = bswap_64(mem_free); + } + + strbuf_addf(&sb, "%u:%"PRIu64":%"PRIu64":", + node, mem_total, mem_free); + + str = do_read_string(fd, ph); + if (!str) + goto error; + + /* include a NULL character at the end */ + strbuf_add(&sb, str, strlen(str) + 1); + free(str); + } + ph->env.numa_nodes = strbuf_detach(&sb, NULL); + return 0; + +error: + strbuf_release(&sb); + return -1; +} + +static int process_pmu_mappings(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + size_t ret; + char *name; + u32 pmu_num; + u32 type; + struct strbuf sb; + + ret = readn(fd, &pmu_num, sizeof(pmu_num)); + if (ret != sizeof(pmu_num)) + return -1; + + if (ph->needs_swap) + pmu_num = bswap_32(pmu_num); + + if (!pmu_num) { + pr_debug("pmu mappings not available\n"); + return 0; + } + + ph->env.nr_pmu_mappings = pmu_num; + strbuf_init(&sb, 128); + + while (pmu_num) { + if (readn(fd, &type, sizeof(type)) != sizeof(type)) + goto error; + if (ph->needs_swap) + type = bswap_32(type); + + name = do_read_string(fd, ph); + if (!name) + goto error; + + strbuf_addf(&sb, "%u:%s", type, name); + /* include a NULL character at the end */ + strbuf_add(&sb, "", 1); + + free(name); + pmu_num--; + } + ph->env.pmu_mappings = strbuf_detach(&sb, NULL); + return 0; + +error: + strbuf_release(&sb); + return -1; +} + +static int process_group_desc(struct perf_file_section *section __maybe_unused, + struct perf_header *ph, int fd, + void *data __maybe_unused) +{ + size_t ret = -1; + u32 i, nr, nr_groups; + struct perf_session *session; + struct perf_evsel *evsel, *leader = NULL; + struct group_desc { + char *name; + u32 leader_idx; + u32 nr_members; + } *desc; + + if (readn(fd, &nr_groups, sizeof(nr_groups)) != sizeof(nr_groups)) + return -1; + + if (ph->needs_swap) + nr_groups = bswap_32(nr_groups); + + ph->env.nr_groups = nr_groups; + if (!nr_groups) { + pr_debug("group desc not available\n"); + return 0; + } + + desc = calloc(nr_groups, sizeof(*desc)); + if (!desc) + return -1; + + for (i = 0; i < nr_groups; i++) { + desc[i].name = do_read_string(fd, ph); + if (!desc[i].name) + goto out_free; + + if (readn(fd, &desc[i].leader_idx, sizeof(u32)) != sizeof(u32)) + goto out_free; + + if (readn(fd, &desc[i].nr_members, sizeof(u32)) != sizeof(u32)) + goto out_free; + + if (ph->needs_swap) { + desc[i].leader_idx = bswap_32(desc[i].leader_idx); + desc[i].nr_members = bswap_32(desc[i].nr_members); + } + } + + /* + * Rebuild group relationship based on the group_desc + */ + session = container_of(ph, struct perf_session, header); + session->evlist->nr_groups = nr_groups; + + i = nr = 0; + list_for_each_entry(evsel, &session->evlist->entries, node) { + if (evsel->idx == (int) desc[i].leader_idx) { + evsel->leader = evsel; + /* {anon_group} is a dummy name */ + if (strcmp(desc[i].name, "{anon_group}")) + evsel->group_name = desc[i].name; + evsel->nr_members = desc[i].nr_members; + + if (i >= nr_groups || nr > 0) { + pr_debug("invalid group desc\n"); + goto out_free; + } + + leader = evsel; + nr = evsel->nr_members - 1; + i++; + } else if (nr) { + /* This is a group member */ + evsel->leader = leader; + + nr--; + } + } + + if (i != nr_groups || nr != 0) { + pr_debug("invalid group desc\n"); + goto out_free; + } + + ret = 0; +out_free: + while ((int) --i >= 0) + free(desc[i].name); + free(desc); + + return ret; +} + +struct feature_ops { + int (*write)(int fd, struct perf_header *h, struct perf_evlist *evlist); + void (*print)(struct perf_header *h, int fd, FILE *fp); + int (*process)(struct perf_file_section *section, + struct perf_header *h, int fd, void *data); + const char *name; + bool full_only; +}; + +#define FEAT_OPA(n, func) \ + [n] = { .name = #n, .write = write_##func, .print = print_##func } +#define FEAT_OPP(n, func) \ + [n] = { .name = #n, .write = write_##func, .print = print_##func, \ + .process = process_##func } +#define FEAT_OPF(n, func) \ + [n] = { .name = #n, .write = write_##func, .print = print_##func, \ + .process = process_##func, .full_only = true } + +/* feature_ops not implemented: */ +#define print_tracing_data NULL +#define print_build_id NULL + +static const struct feature_ops feat_ops[HEADER_LAST_FEATURE] = { + FEAT_OPP(HEADER_TRACING_DATA, tracing_data), + FEAT_OPP(HEADER_BUILD_ID, build_id), + FEAT_OPP(HEADER_HOSTNAME, hostname), + FEAT_OPP(HEADER_OSRELEASE, osrelease), + FEAT_OPP(HEADER_VERSION, version), + FEAT_OPP(HEADER_ARCH, arch), + FEAT_OPP(HEADER_NRCPUS, nrcpus), + FEAT_OPP(HEADER_CPUDESC, cpudesc), + FEAT_OPP(HEADER_CPUID, cpuid), + FEAT_OPP(HEADER_TOTAL_MEM, total_mem), + FEAT_OPP(HEADER_EVENT_DESC, event_desc), + FEAT_OPP(HEADER_CMDLINE, cmdline), + FEAT_OPF(HEADER_CPU_TOPOLOGY, cpu_topology), + FEAT_OPF(HEADER_NUMA_TOPOLOGY, numa_topology), + FEAT_OPA(HEADER_BRANCH_STACK, branch_stack), + FEAT_OPP(HEADER_PMU_MAPPINGS, pmu_mappings), + FEAT_OPP(HEADER_GROUP_DESC, group_desc), +}; + +struct header_print_data { + FILE *fp; + bool full; /* extended list of headers */ +}; + +static int perf_file_section__fprintf_info(struct perf_file_section *section, + struct perf_header *ph, + int feat, int fd, void *data) +{ + struct header_print_data *hd = data; + + if (lseek(fd, section->offset, SEEK_SET) == (off_t)-1) { + pr_debug("Failed to lseek to %" PRIu64 " offset for feature " + "%d, continuing...\n", section->offset, feat); + return 0; + } + if (feat >= HEADER_LAST_FEATURE) { + pr_warning("unknown feature %d\n", feat); + return 0; + } + if (!feat_ops[feat].print) + return 0; + + if (!feat_ops[feat].full_only || hd->full) + feat_ops[feat].print(ph, fd, hd->fp); + else + fprintf(hd->fp, "# %s info available, use -I to display\n", + feat_ops[feat].name); + + return 0; +} + +int perf_header__fprintf_info(struct perf_session *session, FILE *fp, bool full) +{ + struct header_print_data hd; + struct perf_header *header = &session->header; + int fd = session->fd; + hd.fp = fp; + hd.full = full; + + perf_header__process_sections(header, fd, &hd, + perf_file_section__fprintf_info); + return 0; +} + +static int do_write_feat(int fd, struct perf_header *h, int type, + struct perf_file_section **p, + struct perf_evlist *evlist) +{ + int err; + int ret = 0; + + if (perf_header__has_feat(h, type)) { + if (!feat_ops[type].write) + return -1; + + (*p)->offset = lseek(fd, 0, SEEK_CUR); + + err = feat_ops[type].write(fd, h, evlist); + if (err < 0) { + pr_debug("failed to write feature %d\n", type); + + /* undo anything written */ + lseek(fd, (*p)->offset, SEEK_SET); + + return -1; + } + (*p)->size = lseek(fd, 0, SEEK_CUR) - (*p)->offset; + (*p)++; + } + return ret; +} + +static int perf_header__adds_write(struct perf_header *header, + struct perf_evlist *evlist, int fd) +{ + int nr_sections; + struct perf_file_section *feat_sec, *p; + int sec_size; + u64 sec_start; + int feat; + int err; + + nr_sections = bitmap_weight(header->adds_features, HEADER_FEAT_BITS); + if (!nr_sections) + return 0; + + feat_sec = p = calloc(nr_sections, sizeof(*feat_sec)); + if (feat_sec == NULL) + return -ENOMEM; + + sec_size = sizeof(*feat_sec) * nr_sections; + + sec_start = header->data_offset + header->data_size; + lseek(fd, sec_start + sec_size, SEEK_SET); + + for_each_set_bit(feat, header->adds_features, HEADER_FEAT_BITS) { + if (do_write_feat(fd, header, feat, &p, evlist)) + perf_header__clear_feat(header, feat); + } + + lseek(fd, sec_start, SEEK_SET); + /* + * may write more than needed due to dropped feature, but + * this is okay, reader will skip the mising entries + */ + err = do_write(fd, feat_sec, sec_size); + if (err < 0) + pr_debug("failed to write feature section\n"); + free(feat_sec); + return err; +} + +int perf_header__write_pipe(int fd) +{ + struct perf_pipe_file_header f_header; + int err; + + f_header = (struct perf_pipe_file_header){ + .magic = PERF_MAGIC, + .size = sizeof(f_header), + }; + + err = do_write(fd, &f_header, sizeof(f_header)); + if (err < 0) { + pr_debug("failed to write perf pipe header\n"); + return err; + } + + return 0; +} + +int perf_session__write_header(struct perf_session *session, + struct perf_evlist *evlist, + int fd, bool at_exit) +{ + struct perf_file_header f_header; + struct perf_file_attr f_attr; + struct perf_header *header = &session->header; + struct perf_evsel *evsel, *pair = NULL; + int err; + + lseek(fd, sizeof(f_header), SEEK_SET); + + if (session->evlist != evlist) + pair = perf_evlist__first(session->evlist); + + list_for_each_entry(evsel, &evlist->entries, node) { + evsel->id_offset = lseek(fd, 0, SEEK_CUR); + err = do_write(fd, evsel->id, evsel->ids * sizeof(u64)); + if (err < 0) { +out_err_write: + pr_debug("failed to write perf header\n"); + return err; + } + if (session->evlist != evlist) { + err = do_write(fd, pair->id, pair->ids * sizeof(u64)); + if (err < 0) + goto out_err_write; + evsel->ids += pair->ids; + pair = perf_evsel__next(pair); + } + } + + header->attr_offset = lseek(fd, 0, SEEK_CUR); + + list_for_each_entry(evsel, &evlist->entries, node) { + f_attr = (struct perf_file_attr){ + .attr = evsel->attr, + .ids = { + .offset = evsel->id_offset, + .size = evsel->ids * sizeof(u64), + } + }; + err = do_write(fd, &f_attr, sizeof(f_attr)); + if (err < 0) { + pr_debug("failed to write perf header attribute\n"); + return err; + } + } + + header->event_offset = lseek(fd, 0, SEEK_CUR); + header->event_size = trace_event_count * sizeof(struct perf_trace_event_type); + if (trace_events) { + err = do_write(fd, trace_events, header->event_size); + if (err < 0) { + pr_debug("failed to write perf header events\n"); + return err; + } + } + + header->data_offset = lseek(fd, 0, SEEK_CUR); + + if (at_exit) { + err = perf_header__adds_write(header, evlist, fd); + if (err < 0) + return err; + } + + f_header = (struct perf_file_header){ + .magic = PERF_MAGIC, + .size = sizeof(f_header), + .attr_size = sizeof(f_attr), + .attrs = { + .offset = header->attr_offset, + .size = evlist->nr_entries * sizeof(f_attr), + }, + .data = { + .offset = header->data_offset, + .size = header->data_size, + }, + .event_types = { + .offset = header->event_offset, + .size = header->event_size, + }, + }; + + memcpy(&f_header.adds_features, &header->adds_features, sizeof(header->adds_features)); + + lseek(fd, 0, SEEK_SET); + err = do_write(fd, &f_header, sizeof(f_header)); + if (err < 0) { + pr_debug("failed to write perf header\n"); + return err; + } + lseek(fd, header->data_offset + header->data_size, SEEK_SET); + + header->frozen = 1; + return 0; +} + +static int perf_header__getbuffer64(struct perf_header *header, + int fd, void *buf, size_t size) +{ + if (readn(fd, buf, size) <= 0) + return -1; + + if (header->needs_swap) + mem_bswap_64(buf, size); + + return 0; +} + +int perf_header__process_sections(struct perf_header *header, int fd, + void *data, + int (*process)(struct perf_file_section *section, + struct perf_header *ph, + int feat, int fd, void *data)) +{ + struct perf_file_section *feat_sec, *sec; + int nr_sections; + int sec_size; + int feat; + int err; + + nr_sections = bitmap_weight(header->adds_features, HEADER_FEAT_BITS); + if (!nr_sections) + return 0; + + feat_sec = sec = calloc(nr_sections, sizeof(*feat_sec)); + if (!feat_sec) + return -1; + + sec_size = sizeof(*feat_sec) * nr_sections; + + lseek(fd, header->data_offset + header->data_size, SEEK_SET); + + err = perf_header__getbuffer64(header, fd, feat_sec, sec_size); + if (err < 0) + goto out_free; + + for_each_set_bit(feat, header->adds_features, HEADER_LAST_FEATURE) { + err = process(sec++, header, feat, fd, data); + if (err < 0) + goto out_free; + } + err = 0; +out_free: + free(feat_sec); + return err; +} + +static const int attr_file_abi_sizes[] = { + [0] = PERF_ATTR_SIZE_VER0, + [1] = PERF_ATTR_SIZE_VER1, + [2] = PERF_ATTR_SIZE_VER2, + [3] = PERF_ATTR_SIZE_VER3, + 0, +}; + +/* + * In the legacy file format, the magic number is not used to encode endianness. + * hdr_sz was used to encode endianness. But given that hdr_sz can vary based + * on ABI revisions, we need to try all combinations for all endianness to + * detect the endianness. + */ +static int try_all_file_abis(uint64_t hdr_sz, struct perf_header *ph) +{ + uint64_t ref_size, attr_size; + int i; + + for (i = 0 ; attr_file_abi_sizes[i]; i++) { + ref_size = attr_file_abi_sizes[i] + + sizeof(struct perf_file_section); + if (hdr_sz != ref_size) { + attr_size = bswap_64(hdr_sz); + if (attr_size != ref_size) + continue; + + ph->needs_swap = true; + } + pr_debug("ABI%d perf.data file detected, need_swap=%d\n", + i, + ph->needs_swap); + return 0; + } + /* could not determine endianness */ + return -1; +} + +#define PERF_PIPE_HDR_VER0 16 + +static const size_t attr_pipe_abi_sizes[] = { + [0] = PERF_PIPE_HDR_VER0, + 0, +}; + +/* + * In the legacy pipe format, there is an implicit assumption that endiannesss + * between host recording the samples, and host parsing the samples is the + * same. This is not always the case given that the pipe output may always be + * redirected into a file and analyzed on a different machine with possibly a + * different endianness and perf_event ABI revsions in the perf tool itself. + */ +static int try_all_pipe_abis(uint64_t hdr_sz, struct perf_header *ph) +{ + u64 attr_size; + int i; + + for (i = 0 ; attr_pipe_abi_sizes[i]; i++) { + if (hdr_sz != attr_pipe_abi_sizes[i]) { + attr_size = bswap_64(hdr_sz); + if (attr_size != hdr_sz) + continue; + + ph->needs_swap = true; + } + pr_debug("Pipe ABI%d perf.data file detected\n", i); + return 0; + } + return -1; +} + +bool is_perf_magic(u64 magic) +{ + if (!memcmp(&magic, __perf_magic1, sizeof(magic)) + || magic == __perf_magic2 + || magic == __perf_magic2_sw) + return true; + + return false; +} + +static int check_magic_endian(u64 magic, uint64_t hdr_sz, + bool is_pipe, struct perf_header *ph) +{ + int ret; + + /* check for legacy format */ + ret = memcmp(&magic, __perf_magic1, sizeof(magic)); + if (ret == 0) { + pr_debug("legacy perf.data format\n"); + if (is_pipe) + return try_all_pipe_abis(hdr_sz, ph); + + return try_all_file_abis(hdr_sz, ph); + } + /* + * the new magic number serves two purposes: + * - unique number to identify actual perf.data files + * - encode endianness of file + */ + + /* check magic number with one endianness */ + if (magic == __perf_magic2) + return 0; + + /* check magic number with opposite endianness */ + if (magic != __perf_magic2_sw) + return -1; + + ph->needs_swap = true; + + return 0; +} + +int perf_file_header__read(struct perf_file_header *header, + struct perf_header *ph, int fd) +{ + int ret; + + lseek(fd, 0, SEEK_SET); + + ret = readn(fd, header, sizeof(*header)); + if (ret <= 0) + return -1; + + if (check_magic_endian(header->magic, + header->attr_size, false, ph) < 0) { + pr_debug("magic/endian check failed\n"); + return -1; + } + + if (ph->needs_swap) { + mem_bswap_64(header, offsetof(struct perf_file_header, + adds_features)); + } + + if (header->size != sizeof(*header)) { + /* Support the previous format */ + if (header->size == offsetof(typeof(*header), adds_features)) + bitmap_zero(header->adds_features, HEADER_FEAT_BITS); + else + return -1; + } else if (ph->needs_swap) { + /* + * feature bitmap is declared as an array of unsigned longs -- + * not good since its size can differ between the host that + * generated the data file and the host analyzing the file. + * + * We need to handle endianness, but we don't know the size of + * the unsigned long where the file was generated. Take a best + * guess at determining it: try 64-bit swap first (ie., file + * created on a 64-bit host), and check if the hostname feature + * bit is set (this feature bit is forced on as of fbe96f2). + * If the bit is not, undo the 64-bit swap and try a 32-bit + * swap. If the hostname bit is still not set (e.g., older data + * file), punt and fallback to the original behavior -- + * clearing all feature bits and setting buildid. + */ + mem_bswap_64(&header->adds_features, + BITS_TO_U64(HEADER_FEAT_BITS)); + + if (!test_bit(HEADER_HOSTNAME, header->adds_features)) { + /* unswap as u64 */ + mem_bswap_64(&header->adds_features, + BITS_TO_U64(HEADER_FEAT_BITS)); + + /* unswap as u32 */ + mem_bswap_32(&header->adds_features, + BITS_TO_U32(HEADER_FEAT_BITS)); + } + + if (!test_bit(HEADER_HOSTNAME, header->adds_features)) { + bitmap_zero(header->adds_features, HEADER_FEAT_BITS); + set_bit(HEADER_BUILD_ID, header->adds_features); + } + } + + memcpy(&ph->adds_features, &header->adds_features, + sizeof(ph->adds_features)); + + ph->event_offset = header->event_types.offset; + ph->event_size = header->event_types.size; + ph->data_offset = header->data.offset; + ph->data_size = header->data.size; + return 0; +} + +static int perf_file_section__process(struct perf_file_section *section, + struct perf_header *ph, + int feat, int fd, void *data) +{ + if (lseek(fd, section->offset, SEEK_SET) == (off_t)-1) { + pr_debug("Failed to lseek to %" PRIu64 " offset for feature " + "%d, continuing...\n", section->offset, feat); + return 0; + } + + if (feat >= HEADER_LAST_FEATURE) { + pr_debug("unknown feature %d, continuing...\n", feat); + return 0; + } + + if (!feat_ops[feat].process) + return 0; + + return feat_ops[feat].process(section, ph, fd, data); +} + +static int perf_file_header__read_pipe(struct perf_pipe_file_header *header, + struct perf_header *ph, int fd, + bool repipe) +{ + int ret; + + ret = readn(fd, header, sizeof(*header)); + if (ret <= 0) + return -1; + + if (check_magic_endian(header->magic, header->size, true, ph) < 0) { + pr_debug("endian/magic failed\n"); + return -1; + } + + if (ph->needs_swap) + header->size = bswap_64(header->size); + + if (repipe && do_write(STDOUT_FILENO, header, sizeof(*header)) < 0) + return -1; + + return 0; +} + +static int perf_header__read_pipe(struct perf_session *session, int fd) +{ + struct perf_header *header = &session->header; + struct perf_pipe_file_header f_header; + + if (perf_file_header__read_pipe(&f_header, header, fd, + session->repipe) < 0) { + pr_debug("incompatible file format\n"); + return -EINVAL; + } + + session->fd = fd; + + return 0; +} + +static int read_attr(int fd, struct perf_header *ph, + struct perf_file_attr *f_attr) +{ + struct perf_event_attr *attr = &f_attr->attr; + size_t sz, left; + size_t our_sz = sizeof(f_attr->attr); + int ret; + + memset(f_attr, 0, sizeof(*f_attr)); + + /* read minimal guaranteed structure */ + ret = readn(fd, attr, PERF_ATTR_SIZE_VER0); + if (ret <= 0) { + pr_debug("cannot read %d bytes of header attr\n", + PERF_ATTR_SIZE_VER0); + return -1; + } + + /* on file perf_event_attr size */ + sz = attr->size; + + if (ph->needs_swap) + sz = bswap_32(sz); + + if (sz == 0) { + /* assume ABI0 */ + sz = PERF_ATTR_SIZE_VER0; + } else if (sz > our_sz) { + pr_debug("file uses a more recent and unsupported ABI" + " (%zu bytes extra)\n", sz - our_sz); + return -1; + } + /* what we have not yet read and that we know about */ + left = sz - PERF_ATTR_SIZE_VER0; + if (left) { + void *ptr = attr; + ptr += PERF_ATTR_SIZE_VER0; + + ret = readn(fd, ptr, left); + } + /* read perf_file_section, ids are read in caller */ + ret = readn(fd, &f_attr->ids, sizeof(f_attr->ids)); + + return ret <= 0 ? -1 : 0; +} + +static int perf_evsel__prepare_tracepoint_event(struct perf_evsel *evsel, + struct pevent *pevent) +{ + struct event_format *event; + char bf[128]; + + /* already prepared */ + if (evsel->tp_format) + return 0; + + event = pevent_find_event(pevent, evsel->attr.config); + if (event == NULL) + return -1; + + if (!evsel->name) { + snprintf(bf, sizeof(bf), "%s:%s", event->system, event->name); + evsel->name = strdup(bf); + if (evsel->name == NULL) + return -1; + } + + evsel->tp_format = event; + return 0; +} + +static int perf_evlist__prepare_tracepoint_events(struct perf_evlist *evlist, + struct pevent *pevent) +{ + struct perf_evsel *pos; + + list_for_each_entry(pos, &evlist->entries, node) { + if (pos->attr.type == PERF_TYPE_TRACEPOINT && + perf_evsel__prepare_tracepoint_event(pos, pevent)) + return -1; + } + + return 0; +} + +int perf_session__read_header(struct perf_session *session, int fd) +{ + struct perf_header *header = &session->header; + struct perf_file_header f_header; + struct perf_file_attr f_attr; + u64 f_id; + int nr_attrs, nr_ids, i, j; + + session->evlist = perf_evlist__new(NULL, NULL); + if (session->evlist == NULL) + return -ENOMEM; + + if (session->fd_pipe) + return perf_header__read_pipe(session, fd); + + if (perf_file_header__read(&f_header, header, fd) < 0) + return -EINVAL; + + nr_attrs = f_header.attrs.size / f_header.attr_size; + lseek(fd, f_header.attrs.offset, SEEK_SET); + + for (i = 0; i < nr_attrs; i++) { + struct perf_evsel *evsel; + off_t tmp; + + if (read_attr(fd, header, &f_attr) < 0) + goto out_errno; + + if (header->needs_swap) + perf_event__attr_swap(&f_attr.attr); + + tmp = lseek(fd, 0, SEEK_CUR); + evsel = perf_evsel__new(&f_attr.attr, i); + + if (evsel == NULL) + goto out_delete_evlist; + + evsel->needs_swap = header->needs_swap; + /* + * Do it before so that if perf_evsel__alloc_id fails, this + * entry gets purged too at perf_evlist__delete(). + */ + perf_evlist__add(session->evlist, evsel); + + nr_ids = f_attr.ids.size / sizeof(u64); + /* + * We don't have the cpu and thread maps on the header, so + * for allocating the perf_sample_id table we fake 1 cpu and + * hattr->ids threads. + */ + if (perf_evsel__alloc_id(evsel, 1, nr_ids)) + goto out_delete_evlist; + + lseek(fd, f_attr.ids.offset, SEEK_SET); + + for (j = 0; j < nr_ids; j++) { + if (perf_header__getbuffer64(header, fd, &f_id, sizeof(f_id))) + goto out_errno; + + perf_evlist__id_add(session->evlist, evsel, 0, j, f_id); + } + + lseek(fd, tmp, SEEK_SET); + } + + symbol_conf.nr_events = nr_attrs; + + if (f_header.event_types.size) { + lseek(fd, f_header.event_types.offset, SEEK_SET); + trace_events = malloc(f_header.event_types.size); + if (trace_events == NULL) + return -ENOMEM; + if (perf_header__getbuffer64(header, fd, trace_events, + f_header.event_types.size)) + goto out_errno; + trace_event_count = f_header.event_types.size / sizeof(struct perf_trace_event_type); + } + + perf_header__process_sections(header, fd, &session->pevent, + perf_file_section__process); + + lseek(fd, header->data_offset, SEEK_SET); + + if (perf_evlist__prepare_tracepoint_events(session->evlist, + session->pevent)) + goto out_delete_evlist; + + header->frozen = 1; + return 0; +out_errno: + return -errno; + +out_delete_evlist: + perf_evlist__delete(session->evlist); + session->evlist = NULL; + return -ENOMEM; +} + +int perf_event__synthesize_attr(struct perf_tool *tool, + struct perf_event_attr *attr, u32 ids, u64 *id, + perf_event__handler_t process) +{ + union perf_event *ev; + size_t size; + int err; + + size = sizeof(struct perf_event_attr); + size = PERF_ALIGN(size, sizeof(u64)); + size += sizeof(struct perf_event_header); + size += ids * sizeof(u64); + + ev = malloc(size); + + if (ev == NULL) + return -ENOMEM; + + ev->attr.attr = *attr; + memcpy(ev->attr.id, id, ids * sizeof(u64)); + + ev->attr.header.type = PERF_RECORD_HEADER_ATTR; + ev->attr.header.size = (u16)size; + + if (ev->attr.header.size == size) + err = process(tool, ev, NULL, NULL); + else + err = -E2BIG; + + free(ev); + + return err; +} + +int perf_event__synthesize_attrs(struct perf_tool *tool, + struct perf_session *session, + perf_event__handler_t process) +{ + struct perf_evsel *evsel; + int err = 0; + + list_for_each_entry(evsel, &session->evlist->entries, node) { + err = perf_event__synthesize_attr(tool, &evsel->attr, evsel->ids, + evsel->id, process); + if (err) { + pr_debug("failed to create perf header attribute\n"); + return err; + } + } + + return err; +} + +int perf_event__process_attr(union perf_event *event, + struct perf_evlist **pevlist) +{ + u32 i, ids, n_ids; + struct perf_evsel *evsel; + struct perf_evlist *evlist = *pevlist; + + if (evlist == NULL) { + *pevlist = evlist = perf_evlist__new(NULL, NULL); + if (evlist == NULL) + return -ENOMEM; + } + + evsel = perf_evsel__new(&event->attr.attr, evlist->nr_entries); + if (evsel == NULL) + return -ENOMEM; + + perf_evlist__add(evlist, evsel); + + ids = event->header.size; + ids -= (void *)&event->attr.id - (void *)event; + n_ids = ids / sizeof(u64); + /* + * We don't have the cpu and thread maps on the header, so + * for allocating the perf_sample_id table we fake 1 cpu and + * hattr->ids threads. + */ + if (perf_evsel__alloc_id(evsel, 1, n_ids)) + return -ENOMEM; + + for (i = 0; i < n_ids; i++) { + perf_evlist__id_add(evlist, evsel, 0, i, event->attr.id[i]); + } + + return 0; +} + +int perf_event__synthesize_event_type(struct perf_tool *tool, + u64 event_id, char *name, + perf_event__handler_t process, + struct machine *machine) +{ + union perf_event ev; + size_t size = 0; + int err = 0; + + memset(&ev, 0, sizeof(ev)); + + ev.event_type.event_type.event_id = event_id; + memset(ev.event_type.event_type.name, 0, MAX_EVENT_NAME); + strncpy(ev.event_type.event_type.name, name, MAX_EVENT_NAME - 1); + + ev.event_type.header.type = PERF_RECORD_HEADER_EVENT_TYPE; + size = strlen(ev.event_type.event_type.name); + size = PERF_ALIGN(size, sizeof(u64)); + ev.event_type.header.size = sizeof(ev.event_type) - + (sizeof(ev.event_type.event_type.name) - size); + + err = process(tool, &ev, NULL, machine); + + return err; +} + +int perf_event__synthesize_event_types(struct perf_tool *tool, + perf_event__handler_t process, + struct machine *machine) +{ + struct perf_trace_event_type *type; + int i, err = 0; + + for (i = 0; i < trace_event_count; i++) { + type = &trace_events[i]; + + err = perf_event__synthesize_event_type(tool, type->event_id, + type->name, process, + machine); + if (err) { + pr_debug("failed to create perf header event type\n"); + return err; + } + } + + return err; +} + +int perf_event__process_event_type(struct perf_tool *tool __maybe_unused, + union perf_event *event) +{ + if (perf_header__push_event(event->event_type.event_type.event_id, + event->event_type.event_type.name) < 0) + return -ENOMEM; + + return 0; +} + +int perf_event__synthesize_tracing_data(struct perf_tool *tool, int fd, + struct perf_evlist *evlist, + perf_event__handler_t process) +{ + union perf_event ev; + struct tracing_data *tdata; + ssize_t size = 0, aligned_size = 0, padding; + int err __maybe_unused = 0; + + /* + * We are going to store the size of the data followed + * by the data contents. Since the fd descriptor is a pipe, + * we cannot seek back to store the size of the data once + * we know it. Instead we: + * + * - write the tracing data to the temp file + * - get/write the data size to pipe + * - write the tracing data from the temp file + * to the pipe + */ + tdata = tracing_data_get(&evlist->entries, fd, true); + if (!tdata) + return -1; + + memset(&ev, 0, sizeof(ev)); + + ev.tracing_data.header.type = PERF_RECORD_HEADER_TRACING_DATA; + size = tdata->size; + aligned_size = PERF_ALIGN(size, sizeof(u64)); + padding = aligned_size - size; + ev.tracing_data.header.size = sizeof(ev.tracing_data); + ev.tracing_data.size = aligned_size; + + process(tool, &ev, NULL, NULL); + + /* + * The put function will copy all the tracing data + * stored in temp file to the pipe. + */ + tracing_data_put(tdata); + + write_padded(fd, NULL, 0, padding); + + return aligned_size; +} + +int perf_event__process_tracing_data(union perf_event *event, + struct perf_session *session) +{ + ssize_t size_read, padding, size = event->tracing_data.size; + off_t offset = lseek(session->fd, 0, SEEK_CUR); + char buf[BUFSIZ]; + + /* setup for reading amidst mmap */ + lseek(session->fd, offset + sizeof(struct tracing_data_event), + SEEK_SET); + + size_read = trace_report(session->fd, &session->pevent, + session->repipe); + padding = PERF_ALIGN(size_read, sizeof(u64)) - size_read; + + if (readn(session->fd, buf, padding) < 0) { + pr_err("%s: reading input file", __func__); + return -1; + } + if (session->repipe) { + int retw = write(STDOUT_FILENO, buf, padding); + if (retw <= 0 || retw != padding) { + pr_err("%s: repiping tracing data padding", __func__); + return -1; + } + } + + if (size_read + padding != size) { + pr_err("%s: tracing data size mismatch", __func__); + return -1; + } + + perf_evlist__prepare_tracepoint_events(session->evlist, + session->pevent); + + return size_read + padding; +} + +int perf_event__synthesize_build_id(struct perf_tool *tool, + struct dso *pos, u16 misc, + perf_event__handler_t process, + struct machine *machine) +{ + union perf_event ev; + size_t len; + int err = 0; + + if (!pos->hit) + return err; + + memset(&ev, 0, sizeof(ev)); + + len = pos->long_name_len + 1; + len = PERF_ALIGN(len, NAME_ALIGN); + memcpy(&ev.build_id.build_id, pos->build_id, sizeof(pos->build_id)); + ev.build_id.header.type = PERF_RECORD_HEADER_BUILD_ID; + ev.build_id.header.misc = misc; + ev.build_id.pid = machine->pid; + ev.build_id.header.size = sizeof(ev.build_id) + len; + memcpy(&ev.build_id.filename, pos->long_name, pos->long_name_len); + + err = process(tool, &ev, NULL, machine); + + return err; +} + +int perf_event__process_build_id(struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_session *session) +{ + __event_process_build_id(&event->build_id, + event->build_id.filename, + session); + return 0; +} + +void disable_buildid_cache(void) +{ + no_buildid_cache = true; +} diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h new file mode 100644 index 00000000..c9fc55ca --- /dev/null +++ b/tools/perf/util/header.h @@ -0,0 +1,166 @@ +#ifndef __PERF_HEADER_H +#define __PERF_HEADER_H + +#include <linux/perf_event.h> +#include <sys/types.h> +#include <stdbool.h> +#include "types.h" +#include "event.h" + +#include <linux/bitmap.h> + +enum { + HEADER_RESERVED = 0, /* always cleared */ + HEADER_FIRST_FEATURE = 1, + HEADER_TRACING_DATA = 1, + HEADER_BUILD_ID, + + HEADER_HOSTNAME, + HEADER_OSRELEASE, + HEADER_VERSION, + HEADER_ARCH, + HEADER_NRCPUS, + HEADER_CPUDESC, + HEADER_CPUID, + HEADER_TOTAL_MEM, + HEADER_CMDLINE, + HEADER_EVENT_DESC, + HEADER_CPU_TOPOLOGY, + HEADER_NUMA_TOPOLOGY, + HEADER_BRANCH_STACK, + HEADER_PMU_MAPPINGS, + HEADER_GROUP_DESC, + HEADER_LAST_FEATURE, + HEADER_FEAT_BITS = 256, +}; + +struct perf_file_section { + u64 offset; + u64 size; +}; + +struct perf_file_header { + u64 magic; + u64 size; + u64 attr_size; + struct perf_file_section attrs; + struct perf_file_section data; + struct perf_file_section event_types; + DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS); +}; + +struct perf_pipe_file_header { + u64 magic; + u64 size; +}; + +struct perf_header; + +int perf_file_header__read(struct perf_file_header *header, + struct perf_header *ph, int fd); + +struct perf_session_env { + char *hostname; + char *os_release; + char *version; + char *arch; + int nr_cpus_online; + int nr_cpus_avail; + char *cpu_desc; + char *cpuid; + unsigned long long total_mem; + + int nr_cmdline; + char *cmdline; + int nr_sibling_cores; + char *sibling_cores; + int nr_sibling_threads; + char *sibling_threads; + int nr_numa_nodes; + char *numa_nodes; + int nr_pmu_mappings; + char *pmu_mappings; + int nr_groups; +}; + +struct perf_header { + int frozen; + bool needs_swap; + s64 attr_offset; + u64 data_offset; + u64 data_size; + u64 event_offset; + u64 event_size; + DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS); + struct perf_session_env env; +}; + +struct perf_evlist; +struct perf_session; + +int perf_session__read_header(struct perf_session *session, int fd); +int perf_session__write_header(struct perf_session *session, + struct perf_evlist *evlist, + int fd, bool at_exit); +int perf_header__write_pipe(int fd); + +int perf_header__push_event(u64 id, const char *name); +char *perf_header__find_event(u64 id); + +void perf_header__set_feat(struct perf_header *header, int feat); +void perf_header__clear_feat(struct perf_header *header, int feat); +bool perf_header__has_feat(const struct perf_header *header, int feat); + +int perf_header__set_cmdline(int argc, const char **argv); + +int perf_header__process_sections(struct perf_header *header, int fd, + void *data, + int (*process)(struct perf_file_section *section, + struct perf_header *ph, + int feat, int fd, void *data)); + +int perf_header__fprintf_info(struct perf_session *s, FILE *fp, bool full); + +int build_id_cache__add_s(const char *sbuild_id, const char *debugdir, + const char *name, bool is_kallsyms, bool is_vdso); +int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir); + +int perf_event__synthesize_attr(struct perf_tool *tool, + struct perf_event_attr *attr, u32 ids, u64 *id, + perf_event__handler_t process); +int perf_event__synthesize_attrs(struct perf_tool *tool, + struct perf_session *session, + perf_event__handler_t process); +int perf_event__process_attr(union perf_event *event, struct perf_evlist **pevlist); + +int perf_event__synthesize_event_type(struct perf_tool *tool, + u64 event_id, char *name, + perf_event__handler_t process, + struct machine *machine); +int perf_event__synthesize_event_types(struct perf_tool *tool, + perf_event__handler_t process, + struct machine *machine); +int perf_event__process_event_type(struct perf_tool *tool, + union perf_event *event); + +int perf_event__synthesize_tracing_data(struct perf_tool *tool, + int fd, struct perf_evlist *evlist, + perf_event__handler_t process); +int perf_event__process_tracing_data(union perf_event *event, + struct perf_session *session); + +int perf_event__synthesize_build_id(struct perf_tool *tool, + struct dso *pos, u16 misc, + perf_event__handler_t process, + struct machine *machine); +int perf_event__process_build_id(struct perf_tool *tool, + union perf_event *event, + struct perf_session *session); +bool is_perf_magic(u64 magic); + +/* + * arch specific callback + */ +int get_cpuid(char *buffer, size_t sz); + +#endif /* __PERF_HEADER_H */ diff --git a/tools/perf/util/help.c b/tools/perf/util/help.c new file mode 100644 index 00000000..8b1f6e89 --- /dev/null +++ b/tools/perf/util/help.c @@ -0,0 +1,340 @@ +#include "cache.h" +#include "../builtin.h" +#include "exec_cmd.h" +#include "levenshtein.h" +#include "help.h" +#include <termios.h> + +void add_cmdname(struct cmdnames *cmds, const char *name, size_t 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) +{ + unsigned 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) +{ + unsigned 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) +{ + size_t 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 */ + struct winsize win; + int max_cols; + int i, j; + + get_term_dimensions(&win); + max_cols = win.ws_col - 1; /* don't print *on* the edge */ + + if (space < max_cols) + cols = max_cols / space; + rows = (cmds->cnt + cols - 1) / cols; + + for (i = 0; i < rows; i++) { + printf(" "); + + for (j = 0; j < cols; j++) { + unsigned int n = j * rows + i; + unsigned 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; + + 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) +{ + unsigned 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) +{ + unsigned 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) +{ + unsigned 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) +{ + unsigned 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 perf 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 __maybe_unused, const char **argv __maybe_unused, + const char *prefix __maybe_unused) +{ + 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 00000000..7f5c6ded --- /dev/null +++ b/tools/perf/util/help.h @@ -0,0 +1,29 @@ +#ifndef __PERF_HELP_H +#define __PERF_HELP_H + +struct cmdnames { + size_t alloc; + size_t 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, size_t len); +/* Here we require that excludes is a sorted list. */ +void exclude_cmds(struct cmdnames *cmds, struct cmdnames *excludes); +int is_in_cmdlist(struct cmdnames *c, const char *s); +void list_commands(const char *title, struct cmdnames *main_cmds, + struct cmdnames *other_cmds); + +#endif /* __PERF_HELP_H */ diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c new file mode 100644 index 00000000..f855941b --- /dev/null +++ b/tools/perf/util/hist.c @@ -0,0 +1,918 @@ +#include "annotate.h" +#include "util.h" +#include "build-id.h" +#include "hist.h" +#include "session.h" +#include "sort.h" +#include "evsel.h" +#include <math.h> + +static bool hists__filter_entry_by_dso(struct hists *hists, + struct hist_entry *he); +static bool hists__filter_entry_by_thread(struct hists *hists, + struct hist_entry *he); +static bool hists__filter_entry_by_symbol(struct hists *hists, + struct hist_entry *he); + +enum hist_filter { + HIST_FILTER__DSO, + HIST_FILTER__THREAD, + HIST_FILTER__PARENT, + HIST_FILTER__SYMBOL, +}; + +struct callchain_param callchain_param = { + .mode = CHAIN_GRAPH_REL, + .min_percent = 0.5, + .order = ORDER_CALLEE +}; + +u16 hists__col_len(struct hists *hists, enum hist_column col) +{ + return hists->col_len[col]; +} + +void hists__set_col_len(struct hists *hists, enum hist_column col, u16 len) +{ + hists->col_len[col] = len; +} + +bool hists__new_col_len(struct hists *hists, enum hist_column col, u16 len) +{ + if (len > hists__col_len(hists, col)) { + hists__set_col_len(hists, col, len); + return true; + } + return false; +} + +void hists__reset_col_len(struct hists *hists) +{ + enum hist_column col; + + for (col = 0; col < HISTC_NR_COLS; ++col) + hists__set_col_len(hists, col, 0); +} + +static void hists__set_unres_dso_col_len(struct hists *hists, int dso) +{ + const unsigned int unresolved_col_width = BITS_PER_LONG / 4; + + if (hists__col_len(hists, dso) < unresolved_col_width && + !symbol_conf.col_width_list_str && !symbol_conf.field_sep && + !symbol_conf.dso_list) + hists__set_col_len(hists, dso, unresolved_col_width); +} + +void hists__calc_col_len(struct hists *hists, struct hist_entry *h) +{ + const unsigned int unresolved_col_width = BITS_PER_LONG / 4; + u16 len; + + if (h->ms.sym) + hists__new_col_len(hists, HISTC_SYMBOL, h->ms.sym->namelen + 4); + else + hists__set_unres_dso_col_len(hists, HISTC_DSO); + + len = thread__comm_len(h->thread); + if (hists__new_col_len(hists, HISTC_COMM, len)) + hists__set_col_len(hists, HISTC_THREAD, len + 6); + + if (h->ms.map) { + len = dso__name_len(h->ms.map->dso); + hists__new_col_len(hists, HISTC_DSO, len); + } + + if (h->parent) + hists__new_col_len(hists, HISTC_PARENT, h->parent->namelen); + + if (h->branch_info) { + int symlen; + /* + * +4 accounts for '[x] ' priv level info + * +2 account of 0x prefix on raw addresses + */ + if (h->branch_info->from.sym) { + symlen = (int)h->branch_info->from.sym->namelen + 4; + hists__new_col_len(hists, HISTC_SYMBOL_FROM, symlen); + + symlen = dso__name_len(h->branch_info->from.map->dso); + hists__new_col_len(hists, HISTC_DSO_FROM, symlen); + } else { + symlen = unresolved_col_width + 4 + 2; + hists__new_col_len(hists, HISTC_SYMBOL_FROM, symlen); + hists__set_unres_dso_col_len(hists, HISTC_DSO_FROM); + } + + if (h->branch_info->to.sym) { + symlen = (int)h->branch_info->to.sym->namelen + 4; + hists__new_col_len(hists, HISTC_SYMBOL_TO, symlen); + + symlen = dso__name_len(h->branch_info->to.map->dso); + hists__new_col_len(hists, HISTC_DSO_TO, symlen); + } else { + symlen = unresolved_col_width + 4 + 2; + hists__new_col_len(hists, HISTC_SYMBOL_TO, symlen); + hists__set_unres_dso_col_len(hists, HISTC_DSO_TO); + } + } +} + +void hists__output_recalc_col_len(struct hists *hists, int max_rows) +{ + struct rb_node *next = rb_first(&hists->entries); + struct hist_entry *n; + int row = 0; + + hists__reset_col_len(hists); + + while (next && row++ < max_rows) { + n = rb_entry(next, struct hist_entry, rb_node); + if (!n->filtered) + hists__calc_col_len(hists, n); + next = rb_next(&n->rb_node); + } +} + +static void hist_entry__add_cpumode_period(struct hist_entry *he, + unsigned int cpumode, u64 period) +{ + switch (cpumode) { + case PERF_RECORD_MISC_KERNEL: + he->stat.period_sys += period; + break; + case PERF_RECORD_MISC_USER: + he->stat.period_us += period; + break; + case PERF_RECORD_MISC_GUEST_KERNEL: + he->stat.period_guest_sys += period; + break; + case PERF_RECORD_MISC_GUEST_USER: + he->stat.period_guest_us += period; + break; + default: + break; + } +} + +static void he_stat__add_period(struct he_stat *he_stat, u64 period) +{ + he_stat->period += period; + he_stat->nr_events += 1; +} + +static void he_stat__add_stat(struct he_stat *dest, struct he_stat *src) +{ + dest->period += src->period; + dest->period_sys += src->period_sys; + dest->period_us += src->period_us; + dest->period_guest_sys += src->period_guest_sys; + dest->period_guest_us += src->period_guest_us; + dest->nr_events += src->nr_events; +} + +static void hist_entry__decay(struct hist_entry *he) +{ + he->stat.period = (he->stat.period * 7) / 8; + he->stat.nr_events = (he->stat.nr_events * 7) / 8; +} + +static bool hists__decay_entry(struct hists *hists, struct hist_entry *he) +{ + u64 prev_period = he->stat.period; + + if (prev_period == 0) + return true; + + hist_entry__decay(he); + + if (!he->filtered) + hists->stats.total_period -= prev_period - he->stat.period; + + return he->stat.period == 0; +} + +static void __hists__decay_entries(struct hists *hists, bool zap_user, + bool zap_kernel, bool threaded) +{ + struct rb_node *next = rb_first(&hists->entries); + struct hist_entry *n; + + while (next) { + n = rb_entry(next, struct hist_entry, rb_node); + next = rb_next(&n->rb_node); + /* + * We may be annotating this, for instance, so keep it here in + * case some it gets new samples, we'll eventually free it when + * the user stops browsing and it agains gets fully decayed. + */ + if (((zap_user && n->level == '.') || + (zap_kernel && n->level != '.') || + hists__decay_entry(hists, n)) && + !n->used) { + rb_erase(&n->rb_node, &hists->entries); + + if (sort__need_collapse || threaded) + rb_erase(&n->rb_node_in, &hists->entries_collapsed); + + hist_entry__free(n); + --hists->nr_entries; + } + } +} + +void hists__decay_entries(struct hists *hists, bool zap_user, bool zap_kernel) +{ + return __hists__decay_entries(hists, zap_user, zap_kernel, false); +} + +void hists__decay_entries_threaded(struct hists *hists, + bool zap_user, bool zap_kernel) +{ + return __hists__decay_entries(hists, zap_user, zap_kernel, true); +} + +/* + * histogram, sorted on item, collects periods + */ + +static struct hist_entry *hist_entry__new(struct hist_entry *template) +{ + size_t callchain_size = symbol_conf.use_callchain ? sizeof(struct callchain_root) : 0; + struct hist_entry *he = malloc(sizeof(*he) + callchain_size); + + if (he != NULL) { + *he = *template; + + if (he->ms.map) + he->ms.map->referenced = true; + + if (he->branch_info) { + if (he->branch_info->from.map) + he->branch_info->from.map->referenced = true; + if (he->branch_info->to.map) + he->branch_info->to.map->referenced = true; + } + + if (symbol_conf.use_callchain) + callchain_init(he->callchain); + + INIT_LIST_HEAD(&he->pairs.node); + } + + return he; +} + +void hists__inc_nr_entries(struct hists *hists, struct hist_entry *h) +{ + if (!h->filtered) { + hists__calc_col_len(hists, h); + ++hists->nr_entries; + hists->stats.total_period += h->stat.period; + } +} + +static u8 symbol__parent_filter(const struct symbol *parent) +{ + if (symbol_conf.exclude_other && parent == NULL) + return 1 << HIST_FILTER__PARENT; + return 0; +} + +static struct hist_entry *add_hist_entry(struct hists *hists, + struct hist_entry *entry, + struct addr_location *al, + u64 period) +{ + struct rb_node **p; + struct rb_node *parent = NULL; + struct hist_entry *he; + int cmp; + + pthread_mutex_lock(&hists->lock); + + p = &hists->entries_in->rb_node; + + while (*p != NULL) { + parent = *p; + he = rb_entry(parent, struct hist_entry, rb_node_in); + + /* + * Make sure that it receives arguments in a same order as + * hist_entry__collapse() so that we can use an appropriate + * function when searching an entry regardless which sort + * keys were used. + */ + cmp = hist_entry__cmp(he, entry); + + if (!cmp) { + he_stat__add_period(&he->stat, period); + + /* If the map of an existing hist_entry has + * become out-of-date due to an exec() or + * similar, update it. Otherwise we will + * mis-adjust symbol addresses when computing + * the history counter to increment. + */ + if (he->ms.map != entry->ms.map) { + he->ms.map = entry->ms.map; + if (he->ms.map) + he->ms.map->referenced = true; + } + goto out; + } + + if (cmp < 0) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + he = hist_entry__new(entry); + if (!he) + goto out_unlock; + + rb_link_node(&he->rb_node_in, parent, p); + rb_insert_color(&he->rb_node_in, hists->entries_in); +out: + hist_entry__add_cpumode_period(he, al->cpumode, period); +out_unlock: + pthread_mutex_unlock(&hists->lock); + return he; +} + +struct hist_entry *__hists__add_branch_entry(struct hists *self, + struct addr_location *al, + struct symbol *sym_parent, + struct branch_info *bi, + u64 period) +{ + struct hist_entry entry = { + .thread = al->thread, + .ms = { + .map = bi->to.map, + .sym = bi->to.sym, + }, + .cpu = al->cpu, + .ip = bi->to.addr, + .level = al->level, + .stat = { + .period = period, + .nr_events = 1, + }, + .parent = sym_parent, + .filtered = symbol__parent_filter(sym_parent), + .branch_info = bi, + .hists = self, + }; + + return add_hist_entry(self, &entry, al, period); +} + +struct hist_entry *__hists__add_entry(struct hists *self, + struct addr_location *al, + struct symbol *sym_parent, u64 period) +{ + struct hist_entry entry = { + .thread = al->thread, + .ms = { + .map = al->map, + .sym = al->sym, + }, + .cpu = al->cpu, + .ip = al->addr, + .level = al->level, + .stat = { + .period = period, + .nr_events = 1, + }, + .parent = sym_parent, + .filtered = symbol__parent_filter(sym_parent), + .hists = self, + }; + + return add_hist_entry(self, &entry, al, period); +} + +int64_t +hist_entry__cmp(struct hist_entry *left, struct hist_entry *right) +{ + struct sort_entry *se; + int64_t cmp = 0; + + list_for_each_entry(se, &hist_entry__sort_list, list) { + cmp = se->se_cmp(left, right); + if (cmp) + break; + } + + return cmp; +} + +int64_t +hist_entry__collapse(struct hist_entry *left, struct hist_entry *right) +{ + struct sort_entry *se; + int64_t cmp = 0; + + list_for_each_entry(se, &hist_entry__sort_list, list) { + int64_t (*f)(struct hist_entry *, struct hist_entry *); + + f = se->se_collapse ?: se->se_cmp; + + cmp = f(left, right); + if (cmp) + break; + } + + return cmp; +} + +void hist_entry__free(struct hist_entry *he) +{ + free(he->branch_info); + free(he); +} + +/* + * collapse the histogram + */ + +static bool hists__collapse_insert_entry(struct hists *hists __maybe_unused, + struct rb_root *root, + struct hist_entry *he) +{ + struct rb_node **p = &root->rb_node; + struct rb_node *parent = NULL; + struct hist_entry *iter; + int64_t cmp; + + while (*p != NULL) { + parent = *p; + iter = rb_entry(parent, struct hist_entry, rb_node_in); + + cmp = hist_entry__collapse(iter, he); + + if (!cmp) { + he_stat__add_stat(&iter->stat, &he->stat); + + if (symbol_conf.use_callchain) { + callchain_cursor_reset(&callchain_cursor); + callchain_merge(&callchain_cursor, + iter->callchain, + he->callchain); + } + hist_entry__free(he); + return false; + } + + if (cmp < 0) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + rb_link_node(&he->rb_node_in, parent, p); + rb_insert_color(&he->rb_node_in, root); + return true; +} + +static struct rb_root *hists__get_rotate_entries_in(struct hists *hists) +{ + struct rb_root *root; + + pthread_mutex_lock(&hists->lock); + + root = hists->entries_in; + if (++hists->entries_in > &hists->entries_in_array[1]) + hists->entries_in = &hists->entries_in_array[0]; + + pthread_mutex_unlock(&hists->lock); + + return root; +} + +static void hists__apply_filters(struct hists *hists, struct hist_entry *he) +{ + hists__filter_entry_by_dso(hists, he); + hists__filter_entry_by_thread(hists, he); + hists__filter_entry_by_symbol(hists, he); +} + +static void __hists__collapse_resort(struct hists *hists, bool threaded) +{ + struct rb_root *root; + struct rb_node *next; + struct hist_entry *n; + + if (!sort__need_collapse && !threaded) + return; + + root = hists__get_rotate_entries_in(hists); + next = rb_first(root); + + while (next) { + n = rb_entry(next, struct hist_entry, rb_node_in); + next = rb_next(&n->rb_node_in); + + rb_erase(&n->rb_node_in, root); + if (hists__collapse_insert_entry(hists, &hists->entries_collapsed, n)) { + /* + * If it wasn't combined with one of the entries already + * collapsed, we need to apply the filters that may have + * been set by, say, the hist_browser. + */ + hists__apply_filters(hists, n); + } + } +} + +void hists__collapse_resort(struct hists *hists) +{ + return __hists__collapse_resort(hists, false); +} + +void hists__collapse_resort_threaded(struct hists *hists) +{ + return __hists__collapse_resort(hists, true); +} + +/* + * reverse the map, sort on period. + */ + +static int period_cmp(u64 period_a, u64 period_b) +{ + if (period_a > period_b) + return 1; + if (period_a < period_b) + return -1; + return 0; +} + +static int hist_entry__sort_on_period(struct hist_entry *a, + struct hist_entry *b) +{ + int ret; + int i, nr_members; + struct perf_evsel *evsel; + struct hist_entry *pair; + u64 *periods_a, *periods_b; + + ret = period_cmp(a->stat.period, b->stat.period); + if (ret || !symbol_conf.event_group) + return ret; + + evsel = hists_to_evsel(a->hists); + nr_members = evsel->nr_members; + if (nr_members <= 1) + return ret; + + periods_a = zalloc(sizeof(periods_a) * nr_members); + periods_b = zalloc(sizeof(periods_b) * nr_members); + + if (!periods_a || !periods_b) + goto out; + + list_for_each_entry(pair, &a->pairs.head, pairs.node) { + evsel = hists_to_evsel(pair->hists); + periods_a[perf_evsel__group_idx(evsel)] = pair->stat.period; + } + + list_for_each_entry(pair, &b->pairs.head, pairs.node) { + evsel = hists_to_evsel(pair->hists); + periods_b[perf_evsel__group_idx(evsel)] = pair->stat.period; + } + + for (i = 1; i < nr_members; i++) { + ret = period_cmp(periods_a[i], periods_b[i]); + if (ret) + break; + } + +out: + free(periods_a); + free(periods_b); + + return ret; +} + +static void __hists__insert_output_entry(struct rb_root *entries, + struct hist_entry *he, + u64 min_callchain_hits) +{ + struct rb_node **p = &entries->rb_node; + struct rb_node *parent = NULL; + struct hist_entry *iter; + + if (symbol_conf.use_callchain) + callchain_param.sort(&he->sorted_chain, he->callchain, + min_callchain_hits, &callchain_param); + + while (*p != NULL) { + parent = *p; + iter = rb_entry(parent, struct hist_entry, rb_node); + + if (hist_entry__sort_on_period(he, iter) > 0) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + rb_link_node(&he->rb_node, parent, p); + rb_insert_color(&he->rb_node, entries); +} + +static void __hists__output_resort(struct hists *hists, bool threaded) +{ + struct rb_root *root; + struct rb_node *next; + struct hist_entry *n; + u64 min_callchain_hits; + + min_callchain_hits = hists->stats.total_period * (callchain_param.min_percent / 100); + + if (sort__need_collapse || threaded) + root = &hists->entries_collapsed; + else + root = hists->entries_in; + + next = rb_first(root); + hists->entries = RB_ROOT; + + hists->nr_entries = 0; + hists->stats.total_period = 0; + hists__reset_col_len(hists); + + while (next) { + n = rb_entry(next, struct hist_entry, rb_node_in); + next = rb_next(&n->rb_node_in); + + __hists__insert_output_entry(&hists->entries, n, min_callchain_hits); + hists__inc_nr_entries(hists, n); + } +} + +void hists__output_resort(struct hists *hists) +{ + return __hists__output_resort(hists, false); +} + +void hists__output_resort_threaded(struct hists *hists) +{ + return __hists__output_resort(hists, true); +} + +static void hists__remove_entry_filter(struct hists *hists, struct hist_entry *h, + enum hist_filter filter) +{ + h->filtered &= ~(1 << filter); + if (h->filtered) + return; + + ++hists->nr_entries; + if (h->ms.unfolded) + hists->nr_entries += h->nr_rows; + h->row_offset = 0; + hists->stats.total_period += h->stat.period; + hists->stats.nr_events[PERF_RECORD_SAMPLE] += h->stat.nr_events; + + hists__calc_col_len(hists, h); +} + + +static bool hists__filter_entry_by_dso(struct hists *hists, + struct hist_entry *he) +{ + if (hists->dso_filter != NULL && + (he->ms.map == NULL || he->ms.map->dso != hists->dso_filter)) { + he->filtered |= (1 << HIST_FILTER__DSO); + return true; + } + + return false; +} + +void hists__filter_by_dso(struct hists *hists) +{ + struct rb_node *nd; + + hists->nr_entries = hists->stats.total_period = 0; + hists->stats.nr_events[PERF_RECORD_SAMPLE] = 0; + hists__reset_col_len(hists); + + for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + + if (symbol_conf.exclude_other && !h->parent) + continue; + + if (hists__filter_entry_by_dso(hists, h)) + continue; + + hists__remove_entry_filter(hists, h, HIST_FILTER__DSO); + } +} + +static bool hists__filter_entry_by_thread(struct hists *hists, + struct hist_entry *he) +{ + if (hists->thread_filter != NULL && + he->thread != hists->thread_filter) { + he->filtered |= (1 << HIST_FILTER__THREAD); + return true; + } + + return false; +} + +void hists__filter_by_thread(struct hists *hists) +{ + struct rb_node *nd; + + hists->nr_entries = hists->stats.total_period = 0; + hists->stats.nr_events[PERF_RECORD_SAMPLE] = 0; + hists__reset_col_len(hists); + + for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + + if (hists__filter_entry_by_thread(hists, h)) + continue; + + hists__remove_entry_filter(hists, h, HIST_FILTER__THREAD); + } +} + +static bool hists__filter_entry_by_symbol(struct hists *hists, + struct hist_entry *he) +{ + if (hists->symbol_filter_str != NULL && + (!he->ms.sym || strstr(he->ms.sym->name, + hists->symbol_filter_str) == NULL)) { + he->filtered |= (1 << HIST_FILTER__SYMBOL); + return true; + } + + return false; +} + +void hists__filter_by_symbol(struct hists *hists) +{ + struct rb_node *nd; + + hists->nr_entries = hists->stats.total_period = 0; + hists->stats.nr_events[PERF_RECORD_SAMPLE] = 0; + hists__reset_col_len(hists); + + for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) { + struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node); + + if (hists__filter_entry_by_symbol(hists, h)) + continue; + + hists__remove_entry_filter(hists, h, HIST_FILTER__SYMBOL); + } +} + +int hist_entry__inc_addr_samples(struct hist_entry *he, int evidx, u64 ip) +{ + return symbol__inc_addr_samples(he->ms.sym, he->ms.map, evidx, ip); +} + +int hist_entry__annotate(struct hist_entry *he, size_t privsize) +{ + return symbol__annotate(he->ms.sym, he->ms.map, privsize); +} + +void events_stats__inc(struct events_stats *stats, u32 type) +{ + ++stats->nr_events[0]; + ++stats->nr_events[type]; +} + +void hists__inc_nr_events(struct hists *hists, u32 type) +{ + events_stats__inc(&hists->stats, type); +} + +static struct hist_entry *hists__add_dummy_entry(struct hists *hists, + struct hist_entry *pair) +{ + struct rb_root *root; + struct rb_node **p; + struct rb_node *parent = NULL; + struct hist_entry *he; + int cmp; + + if (sort__need_collapse) + root = &hists->entries_collapsed; + else + root = hists->entries_in; + + p = &root->rb_node; + + while (*p != NULL) { + parent = *p; + he = rb_entry(parent, struct hist_entry, rb_node_in); + + cmp = hist_entry__collapse(he, pair); + + if (!cmp) + goto out; + + if (cmp < 0) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + he = hist_entry__new(pair); + if (he) { + memset(&he->stat, 0, sizeof(he->stat)); + he->hists = hists; + rb_link_node(&he->rb_node_in, parent, p); + rb_insert_color(&he->rb_node_in, root); + hists__inc_nr_entries(hists, he); + } +out: + return he; +} + +static struct hist_entry *hists__find_entry(struct hists *hists, + struct hist_entry *he) +{ + struct rb_node *n; + + if (sort__need_collapse) + n = hists->entries_collapsed.rb_node; + else + n = hists->entries_in->rb_node; + + while (n) { + struct hist_entry *iter = rb_entry(n, struct hist_entry, rb_node_in); + int64_t cmp = hist_entry__collapse(iter, he); + + if (cmp < 0) + n = n->rb_left; + else if (cmp > 0) + n = n->rb_right; + else + return iter; + } + + return NULL; +} + +/* + * Look for pairs to link to the leader buckets (hist_entries): + */ +void hists__match(struct hists *leader, struct hists *other) +{ + struct rb_root *root; + struct rb_node *nd; + struct hist_entry *pos, *pair; + + if (sort__need_collapse) + root = &leader->entries_collapsed; + else + root = leader->entries_in; + + for (nd = rb_first(root); nd; nd = rb_next(nd)) { + pos = rb_entry(nd, struct hist_entry, rb_node_in); + pair = hists__find_entry(other, pos); + + if (pair) + hist_entry__add_pair(pair, pos); + } +} + +/* + * Look for entries in the other hists that are not present in the leader, if + * we find them, just add a dummy entry on the leader hists, with period=0, + * nr_events=0, to serve as the list header. + */ +int hists__link(struct hists *leader, struct hists *other) +{ + struct rb_root *root; + struct rb_node *nd; + struct hist_entry *pos, *pair; + + if (sort__need_collapse) + root = &other->entries_collapsed; + else + root = other->entries_in; + + for (nd = rb_first(root); nd; nd = rb_next(nd)) { + pos = rb_entry(nd, struct hist_entry, rb_node_in); + + if (!hist_entry__has_pairs(pos)) { + pair = hists__add_dummy_entry(leader, pos); + if (pair == NULL) + return -1; + hist_entry__add_pair(pos, pair); + } + } + + return 0; +} diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h new file mode 100644 index 00000000..38624686 --- /dev/null +++ b/tools/perf/util/hist.h @@ -0,0 +1,236 @@ +#ifndef __PERF_HIST_H +#define __PERF_HIST_H + +#include <linux/types.h> +#include <pthread.h> +#include "callchain.h" +#include "header.h" + +extern struct callchain_param callchain_param; + +struct hist_entry; +struct addr_location; +struct symbol; + +/* + * The kernel collects the number of events it couldn't send in a stretch and + * when possible sends this number in a PERF_RECORD_LOST event. The number of + * such "chunks" of lost events is stored in .nr_events[PERF_EVENT_LOST] while + * total_lost tells exactly how many events the kernel in fact lost, i.e. it is + * the sum of all struct lost_event.lost fields reported. + * + * The total_period is needed because by default auto-freq is used, so + * multipling nr_events[PERF_EVENT_SAMPLE] by a frequency isn't possible to get + * the total number of low level events, it is necessary to to sum all struct + * sample_event.period and stash the result in total_period. + */ +struct events_stats { + u64 total_period; + u64 total_lost; + u64 total_invalid_chains; + u32 nr_events[PERF_RECORD_HEADER_MAX]; + u32 nr_lost_warned; + u32 nr_unknown_events; + u32 nr_invalid_chains; + u32 nr_unknown_id; + u32 nr_unprocessable_samples; +}; + +enum hist_column { + HISTC_SYMBOL, + HISTC_DSO, + HISTC_THREAD, + HISTC_COMM, + HISTC_PARENT, + HISTC_CPU, + HISTC_MISPREDICT, + HISTC_SYMBOL_FROM, + HISTC_SYMBOL_TO, + HISTC_DSO_FROM, + HISTC_DSO_TO, + HISTC_SRCLINE, + HISTC_NR_COLS, /* Last entry */ +}; + +struct thread; +struct dso; + +struct hists { + struct rb_root entries_in_array[2]; + struct rb_root *entries_in; + struct rb_root entries; + struct rb_root entries_collapsed; + u64 nr_entries; + const struct thread *thread_filter; + const struct dso *dso_filter; + const char *uid_filter_str; + const char *symbol_filter_str; + pthread_mutex_t lock; + struct events_stats stats; + u64 event_stream; + u16 col_len[HISTC_NR_COLS]; +}; + +struct hist_entry *__hists__add_entry(struct hists *self, + struct addr_location *al, + struct symbol *parent, u64 period); +int64_t hist_entry__cmp(struct hist_entry *left, struct hist_entry *right); +int64_t hist_entry__collapse(struct hist_entry *left, struct hist_entry *right); +int hist_entry__sort_snprintf(struct hist_entry *self, char *bf, size_t size, + struct hists *hists); +void hist_entry__free(struct hist_entry *); + +struct hist_entry *__hists__add_branch_entry(struct hists *self, + struct addr_location *al, + struct symbol *sym_parent, + struct branch_info *bi, + u64 period); + +void hists__output_resort(struct hists *self); +void hists__output_resort_threaded(struct hists *hists); +void hists__collapse_resort(struct hists *self); +void hists__collapse_resort_threaded(struct hists *hists); + +void hists__decay_entries(struct hists *hists, bool zap_user, bool zap_kernel); +void hists__decay_entries_threaded(struct hists *hists, bool zap_user, + bool zap_kernel); +void hists__output_recalc_col_len(struct hists *hists, int max_rows); + +void hists__inc_nr_entries(struct hists *hists, struct hist_entry *h); +void hists__inc_nr_events(struct hists *self, u32 type); +void events_stats__inc(struct events_stats *stats, u32 type); +size_t events_stats__fprintf(struct events_stats *stats, FILE *fp); + +size_t hists__fprintf(struct hists *self, bool show_header, int max_rows, + int max_cols, FILE *fp); + +int hist_entry__inc_addr_samples(struct hist_entry *self, int evidx, u64 addr); +int hist_entry__annotate(struct hist_entry *self, size_t privsize); + +void hists__filter_by_dso(struct hists *hists); +void hists__filter_by_thread(struct hists *hists); +void hists__filter_by_symbol(struct hists *hists); + +u16 hists__col_len(struct hists *self, enum hist_column col); +void hists__set_col_len(struct hists *self, enum hist_column col, u16 len); +bool hists__new_col_len(struct hists *self, enum hist_column col, u16 len); +void hists__reset_col_len(struct hists *hists); +void hists__calc_col_len(struct hists *hists, struct hist_entry *he); + +void hists__match(struct hists *leader, struct hists *other); +int hists__link(struct hists *leader, struct hists *other); + +struct perf_hpp { + char *buf; + size_t size; + const char *sep; + void *ptr; +}; + +struct perf_hpp_fmt { + int (*header)(struct perf_hpp *hpp); + int (*width)(struct perf_hpp *hpp); + int (*color)(struct perf_hpp *hpp, struct hist_entry *he); + int (*entry)(struct perf_hpp *hpp, struct hist_entry *he); + + struct list_head list; +}; + +extern struct list_head perf_hpp__list; + +#define perf_hpp__for_each_format(format) \ + list_for_each_entry(format, &perf_hpp__list, list) + +extern struct perf_hpp_fmt perf_hpp__format[]; + +enum { + PERF_HPP__BASELINE, + PERF_HPP__OVERHEAD, + PERF_HPP__OVERHEAD_SYS, + PERF_HPP__OVERHEAD_US, + PERF_HPP__OVERHEAD_GUEST_SYS, + PERF_HPP__OVERHEAD_GUEST_US, + PERF_HPP__SAMPLES, + PERF_HPP__PERIOD, + PERF_HPP__PERIOD_BASELINE, + PERF_HPP__DELTA, + PERF_HPP__RATIO, + PERF_HPP__WEIGHTED_DIFF, + PERF_HPP__FORMULA, + + PERF_HPP__MAX_INDEX +}; + +void perf_hpp__init(void); +void perf_hpp__column_register(struct perf_hpp_fmt *format); +void perf_hpp__column_enable(unsigned col); +int hist_entry__period_snprintf(struct perf_hpp *hpp, struct hist_entry *he, + bool color); + +struct perf_evlist; + +struct hist_browser_timer { + void (*timer)(void *arg); + void *arg; + int refresh; +}; + +#ifdef NEWT_SUPPORT +#include "../ui/keysyms.h" +int hist_entry__tui_annotate(struct hist_entry *he, int evidx, + struct hist_browser_timer *hbt); + +int perf_evlist__tui_browse_hists(struct perf_evlist *evlist, const char *help, + struct hist_browser_timer *hbt, + struct perf_session_env *env); +int script_browse(const char *script_opt); +#else +static inline +int perf_evlist__tui_browse_hists(struct perf_evlist *evlist __maybe_unused, + const char *help __maybe_unused, + struct hist_browser_timer *hbt __maybe_unused, + struct perf_session_env *env __maybe_unused) +{ + return 0; +} + +static inline int hist_entry__tui_annotate(struct hist_entry *self + __maybe_unused, + int evidx __maybe_unused, + struct hist_browser_timer *hbt + __maybe_unused) +{ + return 0; +} + +static inline int script_browse(const char *script_opt __maybe_unused) +{ + return 0; +} + +#define K_LEFT -1 +#define K_RIGHT -2 +#endif + +#ifdef GTK2_SUPPORT +int perf_evlist__gtk_browse_hists(struct perf_evlist *evlist, const char *help, + struct hist_browser_timer *hbt __maybe_unused); +#else +static inline +int perf_evlist__gtk_browse_hists(struct perf_evlist *evlist __maybe_unused, + const char *help __maybe_unused, + struct hist_browser_timer *hbt __maybe_unused) +{ + return 0; +} +#endif + +unsigned int hists__sort_list_width(struct hists *self); + +double perf_diff__compute_delta(struct hist_entry *he, struct hist_entry *pair); +double perf_diff__compute_ratio(struct hist_entry *he, struct hist_entry *pair); +s64 perf_diff__compute_wdiff(struct hist_entry *he, struct hist_entry *pair); +int perf_diff__formula(struct hist_entry *he, struct hist_entry *pair, + char *buf, size_t size); +double perf_diff__period_percent(struct hist_entry *he, u64 period); +#endif /* __PERF_HIST_H */ diff --git a/tools/perf/util/hweight.c b/tools/perf/util/hweight.c new file mode 100644 index 00000000..5c1d0d09 --- /dev/null +++ b/tools/perf/util/hweight.c @@ -0,0 +1,31 @@ +#include <linux/bitops.h> + +/** + * hweightN - returns the hamming weight of a N-bit word + * @x: the word to weigh + * + * The Hamming Weight of a number is the total number of bits set in it. + */ + +unsigned int hweight32(unsigned int w) +{ + unsigned int res = w - ((w >> 1) & 0x55555555); + res = (res & 0x33333333) + ((res >> 2) & 0x33333333); + res = (res + (res >> 4)) & 0x0F0F0F0F; + res = res + (res >> 8); + return (res + (res >> 16)) & 0x000000FF; +} + +unsigned long hweight64(__u64 w) +{ +#if BITS_PER_LONG == 32 + return hweight32((unsigned int)(w >> 32)) + hweight32((unsigned int)w); +#elif BITS_PER_LONG == 64 + __u64 res = w - ((w >> 1) & 0x5555555555555555ul); + res = (res & 0x3333333333333333ul) + ((res >> 2) & 0x3333333333333333ul); + res = (res + (res >> 4)) & 0x0F0F0F0F0F0F0F0Ful; + res = res + (res >> 8); + res = res + (res >> 16); + return (res + (res >> 32)) & 0x00000000000000FFul; +#endif +} diff --git a/tools/perf/util/include/asm/alternative-asm.h b/tools/perf/util/include/asm/alternative-asm.h new file mode 100644 index 00000000..6789d788 --- /dev/null +++ b/tools/perf/util/include/asm/alternative-asm.h @@ -0,0 +1,8 @@ +#ifndef _PERF_ASM_ALTERNATIVE_ASM_H +#define _PERF_ASM_ALTERNATIVE_ASM_H + +/* Just disable it so we can build arch/x86/lib/memcpy_64.S for perf bench: */ + +#define altinstruction_entry # + +#endif diff --git a/tools/perf/util/include/asm/asm-offsets.h b/tools/perf/util/include/asm/asm-offsets.h new file mode 100644 index 00000000..ed538942 --- /dev/null +++ b/tools/perf/util/include/asm/asm-offsets.h @@ -0,0 +1 @@ +/* stub */ diff --git a/tools/perf/util/include/asm/bug.h b/tools/perf/util/include/asm/bug.h new file mode 100644 index 00000000..7fcc6810 --- /dev/null +++ b/tools/perf/util/include/asm/bug.h @@ -0,0 +1,22 @@ +#ifndef _PERF_ASM_GENERIC_BUG_H +#define _PERF_ASM_GENERIC_BUG_H + +#define __WARN_printf(arg...) do { fprintf(stderr, arg); } while (0) + +#define WARN(condition, format...) ({ \ + int __ret_warn_on = !!(condition); \ + if (unlikely(__ret_warn_on)) \ + __WARN_printf(format); \ + unlikely(__ret_warn_on); \ +}) + +#define WARN_ONCE(condition, format...) ({ \ + static int __warned; \ + int __ret_warn_once = !!(condition); \ + \ + if (unlikely(__ret_warn_once)) \ + if (WARN(!__warned, format)) \ + __warned = 1; \ + unlikely(__ret_warn_once); \ +}) +#endif diff --git a/tools/perf/util/include/asm/byteorder.h b/tools/perf/util/include/asm/byteorder.h new file mode 100644 index 00000000..2a9bdc06 --- /dev/null +++ b/tools/perf/util/include/asm/byteorder.h @@ -0,0 +1,2 @@ +#include <asm/types.h> +#include "../../../../include/uapi/linux/swab.h" diff --git a/tools/perf/util/include/asm/cpufeature.h b/tools/perf/util/include/asm/cpufeature.h new file mode 100644 index 00000000..acffd5e4 --- /dev/null +++ b/tools/perf/util/include/asm/cpufeature.h @@ -0,0 +1,9 @@ + +#ifndef PERF_CPUFEATURE_H +#define PERF_CPUFEATURE_H + +/* cpufeature.h ... dummy header file for including arch/x86/lib/memcpy_64.S */ + +#define X86_FEATURE_REP_GOOD 0 + +#endif /* PERF_CPUFEATURE_H */ diff --git a/tools/perf/util/include/asm/dwarf2.h b/tools/perf/util/include/asm/dwarf2.h new file mode 100644 index 00000000..afe38199 --- /dev/null +++ b/tools/perf/util/include/asm/dwarf2.h @@ -0,0 +1,13 @@ + +#ifndef PERF_DWARF2_H +#define PERF_DWARF2_H + +/* dwarf2.h ... dummy header file for including arch/x86/lib/mem{cpy,set}_64.S */ + +#define CFI_STARTPROC +#define CFI_ENDPROC +#define CFI_REMEMBER_STATE +#define CFI_RESTORE_STATE + +#endif /* PERF_DWARF2_H */ + diff --git a/tools/perf/util/include/asm/hweight.h b/tools/perf/util/include/asm/hweight.h new file mode 100644 index 00000000..36cf26d4 --- /dev/null +++ b/tools/perf/util/include/asm/hweight.h @@ -0,0 +1,8 @@ +#ifndef PERF_HWEIGHT_H +#define PERF_HWEIGHT_H + +#include <linux/types.h> +unsigned int hweight32(unsigned int w); +unsigned long hweight64(__u64 w); + +#endif /* PERF_HWEIGHT_H */ diff --git a/tools/perf/util/include/asm/swab.h b/tools/perf/util/include/asm/swab.h new file mode 100644 index 00000000..ed538942 --- /dev/null +++ b/tools/perf/util/include/asm/swab.h @@ -0,0 +1 @@ +/* stub */ diff --git a/tools/perf/util/include/asm/system.h b/tools/perf/util/include/asm/system.h new file mode 100644 index 00000000..710cecca --- /dev/null +++ b/tools/perf/util/include/asm/system.h @@ -0,0 +1 @@ +/* Empty */ diff --git a/tools/perf/util/include/asm/uaccess.h b/tools/perf/util/include/asm/uaccess.h new file mode 100644 index 00000000..d0f72b8f --- /dev/null +++ b/tools/perf/util/include/asm/uaccess.h @@ -0,0 +1,14 @@ +#ifndef _PERF_ASM_UACCESS_H_ +#define _PERF_ASM_UACCESS_H_ + +#define __get_user(src, dest) \ +({ \ + (src) = *dest; \ + 0; \ +}) + +#define get_user __get_user + +#define access_ok(type, addr, size) 1 + +#endif diff --git a/tools/perf/util/include/asm/unistd_32.h b/tools/perf/util/include/asm/unistd_32.h new file mode 100644 index 00000000..8b137891 --- /dev/null +++ b/tools/perf/util/include/asm/unistd_32.h @@ -0,0 +1 @@ + diff --git a/tools/perf/util/include/asm/unistd_64.h b/tools/perf/util/include/asm/unistd_64.h new file mode 100644 index 00000000..8b137891 --- /dev/null +++ b/tools/perf/util/include/asm/unistd_64.h @@ -0,0 +1 @@ + diff --git a/tools/perf/util/include/dwarf-regs.h b/tools/perf/util/include/dwarf-regs.h new file mode 100644 index 00000000..cf6727e9 --- /dev/null +++ b/tools/perf/util/include/dwarf-regs.h @@ -0,0 +1,8 @@ +#ifndef _PERF_DWARF_REGS_H_ +#define _PERF_DWARF_REGS_H_ + +#ifdef DWARF_SUPPORT +const char *get_arch_regstr(unsigned int n); +#endif + +#endif diff --git a/tools/perf/util/include/linux/bitmap.h b/tools/perf/util/include/linux/bitmap.h new file mode 100644 index 00000000..bb162e40 --- /dev/null +++ b/tools/perf/util/include/linux/bitmap.h @@ -0,0 +1,46 @@ +#ifndef _PERF_BITOPS_H +#define _PERF_BITOPS_H + +#include <string.h> +#include <linux/bitops.h> + +int __bitmap_weight(const unsigned long *bitmap, int bits); +void __bitmap_or(unsigned long *dst, const unsigned long *bitmap1, + const unsigned long *bitmap2, int bits); + +#define BITMAP_LAST_WORD_MASK(nbits) \ +( \ + ((nbits) % BITS_PER_LONG) ? \ + (1UL<<((nbits) % BITS_PER_LONG))-1 : ~0UL \ +) + +#define small_const_nbits(nbits) \ + (__builtin_constant_p(nbits) && (nbits) <= BITS_PER_LONG) + +static inline void bitmap_zero(unsigned long *dst, int nbits) +{ + if (small_const_nbits(nbits)) + *dst = 0UL; + else { + int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long); + memset(dst, 0, len); + } +} + +static inline int bitmap_weight(const unsigned long *src, int nbits) +{ + if (small_const_nbits(nbits)) + return hweight_long(*src & BITMAP_LAST_WORD_MASK(nbits)); + return __bitmap_weight(src, nbits); +} + +static inline void bitmap_or(unsigned long *dst, const unsigned long *src1, + const unsigned long *src2, int nbits) +{ + if (small_const_nbits(nbits)) + *dst = *src1 | *src2; + else + __bitmap_or(dst, src1, src2, nbits); +} + +#endif /* _PERF_BITOPS_H */ diff --git a/tools/perf/util/include/linux/bitops.h b/tools/perf/util/include/linux/bitops.h new file mode 100644 index 00000000..45cf10a5 --- /dev/null +++ b/tools/perf/util/include/linux/bitops.h @@ -0,0 +1,158 @@ +#ifndef _PERF_LINUX_BITOPS_H_ +#define _PERF_LINUX_BITOPS_H_ + +#include <linux/kernel.h> +#include <linux/compiler.h> +#include <asm/hweight.h> + +#ifndef __WORDSIZE +#define __WORDSIZE (__SIZEOF_LONG__ * 8) +#endif + +#define BITS_PER_LONG __WORDSIZE +#define BITS_PER_BYTE 8 +#define BITS_TO_LONGS(nr) DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(long)) +#define BITS_TO_U64(nr) DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u64)) +#define BITS_TO_U32(nr) DIV_ROUND_UP(nr, BITS_PER_BYTE * sizeof(u32)) +#define BITS_TO_BYTES(nr) DIV_ROUND_UP(nr, BITS_PER_BYTE) + +#define for_each_set_bit(bit, addr, size) \ + for ((bit) = find_first_bit((addr), (size)); \ + (bit) < (size); \ + (bit) = find_next_bit((addr), (size), (bit) + 1)) + +/* same as for_each_set_bit() but use bit as value to start with */ +#define for_each_set_bit_from(bit, addr, size) \ + for ((bit) = find_next_bit((addr), (size), (bit)); \ + (bit) < (size); \ + (bit) = find_next_bit((addr), (size), (bit) + 1)) + +static inline void set_bit(int nr, unsigned long *addr) +{ + addr[nr / BITS_PER_LONG] |= 1UL << (nr % BITS_PER_LONG); +} + +static inline void clear_bit(int nr, unsigned long *addr) +{ + addr[nr / BITS_PER_LONG] &= ~(1UL << (nr % BITS_PER_LONG)); +} + +static __always_inline int test_bit(unsigned int nr, const unsigned long *addr) +{ + return ((1UL << (nr % BITS_PER_LONG)) & + (((unsigned long *)addr)[nr / BITS_PER_LONG])) != 0; +} + +static inline unsigned long hweight_long(unsigned long w) +{ + return sizeof(w) == 4 ? hweight32(w) : hweight64(w); +} + +#define BITOP_WORD(nr) ((nr) / BITS_PER_LONG) + +/** + * __ffs - find first bit in word. + * @word: The word to search + * + * Undefined if no bit exists, so code should check against 0 first. + */ +static __always_inline unsigned long __ffs(unsigned long word) +{ + int num = 0; + +#if BITS_PER_LONG == 64 + if ((word & 0xffffffff) == 0) { + num += 32; + word >>= 32; + } +#endif + if ((word & 0xffff) == 0) { + num += 16; + word >>= 16; + } + if ((word & 0xff) == 0) { + num += 8; + word >>= 8; + } + if ((word & 0xf) == 0) { + num += 4; + word >>= 4; + } + if ((word & 0x3) == 0) { + num += 2; + word >>= 2; + } + if ((word & 0x1) == 0) + num += 1; + return num; +} + +/* + * Find the first set bit in a memory region. + */ +static inline unsigned long +find_first_bit(const unsigned long *addr, unsigned long size) +{ + const unsigned long *p = addr; + unsigned long result = 0; + unsigned long tmp; + + while (size & ~(BITS_PER_LONG-1)) { + if ((tmp = *(p++))) + goto found; + result += BITS_PER_LONG; + size -= BITS_PER_LONG; + } + if (!size) + return result; + + tmp = (*p) & (~0UL >> (BITS_PER_LONG - size)); + if (tmp == 0UL) /* Are any bits set? */ + return result + size; /* Nope. */ +found: + return result + __ffs(tmp); +} + +/* + * Find the next set bit in a memory region. + */ +static inline unsigned long +find_next_bit(const unsigned long *addr, unsigned long size, unsigned long offset) +{ + const unsigned long *p = addr + BITOP_WORD(offset); + unsigned long result = offset & ~(BITS_PER_LONG-1); + unsigned long tmp; + + if (offset >= size) + return size; + size -= result; + offset %= BITS_PER_LONG; + if (offset) { + tmp = *(p++); + tmp &= (~0UL << offset); + if (size < BITS_PER_LONG) + goto found_first; + if (tmp) + goto found_middle; + size -= BITS_PER_LONG; + result += BITS_PER_LONG; + } + while (size & ~(BITS_PER_LONG-1)) { + if ((tmp = *(p++))) + goto found_middle; + result += BITS_PER_LONG; + size -= BITS_PER_LONG; + } + if (!size) + return result; + tmp = *p; + +found_first: + tmp &= (~0UL >> (BITS_PER_LONG - size)); + if (tmp == 0UL) /* Are any bits set? */ + return result + size; /* Nope. */ +found_middle: + return result + __ffs(tmp); +} + +#endif diff --git a/tools/perf/util/include/linux/compiler.h b/tools/perf/util/include/linux/compiler.h new file mode 100644 index 00000000..96b919da --- /dev/null +++ b/tools/perf/util/include/linux/compiler.h @@ -0,0 +1,21 @@ +#ifndef _PERF_LINUX_COMPILER_H_ +#define _PERF_LINUX_COMPILER_H_ + +#ifndef __always_inline +#define __always_inline inline +#endif +#define __user +#ifndef __attribute_const__ +#define __attribute_const__ +#endif + +#ifndef __maybe_unused +#define __maybe_unused __attribute__((unused)) +#endif +#define __packed __attribute__((__packed__)) + +#ifndef __force +#define __force +#endif + +#endif diff --git a/tools/perf/util/include/linux/const.h b/tools/perf/util/include/linux/const.h new file mode 100644 index 00000000..c10a35e1 --- /dev/null +++ b/tools/perf/util/include/linux/const.h @@ -0,0 +1 @@ +#include "../../../../include/uapi/linux/const.h" diff --git a/tools/perf/util/include/linux/ctype.h b/tools/perf/util/include/linux/ctype.h new file mode 100644 index 00000000..a53d4ee1 --- /dev/null +++ b/tools/perf/util/include/linux/ctype.h @@ -0,0 +1 @@ +#include "../util.h" diff --git a/tools/perf/util/include/linux/export.h b/tools/perf/util/include/linux/export.h new file mode 100644 index 00000000..b43e2dc2 --- /dev/null +++ b/tools/perf/util/include/linux/export.h @@ -0,0 +1,6 @@ +#ifndef PERF_LINUX_MODULE_H +#define PERF_LINUX_MODULE_H + +#define EXPORT_SYMBOL(name) + +#endif diff --git a/tools/perf/util/include/linux/hash.h b/tools/perf/util/include/linux/hash.h new file mode 100644 index 00000000..201f5739 --- /dev/null +++ b/tools/perf/util/include/linux/hash.h @@ -0,0 +1,5 @@ +#include "../../../../include/linux/hash.h" + +#ifndef PERF_HASH_H +#define PERF_HASH_H +#endif diff --git a/tools/perf/util/include/linux/kernel.h b/tools/perf/util/include/linux/kernel.h new file mode 100644 index 00000000..d8c927c8 --- /dev/null +++ b/tools/perf/util/include/linux/kernel.h @@ -0,0 +1,134 @@ +#ifndef PERF_LINUX_KERNEL_H_ +#define PERF_LINUX_KERNEL_H_ + +#include <stdarg.h> +#include <stdio.h> +#include <stdlib.h> +#include <assert.h> + +#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d)) + +#define PERF_ALIGN(x, a) __PERF_ALIGN_MASK(x, (typeof(x))(a)-1) +#define __PERF_ALIGN_MASK(x, mask) (((x)+(mask))&~(mask)) + +#ifndef offsetof +#define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER) +#endif + +#ifndef container_of +/** + * container_of - cast a member of a structure out to the containing structure + * @ptr: the pointer to the member. + * @type: the type of the container struct this is embedded in. + * @member: the name of the member within the struct. + * + */ +#define container_of(ptr, type, member) ({ \ + const typeof(((type *)0)->member) * __mptr = (ptr); \ + (type *)((char *)__mptr - offsetof(type, member)); }) +#endif + +#define BUILD_BUG_ON_ZERO(e) (sizeof(struct { int:-!!(e); })) + +#ifndef max +#define max(x, y) ({ \ + typeof(x) _max1 = (x); \ + typeof(y) _max2 = (y); \ + (void) (&_max1 == &_max2); \ + _max1 > _max2 ? _max1 : _max2; }) +#endif + +#ifndef min +#define min(x, y) ({ \ + typeof(x) _min1 = (x); \ + typeof(y) _min2 = (y); \ + (void) (&_min1 == &_min2); \ + _min1 < _min2 ? _min1 : _min2; }) +#endif + +#ifndef roundup +#define roundup(x, y) ( \ +{ \ + const typeof(y) __y = y; \ + (((x) + (__y - 1)) / __y) * __y; \ +} \ +) +#endif + +#ifndef BUG_ON +#ifdef NDEBUG +#define BUG_ON(cond) do { if (cond) {} } while (0) +#else +#define BUG_ON(cond) assert(!(cond)) +#endif +#endif + +/* + * Both need more care to handle endianness + * (Don't use bitmap_copy_le() for now) + */ +#define cpu_to_le64(x) (x) +#define cpu_to_le32(x) (x) + +static inline int +vscnprintf(char *buf, size_t size, const char *fmt, va_list args) +{ + int i; + ssize_t ssize = size; + + i = vsnprintf(buf, size, fmt, args); + + return (i >= ssize) ? (ssize - 1) : i; +} + +static inline int scnprintf(char * buf, size_t size, const char * fmt, ...) +{ + va_list args; + ssize_t ssize = size; + int i; + + va_start(args, fmt); + i = vsnprintf(buf, size, fmt, args); + va_end(args); + + return (i >= ssize) ? (ssize - 1) : i; +} + +static inline unsigned long +simple_strtoul(const char *nptr, char **endptr, int base) +{ + return strtoul(nptr, endptr, base); +} + +int eprintf(int level, + const char *fmt, ...) __attribute__((format(printf, 2, 3))); + +#ifndef pr_fmt +#define pr_fmt(fmt) fmt +#endif + +#define pr_err(fmt, ...) \ + eprintf(0, pr_fmt(fmt), ##__VA_ARGS__) +#define pr_warning(fmt, ...) \ + eprintf(0, pr_fmt(fmt), ##__VA_ARGS__) +#define pr_info(fmt, ...) \ + eprintf(0, pr_fmt(fmt), ##__VA_ARGS__) +#define pr_debug(fmt, ...) \ + eprintf(1, pr_fmt(fmt), ##__VA_ARGS__) +#define pr_debugN(n, fmt, ...) \ + eprintf(n, pr_fmt(fmt), ##__VA_ARGS__) +#define pr_debug2(fmt, ...) pr_debugN(2, pr_fmt(fmt), ##__VA_ARGS__) +#define pr_debug3(fmt, ...) pr_debugN(3, pr_fmt(fmt), ##__VA_ARGS__) +#define pr_debug4(fmt, ...) pr_debugN(4, pr_fmt(fmt), ##__VA_ARGS__) + +/* + * This looks more complex than it should be. But we need to + * get the type for the ~ right in round_down (it needs to be + * as wide as the result!), and we want to evaluate the macro + * arguments just once each. + */ +#define __round_mask(x, y) ((__typeof__(x))((y)-1)) +#define round_up(x, y) ((((x)-1) | __round_mask(x, y))+1) +#define round_down(x, y) ((x) & ~__round_mask(x, y)) + +#endif diff --git a/tools/perf/util/include/linux/linkage.h b/tools/perf/util/include/linux/linkage.h new file mode 100644 index 00000000..06387cff --- /dev/null +++ b/tools/perf/util/include/linux/linkage.h @@ -0,0 +1,13 @@ + +#ifndef PERF_LINUX_LINKAGE_H_ +#define PERF_LINUX_LINKAGE_H_ + +/* linkage.h ... for including arch/x86/lib/memcpy_64.S */ + +#define ENTRY(name) \ + .globl name; \ + name: + +#define ENDPROC(name) + +#endif /* PERF_LINUX_LINKAGE_H_ */ diff --git a/tools/perf/util/include/linux/list.h b/tools/perf/util/include/linux/list.h new file mode 100644 index 00000000..1d928a0c --- /dev/null +++ b/tools/perf/util/include/linux/list.h @@ -0,0 +1,29 @@ +#include <linux/kernel.h> +#include <linux/prefetch.h> + +#include "../../../../include/linux/list.h" + +#ifndef PERF_LIST_H +#define PERF_LIST_H +/** + * list_del_range - deletes range of entries from list. + * @begin: first element in the range to delete from the list. + * @end: last element in the range to delete from the list. + * Note: list_empty on the range of entries does not return true after this, + * the entries is in an undefined state. + */ +static inline void list_del_range(struct list_head *begin, + struct list_head *end) +{ + begin->prev->next = end->next; + end->next->prev = begin->prev; +} + +/** + * list_for_each_from - iterate over a list from one of its nodes + * @pos: the &struct list_head to use as a loop cursor, from where to start + * @head: the head for your list. + */ +#define list_for_each_from(pos, head) \ + for (; pos != (head); pos = pos->next) +#endif diff --git a/tools/perf/util/include/linux/magic.h b/tools/perf/util/include/linux/magic.h new file mode 100644 index 00000000..58b64ed4 --- /dev/null +++ b/tools/perf/util/include/linux/magic.h @@ -0,0 +1,12 @@ +#ifndef _PERF_LINUX_MAGIC_H_ +#define _PERF_LINUX_MAGIC_H_ + +#ifndef DEBUGFS_MAGIC +#define DEBUGFS_MAGIC 0x64626720 +#endif + +#ifndef SYSFS_MAGIC +#define SYSFS_MAGIC 0x62656572 +#endif + +#endif diff --git a/tools/perf/util/include/linux/poison.h b/tools/perf/util/include/linux/poison.h new file mode 100644 index 00000000..fef6dbc9 --- /dev/null +++ b/tools/perf/util/include/linux/poison.h @@ -0,0 +1 @@ +#include "../../../../include/linux/poison.h" diff --git a/tools/perf/util/include/linux/prefetch.h b/tools/perf/util/include/linux/prefetch.h new file mode 100644 index 00000000..7841e485 --- /dev/null +++ b/tools/perf/util/include/linux/prefetch.h @@ -0,0 +1,6 @@ +#ifndef PERF_LINUX_PREFETCH_H +#define PERF_LINUX_PREFETCH_H + +static inline void prefetch(void *a __attribute__((unused))) { } + +#endif diff --git a/tools/perf/util/include/linux/rbtree.h b/tools/perf/util/include/linux/rbtree.h new file mode 100644 index 00000000..2a030c5a --- /dev/null +++ b/tools/perf/util/include/linux/rbtree.h @@ -0,0 +1,2 @@ +#include <stdbool.h> +#include "../../../../include/linux/rbtree.h" diff --git a/tools/perf/util/include/linux/rbtree_augmented.h b/tools/perf/util/include/linux/rbtree_augmented.h new file mode 100644 index 00000000..9d6fcdf1 --- /dev/null +++ b/tools/perf/util/include/linux/rbtree_augmented.h @@ -0,0 +1,2 @@ +#include <stdbool.h> +#include "../../../../include/linux/rbtree_augmented.h" diff --git a/tools/perf/util/include/linux/string.h b/tools/perf/util/include/linux/string.h new file mode 100644 index 00000000..6f19c548 --- /dev/null +++ b/tools/perf/util/include/linux/string.h @@ -0,0 +1,3 @@ +#include <string.h> + +void *memdup(const void *src, size_t len); diff --git a/tools/perf/util/include/linux/types.h b/tools/perf/util/include/linux/types.h new file mode 100644 index 00000000..eb464786 --- /dev/null +++ b/tools/perf/util/include/linux/types.h @@ -0,0 +1,29 @@ +#ifndef _PERF_LINUX_TYPES_H_ +#define _PERF_LINUX_TYPES_H_ + +#include <asm/types.h> + +#ifndef __bitwise +#define __bitwise +#endif + +#ifndef __le32 +typedef __u32 __bitwise __le32; +#endif + +#define DECLARE_BITMAP(name,bits) \ + unsigned long name[BITS_TO_LONGS(bits)] + +struct list_head { + struct list_head *next, *prev; +}; + +struct hlist_head { + struct hlist_node *first; +}; + +struct hlist_node { + struct hlist_node *next, **pprev; +}; + +#endif diff --git a/tools/perf/util/intlist.c b/tools/perf/util/intlist.c new file mode 100644 index 00000000..11a8d86f --- /dev/null +++ b/tools/perf/util/intlist.c @@ -0,0 +1,131 @@ +/* + * Based on intlist.c by: + * (c) 2009 Arnaldo Carvalho de Melo <acme@redhat.com> + * + * Licensed under the GPLv2. + */ + +#include <errno.h> +#include <stdlib.h> +#include <linux/compiler.h> + +#include "intlist.h" + +static struct rb_node *intlist__node_new(struct rblist *rblist __maybe_unused, + const void *entry) +{ + int i = (int)((long)entry); + struct rb_node *rc = NULL; + struct int_node *node = malloc(sizeof(*node)); + + if (node != NULL) { + node->i = i; + rc = &node->rb_node; + } + + return rc; +} + +static void int_node__delete(struct int_node *ilist) +{ + free(ilist); +} + +static void intlist__node_delete(struct rblist *rblist __maybe_unused, + struct rb_node *rb_node) +{ + struct int_node *node = container_of(rb_node, struct int_node, rb_node); + + int_node__delete(node); +} + +static int intlist__node_cmp(struct rb_node *rb_node, const void *entry) +{ + int i = (int)((long)entry); + struct int_node *node = container_of(rb_node, struct int_node, rb_node); + + return node->i - i; +} + +int intlist__add(struct intlist *ilist, int i) +{ + return rblist__add_node(&ilist->rblist, (void *)((long)i)); +} + +void intlist__remove(struct intlist *ilist, struct int_node *node) +{ + rblist__remove_node(&ilist->rblist, &node->rb_node); +} + +struct int_node *intlist__find(struct intlist *ilist, int i) +{ + struct int_node *node; + struct rb_node *rb_node; + + if (ilist == NULL) + return NULL; + + node = NULL; + rb_node = rblist__find(&ilist->rblist, (void *)((long)i)); + if (rb_node) + node = container_of(rb_node, struct int_node, rb_node); + + return node; +} + +static int intlist__parse_list(struct intlist *ilist, const char *s) +{ + char *sep; + int err; + + do { + long value = strtol(s, &sep, 10); + err = -EINVAL; + if (*sep != ',' && *sep != '\0') + break; + err = intlist__add(ilist, value); + if (err) + break; + s = sep + 1; + } while (*sep != '\0'); + + return err; +} + +struct intlist *intlist__new(const char *slist) +{ + struct intlist *ilist = malloc(sizeof(*ilist)); + + if (ilist != NULL) { + rblist__init(&ilist->rblist); + ilist->rblist.node_cmp = intlist__node_cmp; + ilist->rblist.node_new = intlist__node_new; + ilist->rblist.node_delete = intlist__node_delete; + + if (slist && intlist__parse_list(ilist, slist)) + goto out_delete; + } + + return ilist; +out_delete: + intlist__delete(ilist); + return NULL; +} + +void intlist__delete(struct intlist *ilist) +{ + if (ilist != NULL) + rblist__delete(&ilist->rblist); +} + +struct int_node *intlist__entry(const struct intlist *ilist, unsigned int idx) +{ + struct int_node *node = NULL; + struct rb_node *rb_node; + + rb_node = rblist__entry(&ilist->rblist, idx); + if (rb_node) + node = container_of(rb_node, struct int_node, rb_node); + + return node; +} diff --git a/tools/perf/util/intlist.h b/tools/perf/util/intlist.h new file mode 100644 index 00000000..62351dad --- /dev/null +++ b/tools/perf/util/intlist.h @@ -0,0 +1,75 @@ +#ifndef __PERF_INTLIST_H +#define __PERF_INTLIST_H + +#include <linux/rbtree.h> +#include <stdbool.h> + +#include "rblist.h" + +struct int_node { + struct rb_node rb_node; + int i; +}; + +struct intlist { + struct rblist rblist; +}; + +struct intlist *intlist__new(const char *slist); +void intlist__delete(struct intlist *ilist); + +void intlist__remove(struct intlist *ilist, struct int_node *in); +int intlist__add(struct intlist *ilist, int i); + +struct int_node *intlist__entry(const struct intlist *ilist, unsigned int idx); +struct int_node *intlist__find(struct intlist *ilist, int i); + +static inline bool intlist__has_entry(struct intlist *ilist, int i) +{ + return intlist__find(ilist, i) != NULL; +} + +static inline bool intlist__empty(const struct intlist *ilist) +{ + return rblist__empty(&ilist->rblist); +} + +static inline unsigned int intlist__nr_entries(const struct intlist *ilist) +{ + return rblist__nr_entries(&ilist->rblist); +} + +/* For intlist iteration */ +static inline struct int_node *intlist__first(struct intlist *ilist) +{ + struct rb_node *rn = rb_first(&ilist->rblist.entries); + return rn ? rb_entry(rn, struct int_node, rb_node) : NULL; +} +static inline struct int_node *intlist__next(struct int_node *in) +{ + struct rb_node *rn; + if (!in) + return NULL; + rn = rb_next(&in->rb_node); + return rn ? rb_entry(rn, struct int_node, rb_node) : NULL; +} + +/** + * intlist_for_each - iterate over a intlist + * @pos: the &struct int_node to use as a loop cursor. + * @ilist: the &struct intlist for loop. + */ +#define intlist__for_each(pos, ilist) \ + for (pos = intlist__first(ilist); pos; pos = intlist__next(pos)) + +/** + * intlist_for_each_safe - iterate over a intlist safe against removal of + * int_node + * @pos: the &struct int_node to use as a loop cursor. + * @n: another &struct int_node to use as temporary storage. + * @ilist: the &struct intlist for loop. + */ +#define intlist__for_each_safe(pos, n, ilist) \ + for (pos = intlist__first(ilist), n = intlist__next(pos); pos;\ + pos = n, n = intlist__next(n)) +#endif /* __PERF_INTLIST_H */ diff --git a/tools/perf/util/levenshtein.c b/tools/perf/util/levenshtein.c new file mode 100644 index 00000000..e521d151 --- /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 00000000..b0fcb6d8 --- /dev/null +++ b/tools/perf/util/levenshtein.h @@ -0,0 +1,8 @@ +#ifndef __PERF_LEVENSHTEIN_H +#define __PERF_LEVENSHTEIN_H + +int levenshtein(const char *string1, const char *string2, + int swap_penalty, int substition_penalty, + int insertion_penalty, int deletion_penalty); + +#endif /* __PERF_LEVENSHTEIN_H */ diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c new file mode 100644 index 00000000..efdb38e6 --- /dev/null +++ b/tools/perf/util/machine.c @@ -0,0 +1,1226 @@ +#include "callchain.h" +#include "debug.h" +#include "event.h" +#include "evsel.h" +#include "hist.h" +#include "machine.h" +#include "map.h" +#include "sort.h" +#include "strlist.h" +#include "thread.h" +#include <stdbool.h> +#include "unwind.h" + +int machine__init(struct machine *machine, const char *root_dir, pid_t pid) +{ + map_groups__init(&machine->kmaps); + RB_CLEAR_NODE(&machine->rb_node); + INIT_LIST_HEAD(&machine->user_dsos); + INIT_LIST_HEAD(&machine->kernel_dsos); + + machine->threads = RB_ROOT; + INIT_LIST_HEAD(&machine->dead_threads); + machine->last_match = NULL; + + machine->kmaps.machine = machine; + machine->pid = pid; + + machine->root_dir = strdup(root_dir); + if (machine->root_dir == NULL) + return -ENOMEM; + + if (pid != HOST_KERNEL_ID) { + struct thread *thread = machine__findnew_thread(machine, pid); + char comm[64]; + + if (thread == NULL) + return -ENOMEM; + + snprintf(comm, sizeof(comm), "[guest/%d]", pid); + thread__set_comm(thread, comm); + } + + return 0; +} + +static void dsos__delete(struct list_head *dsos) +{ + struct dso *pos, *n; + + list_for_each_entry_safe(pos, n, dsos, node) { + list_del(&pos->node); + dso__delete(pos); + } +} + +void machine__delete_dead_threads(struct machine *machine) +{ + struct thread *n, *t; + + list_for_each_entry_safe(t, n, &machine->dead_threads, node) { + list_del(&t->node); + thread__delete(t); + } +} + +void machine__delete_threads(struct machine *machine) +{ + struct rb_node *nd = rb_first(&machine->threads); + + while (nd) { + struct thread *t = rb_entry(nd, struct thread, rb_node); + + rb_erase(&t->rb_node, &machine->threads); + nd = rb_next(nd); + thread__delete(t); + } +} + +void machine__exit(struct machine *machine) +{ + map_groups__exit(&machine->kmaps); + dsos__delete(&machine->user_dsos); + dsos__delete(&machine->kernel_dsos); + free(machine->root_dir); + machine->root_dir = NULL; +} + +void machine__delete(struct machine *machine) +{ + machine__exit(machine); + free(machine); +} + +void machines__init(struct machines *machines) +{ + machine__init(&machines->host, "", HOST_KERNEL_ID); + machines->guests = RB_ROOT; +} + +void machines__exit(struct machines *machines) +{ + machine__exit(&machines->host); + /* XXX exit guest */ +} + +struct machine *machines__add(struct machines *machines, pid_t pid, + const char *root_dir) +{ + struct rb_node **p = &machines->guests.rb_node; + struct rb_node *parent = NULL; + struct machine *pos, *machine = malloc(sizeof(*machine)); + + if (machine == NULL) + return NULL; + + if (machine__init(machine, root_dir, pid) != 0) { + free(machine); + return NULL; + } + + while (*p != NULL) { + parent = *p; + pos = rb_entry(parent, struct machine, rb_node); + if (pid < pos->pid) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + rb_link_node(&machine->rb_node, parent, p); + rb_insert_color(&machine->rb_node, &machines->guests); + + return machine; +} + +struct machine *machines__find(struct machines *machines, pid_t pid) +{ + struct rb_node **p = &machines->guests.rb_node; + struct rb_node *parent = NULL; + struct machine *machine; + struct machine *default_machine = NULL; + + if (pid == HOST_KERNEL_ID) + return &machines->host; + + while (*p != NULL) { + parent = *p; + machine = rb_entry(parent, struct machine, rb_node); + if (pid < machine->pid) + p = &(*p)->rb_left; + else if (pid > machine->pid) + p = &(*p)->rb_right; + else + return machine; + if (!machine->pid) + default_machine = machine; + } + + return default_machine; +} + +struct machine *machines__findnew(struct machines *machines, pid_t pid) +{ + char path[PATH_MAX]; + const char *root_dir = ""; + struct machine *machine = machines__find(machines, pid); + + if (machine && (machine->pid == pid)) + goto out; + + if ((pid != HOST_KERNEL_ID) && + (pid != DEFAULT_GUEST_KERNEL_ID) && + (symbol_conf.guestmount)) { + sprintf(path, "%s/%d", symbol_conf.guestmount, pid); + if (access(path, R_OK)) { + static struct strlist *seen; + + if (!seen) + seen = strlist__new(true, NULL); + + if (!strlist__has_entry(seen, path)) { + pr_err("Can't access file %s\n", path); + strlist__add(seen, path); + } + machine = NULL; + goto out; + } + root_dir = path; + } + + machine = machines__add(machines, pid, root_dir); +out: + return machine; +} + +void machines__process_guests(struct machines *machines, + machine__process_t process, void *data) +{ + struct rb_node *nd; + + for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) { + struct machine *pos = rb_entry(nd, struct machine, rb_node); + process(pos, data); + } +} + +char *machine__mmap_name(struct machine *machine, char *bf, size_t size) +{ + if (machine__is_host(machine)) + snprintf(bf, size, "[%s]", "kernel.kallsyms"); + else if (machine__is_default_guest(machine)) + snprintf(bf, size, "[%s]", "guest.kernel.kallsyms"); + else { + snprintf(bf, size, "[%s.%d]", "guest.kernel.kallsyms", + machine->pid); + } + + return bf; +} + +void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size) +{ + struct rb_node *node; + struct machine *machine; + + machines->host.id_hdr_size = id_hdr_size; + + for (node = rb_first(&machines->guests); node; node = rb_next(node)) { + machine = rb_entry(node, struct machine, rb_node); + machine->id_hdr_size = id_hdr_size; + } + + return; +} + +static struct thread *__machine__findnew_thread(struct machine *machine, pid_t pid, + bool create) +{ + struct rb_node **p = &machine->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 (machine->last_match && machine->last_match->pid == pid) + return machine->last_match; + + while (*p != NULL) { + parent = *p; + th = rb_entry(parent, struct thread, rb_node); + + if (th->pid == pid) { + machine->last_match = th; + return th; + } + + if (pid < th->pid) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + if (!create) + return NULL; + + th = thread__new(pid); + if (th != NULL) { + rb_link_node(&th->rb_node, parent, p); + rb_insert_color(&th->rb_node, &machine->threads); + machine->last_match = th; + } + + return th; +} + +struct thread *machine__findnew_thread(struct machine *machine, pid_t pid) +{ + return __machine__findnew_thread(machine, pid, true); +} + +struct thread *machine__find_thread(struct machine *machine, pid_t pid) +{ + return __machine__findnew_thread(machine, pid, false); +} + +int machine__process_comm_event(struct machine *machine, union perf_event *event) +{ + struct thread *thread = machine__findnew_thread(machine, event->comm.tid); + + if (dump_trace) + perf_event__fprintf_comm(event, stdout); + + if (thread == NULL || thread__set_comm(thread, event->comm.comm)) { + dump_printf("problem processing PERF_RECORD_COMM, skipping event.\n"); + return -1; + } + + return 0; +} + +int machine__process_lost_event(struct machine *machine __maybe_unused, + union perf_event *event) +{ + dump_printf(": id:%" PRIu64 ": lost:%" PRIu64 "\n", + event->lost.id, event->lost.lost); + return 0; +} + +struct map *machine__new_module(struct machine *machine, u64 start, + const char *filename) +{ + struct map *map; + struct dso *dso = __dsos__findnew(&machine->kernel_dsos, filename); + + if (dso == NULL) + return NULL; + + map = map__new2(start, dso, MAP__FUNCTION); + if (map == NULL) + return NULL; + + if (machine__is_host(machine)) + dso->symtab_type = DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE; + else + dso->symtab_type = DSO_BINARY_TYPE__GUEST_KMODULE; + map_groups__insert(&machine->kmaps, map); + return map; +} + +size_t machines__fprintf_dsos(struct machines *machines, FILE *fp) +{ + struct rb_node *nd; + size_t ret = __dsos__fprintf(&machines->host.kernel_dsos, fp) + + __dsos__fprintf(&machines->host.user_dsos, fp); + + for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) { + struct machine *pos = rb_entry(nd, struct machine, rb_node); + ret += __dsos__fprintf(&pos->kernel_dsos, fp); + ret += __dsos__fprintf(&pos->user_dsos, fp); + } + + return ret; +} + +size_t machine__fprintf_dsos_buildid(struct machine *machine, FILE *fp, + bool (skip)(struct dso *dso, int parm), int parm) +{ + return __dsos__fprintf_buildid(&machine->kernel_dsos, fp, skip, parm) + + __dsos__fprintf_buildid(&machine->user_dsos, fp, skip, parm); +} + +size_t machines__fprintf_dsos_buildid(struct machines *machines, FILE *fp, + bool (skip)(struct dso *dso, int parm), int parm) +{ + struct rb_node *nd; + size_t ret = machine__fprintf_dsos_buildid(&machines->host, fp, skip, parm); + + for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) { + struct machine *pos = rb_entry(nd, struct machine, rb_node); + ret += machine__fprintf_dsos_buildid(pos, fp, skip, parm); + } + return ret; +} + +size_t machine__fprintf_vmlinux_path(struct machine *machine, FILE *fp) +{ + int i; + size_t printed = 0; + struct dso *kdso = machine->vmlinux_maps[MAP__FUNCTION]->dso; + + if (kdso->has_build_id) { + char filename[PATH_MAX]; + if (dso__build_id_filename(kdso, filename, sizeof(filename))) + printed += fprintf(fp, "[0] %s\n", filename); + } + + for (i = 0; i < vmlinux_path__nr_entries; ++i) + printed += fprintf(fp, "[%d] %s\n", + i + kdso->has_build_id, vmlinux_path[i]); + + return printed; +} + +size_t machine__fprintf(struct machine *machine, FILE *fp) +{ + size_t ret = 0; + struct rb_node *nd; + + for (nd = rb_first(&machine->threads); nd; nd = rb_next(nd)) { + struct thread *pos = rb_entry(nd, struct thread, rb_node); + + ret += thread__fprintf(pos, fp); + } + + return ret; +} + +static struct dso *machine__get_kernel(struct machine *machine) +{ + const char *vmlinux_name = NULL; + struct dso *kernel; + + if (machine__is_host(machine)) { + vmlinux_name = symbol_conf.vmlinux_name; + if (!vmlinux_name) + vmlinux_name = "[kernel.kallsyms]"; + + kernel = dso__kernel_findnew(machine, vmlinux_name, + "[kernel]", + DSO_TYPE_KERNEL); + } else { + char bf[PATH_MAX]; + + if (machine__is_default_guest(machine)) + vmlinux_name = symbol_conf.default_guest_vmlinux_name; + if (!vmlinux_name) + vmlinux_name = machine__mmap_name(machine, bf, + sizeof(bf)); + + kernel = dso__kernel_findnew(machine, vmlinux_name, + "[guest.kernel]", + DSO_TYPE_GUEST_KERNEL); + } + + if (kernel != NULL && (!kernel->has_build_id)) + dso__read_running_kernel_build_id(kernel, machine); + + return kernel; +} + +struct process_args { + u64 start; +}; + +static int symbol__in_kernel(void *arg, const char *name, + char type __maybe_unused, u64 start) +{ + struct process_args *args = arg; + + if (strchr(name, '[')) + return 0; + + args->start = start; + return 1; +} + +/* Figure out the start address of kernel map from /proc/kallsyms */ +static u64 machine__get_kernel_start_addr(struct machine *machine) +{ + const char *filename; + char path[PATH_MAX]; + struct process_args args; + + if (machine__is_host(machine)) { + filename = "/proc/kallsyms"; + } else { + if (machine__is_default_guest(machine)) + filename = (char *)symbol_conf.default_guest_kallsyms; + else { + sprintf(path, "%s/proc/kallsyms", machine->root_dir); + filename = path; + } + } + + if (symbol__restricted_filename(filename, "/proc/kallsyms")) + return 0; + + if (kallsyms__parse(filename, &args, symbol__in_kernel) <= 0) + return 0; + + return args.start; +} + +int __machine__create_kernel_maps(struct machine *machine, struct dso *kernel) +{ + enum map_type type; + u64 start = machine__get_kernel_start_addr(machine); + + for (type = 0; type < MAP__NR_TYPES; ++type) { + struct kmap *kmap; + + machine->vmlinux_maps[type] = map__new2(start, kernel, type); + if (machine->vmlinux_maps[type] == NULL) + return -1; + + machine->vmlinux_maps[type]->map_ip = + machine->vmlinux_maps[type]->unmap_ip = + identity__map_ip; + kmap = map__kmap(machine->vmlinux_maps[type]); + kmap->kmaps = &machine->kmaps; + map_groups__insert(&machine->kmaps, + machine->vmlinux_maps[type]); + } + + return 0; +} + +void machine__destroy_kernel_maps(struct machine *machine) +{ + enum map_type type; + + for (type = 0; type < MAP__NR_TYPES; ++type) { + struct kmap *kmap; + + if (machine->vmlinux_maps[type] == NULL) + continue; + + kmap = map__kmap(machine->vmlinux_maps[type]); + map_groups__remove(&machine->kmaps, + machine->vmlinux_maps[type]); + if (kmap->ref_reloc_sym) { + /* + * ref_reloc_sym is shared among all maps, so free just + * on one of them. + */ + if (type == MAP__FUNCTION) { + free((char *)kmap->ref_reloc_sym->name); + kmap->ref_reloc_sym->name = NULL; + free(kmap->ref_reloc_sym); + } + kmap->ref_reloc_sym = NULL; + } + + map__delete(machine->vmlinux_maps[type]); + machine->vmlinux_maps[type] = NULL; + } +} + +int machines__create_guest_kernel_maps(struct machines *machines) +{ + int ret = 0; + struct dirent **namelist = NULL; + int i, items = 0; + char path[PATH_MAX]; + pid_t pid; + char *endp; + + if (symbol_conf.default_guest_vmlinux_name || + symbol_conf.default_guest_modules || + symbol_conf.default_guest_kallsyms) { + machines__create_kernel_maps(machines, DEFAULT_GUEST_KERNEL_ID); + } + + if (symbol_conf.guestmount) { + items = scandir(symbol_conf.guestmount, &namelist, NULL, NULL); + if (items <= 0) + return -ENOENT; + for (i = 0; i < items; i++) { + if (!isdigit(namelist[i]->d_name[0])) { + /* Filter out . and .. */ + continue; + } + pid = (pid_t)strtol(namelist[i]->d_name, &endp, 10); + if ((*endp != '\0') || + (endp == namelist[i]->d_name) || + (errno == ERANGE)) { + pr_debug("invalid directory (%s). Skipping.\n", + namelist[i]->d_name); + continue; + } + sprintf(path, "%s/%s/proc/kallsyms", + symbol_conf.guestmount, + namelist[i]->d_name); + ret = access(path, R_OK); + if (ret) { + pr_debug("Can't access file %s\n", path); + goto failure; + } + machines__create_kernel_maps(machines, pid); + } +failure: + free(namelist); + } + + return ret; +} + +void machines__destroy_kernel_maps(struct machines *machines) +{ + struct rb_node *next = rb_first(&machines->guests); + + machine__destroy_kernel_maps(&machines->host); + + while (next) { + struct machine *pos = rb_entry(next, struct machine, rb_node); + + next = rb_next(&pos->rb_node); + rb_erase(&pos->rb_node, &machines->guests); + machine__delete(pos); + } +} + +int machines__create_kernel_maps(struct machines *machines, pid_t pid) +{ + struct machine *machine = machines__findnew(machines, pid); + + if (machine == NULL) + return -1; + + return machine__create_kernel_maps(machine); +} + +int machine__load_kallsyms(struct machine *machine, const char *filename, + enum map_type type, symbol_filter_t filter) +{ + struct map *map = machine->vmlinux_maps[type]; + int ret = dso__load_kallsyms(map->dso, filename, map, filter); + + if (ret > 0) { + dso__set_loaded(map->dso, type); + /* + * Since /proc/kallsyms will have multiple sessions for the + * kernel, with modules between them, fixup the end of all + * sections. + */ + __map_groups__fixup_end(&machine->kmaps, type); + } + + return ret; +} + +int machine__load_vmlinux_path(struct machine *machine, enum map_type type, + symbol_filter_t filter) +{ + struct map *map = machine->vmlinux_maps[type]; + int ret = dso__load_vmlinux_path(map->dso, map, filter); + + if (ret > 0) { + dso__set_loaded(map->dso, type); + map__reloc_vmlinux(map); + } + + return ret; +} + +static void map_groups__fixup_end(struct map_groups *mg) +{ + int i; + for (i = 0; i < MAP__NR_TYPES; ++i) + __map_groups__fixup_end(mg, i); +} + +static char *get_kernel_version(const char *root_dir) +{ + char version[PATH_MAX]; + FILE *file; + char *name, *tmp; + const char *prefix = "Linux version "; + + sprintf(version, "%s/proc/version", root_dir); + file = fopen(version, "r"); + if (!file) + return NULL; + + version[0] = '\0'; + tmp = fgets(version, sizeof(version), file); + fclose(file); + + name = strstr(version, prefix); + if (!name) + return NULL; + name += strlen(prefix); + tmp = strchr(name, ' '); + if (tmp) + *tmp = '\0'; + + return strdup(name); +} + +static int map_groups__set_modules_path_dir(struct map_groups *mg, + const char *dir_name) +{ + struct dirent *dent; + DIR *dir = opendir(dir_name); + int ret = 0; + + if (!dir) { + pr_debug("%s: cannot open %s dir\n", __func__, dir_name); + return -1; + } + + while ((dent = readdir(dir)) != NULL) { + char path[PATH_MAX]; + struct stat st; + + /*sshfs might return bad dent->d_type, so we have to stat*/ + snprintf(path, sizeof(path), "%s/%s", dir_name, dent->d_name); + if (stat(path, &st)) + continue; + + if (S_ISDIR(st.st_mode)) { + if (!strcmp(dent->d_name, ".") || + !strcmp(dent->d_name, "..")) + continue; + + ret = map_groups__set_modules_path_dir(mg, path); + if (ret < 0) + goto out; + } else { + char *dot = strrchr(dent->d_name, '.'), + dso_name[PATH_MAX]; + struct map *map; + char *long_name; + + if (dot == NULL || strcmp(dot, ".ko")) + continue; + snprintf(dso_name, sizeof(dso_name), "[%.*s]", + (int)(dot - dent->d_name), dent->d_name); + + strxfrchar(dso_name, '-', '_'); + map = map_groups__find_by_name(mg, MAP__FUNCTION, + dso_name); + if (map == NULL) + continue; + + long_name = strdup(path); + if (long_name == NULL) { + ret = -1; + goto out; + } + dso__set_long_name(map->dso, long_name); + map->dso->lname_alloc = 1; + dso__kernel_module_get_build_id(map->dso, ""); + } + } + +out: + closedir(dir); + return ret; +} + +static int machine__set_modules_path(struct machine *machine) +{ + char *version; + char modules_path[PATH_MAX]; + + version = get_kernel_version(machine->root_dir); + if (!version) + return -1; + + snprintf(modules_path, sizeof(modules_path), "%s/lib/modules/%s/kernel", + machine->root_dir, version); + free(version); + + return map_groups__set_modules_path_dir(&machine->kmaps, modules_path); +} + +static int machine__create_modules(struct machine *machine) +{ + char *line = NULL; + size_t n; + FILE *file; + struct map *map; + const char *modules; + char path[PATH_MAX]; + + if (machine__is_default_guest(machine)) + modules = symbol_conf.default_guest_modules; + else { + sprintf(path, "%s/proc/modules", machine->root_dir); + modules = path; + } + + if (symbol__restricted_filename(path, "/proc/modules")) + return -1; + + file = fopen(modules, "r"); + if (file == NULL) + return -1; + + while (!feof(file)) { + char name[PATH_MAX]; + u64 start; + char *sep; + int line_len; + + line_len = getline(&line, &n, file); + if (line_len < 0) + break; + + if (!line) + goto out_failure; + + line[--line_len] = '\0'; /* \n */ + + sep = strrchr(line, 'x'); + if (sep == NULL) + continue; + + hex2u64(sep + 1, &start); + + sep = strchr(line, ' '); + if (sep == NULL) + continue; + + *sep = '\0'; + + snprintf(name, sizeof(name), "[%s]", line); + map = machine__new_module(machine, start, name); + if (map == NULL) + goto out_delete_line; + dso__kernel_module_get_build_id(map->dso, machine->root_dir); + } + + free(line); + fclose(file); + + return machine__set_modules_path(machine); + +out_delete_line: + free(line); +out_failure: + return -1; +} + +int machine__create_kernel_maps(struct machine *machine) +{ + struct dso *kernel = machine__get_kernel(machine); + + if (kernel == NULL || + __machine__create_kernel_maps(machine, kernel) < 0) + return -1; + + if (symbol_conf.use_modules && machine__create_modules(machine) < 0) { + if (machine__is_host(machine)) + pr_debug("Problems creating module maps, " + "continuing anyway...\n"); + else + pr_debug("Problems creating module maps for guest %d, " + "continuing anyway...\n", machine->pid); + } + + /* + * Now that we have all the maps created, just set the ->end of them: + */ + map_groups__fixup_end(&machine->kmaps); + return 0; +} + +static void machine__set_kernel_mmap_len(struct machine *machine, + union perf_event *event) +{ + int i; + + for (i = 0; i < MAP__NR_TYPES; i++) { + machine->vmlinux_maps[i]->start = event->mmap.start; + machine->vmlinux_maps[i]->end = (event->mmap.start + + event->mmap.len); + /* + * Be a bit paranoid here, some perf.data file came with + * a zero sized synthesized MMAP event for the kernel. + */ + if (machine->vmlinux_maps[i]->end == 0) + machine->vmlinux_maps[i]->end = ~0ULL; + } +} + +static int machine__process_kernel_mmap_event(struct machine *machine, + union perf_event *event) +{ + struct map *map; + char kmmap_prefix[PATH_MAX]; + enum dso_kernel_type kernel_type; + bool is_kernel_mmap; + + machine__mmap_name(machine, kmmap_prefix, sizeof(kmmap_prefix)); + if (machine__is_host(machine)) + kernel_type = DSO_TYPE_KERNEL; + else + kernel_type = DSO_TYPE_GUEST_KERNEL; + + is_kernel_mmap = memcmp(event->mmap.filename, + kmmap_prefix, + strlen(kmmap_prefix) - 1) == 0; + if (event->mmap.filename[0] == '/' || + (!is_kernel_mmap && event->mmap.filename[0] == '[')) { + + char short_module_name[1024]; + char *name, *dot; + + if (event->mmap.filename[0] == '/') { + name = strrchr(event->mmap.filename, '/'); + if (name == NULL) + goto out_problem; + + ++name; /* skip / */ + dot = strrchr(name, '.'); + if (dot == NULL) + goto out_problem; + snprintf(short_module_name, sizeof(short_module_name), + "[%.*s]", (int)(dot - name), name); + strxfrchar(short_module_name, '-', '_'); + } else + strcpy(short_module_name, event->mmap.filename); + + map = machine__new_module(machine, event->mmap.start, + event->mmap.filename); + if (map == NULL) + goto out_problem; + + name = strdup(short_module_name); + if (name == NULL) + goto out_problem; + + map->dso->short_name = name; + map->dso->sname_alloc = 1; + map->end = map->start + event->mmap.len; + } else if (is_kernel_mmap) { + const char *symbol_name = (event->mmap.filename + + strlen(kmmap_prefix)); + /* + * Should be there already, from the build-id table in + * the header. + */ + struct dso *kernel = __dsos__findnew(&machine->kernel_dsos, + kmmap_prefix); + if (kernel == NULL) + goto out_problem; + + kernel->kernel = kernel_type; + if (__machine__create_kernel_maps(machine, kernel) < 0) + goto out_problem; + + machine__set_kernel_mmap_len(machine, event); + + /* + * Avoid using a zero address (kptr_restrict) for the ref reloc + * symbol. Effectively having zero here means that at record + * time /proc/sys/kernel/kptr_restrict was non zero. + */ + if (event->mmap.pgoff != 0) { + maps__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps, + symbol_name, + event->mmap.pgoff); + } + + if (machine__is_default_guest(machine)) { + /* + * preload dso of guest kernel and modules + */ + dso__load(kernel, machine->vmlinux_maps[MAP__FUNCTION], + NULL); + } + } + return 0; +out_problem: + return -1; +} + +int machine__process_mmap_event(struct machine *machine, union perf_event *event) +{ + u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; + struct thread *thread; + struct map *map; + int ret = 0; + + if (dump_trace) + perf_event__fprintf_mmap(event, stdout); + + if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL || + cpumode == PERF_RECORD_MISC_KERNEL) { + ret = machine__process_kernel_mmap_event(machine, event); + if (ret < 0) + goto out_problem; + return 0; + } + + thread = machine__findnew_thread(machine, event->mmap.pid); + if (thread == NULL) + goto out_problem; + map = map__new(&machine->user_dsos, event->mmap.start, + event->mmap.len, event->mmap.pgoff, + event->mmap.pid, event->mmap.filename, + MAP__FUNCTION); + if (map == NULL) + goto out_problem; + + thread__insert_map(thread, map); + return 0; + +out_problem: + dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n"); + return 0; +} + +int machine__process_fork_event(struct machine *machine, union perf_event *event) +{ + struct thread *thread = machine__findnew_thread(machine, event->fork.tid); + struct thread *parent = machine__findnew_thread(machine, event->fork.ptid); + + if (dump_trace) + perf_event__fprintf_task(event, stdout); + + if (thread == NULL || parent == NULL || + thread__fork(thread, parent) < 0) { + dump_printf("problem processing PERF_RECORD_FORK, skipping event.\n"); + return -1; + } + + return 0; +} + +int machine__process_exit_event(struct machine *machine, union perf_event *event) +{ + struct thread *thread = machine__find_thread(machine, event->fork.tid); + + if (dump_trace) + perf_event__fprintf_task(event, stdout); + + if (thread != NULL) + machine__remove_thread(machine, thread); + + return 0; +} + +int machine__process_event(struct machine *machine, union perf_event *event) +{ + int ret; + + switch (event->header.type) { + case PERF_RECORD_COMM: + ret = machine__process_comm_event(machine, event); break; + case PERF_RECORD_MMAP: + ret = machine__process_mmap_event(machine, event); break; + case PERF_RECORD_FORK: + ret = machine__process_fork_event(machine, event); break; + case PERF_RECORD_EXIT: + ret = machine__process_exit_event(machine, event); break; + case PERF_RECORD_LOST: + ret = machine__process_lost_event(machine, event); break; + default: + ret = -1; + break; + } + + return ret; +} + +void machine__remove_thread(struct machine *machine, struct thread *th) +{ + machine->last_match = NULL; + rb_erase(&th->rb_node, &machine->threads); + /* + * We may have references to this thread, for instance in some hist_entry + * instances, so just move them to a separate list. + */ + list_add_tail(&th->node, &machine->dead_threads); +} + +static bool symbol__match_parent_regex(struct symbol *sym) +{ + if (sym->name && !regexec(&parent_regex, sym->name, 0, NULL, 0)) + return 1; + + return 0; +} + +static const u8 cpumodes[] = { + PERF_RECORD_MISC_USER, + PERF_RECORD_MISC_KERNEL, + PERF_RECORD_MISC_GUEST_USER, + PERF_RECORD_MISC_GUEST_KERNEL +}; +#define NCPUMODES (sizeof(cpumodes)/sizeof(u8)) + +static void ip__resolve_ams(struct machine *machine, struct thread *thread, + struct addr_map_symbol *ams, + u64 ip) +{ + struct addr_location al; + size_t i; + u8 m; + + memset(&al, 0, sizeof(al)); + + for (i = 0; i < NCPUMODES; i++) { + m = cpumodes[i]; + /* + * We cannot use the header.misc hint to determine whether a + * branch stack address is user, kernel, guest, hypervisor. + * Branches may straddle the kernel/user/hypervisor boundaries. + * Thus, we have to try consecutively until we find a match + * or else, the symbol is unknown + */ + thread__find_addr_location(thread, machine, m, MAP__FUNCTION, + ip, &al, NULL); + if (al.sym) + goto found; + } +found: + ams->addr = ip; + ams->al_addr = al.addr; + ams->sym = al.sym; + ams->map = al.map; +} + +struct branch_info *machine__resolve_bstack(struct machine *machine, + struct thread *thr, + struct branch_stack *bs) +{ + struct branch_info *bi; + unsigned int i; + + bi = calloc(bs->nr, sizeof(struct branch_info)); + if (!bi) + return NULL; + + for (i = 0; i < bs->nr; i++) { + ip__resolve_ams(machine, thr, &bi[i].to, bs->entries[i].to); + ip__resolve_ams(machine, thr, &bi[i].from, bs->entries[i].from); + bi[i].flags = bs->entries[i].flags; + } + return bi; +} + +static int machine__resolve_callchain_sample(struct machine *machine, + struct thread *thread, + struct ip_callchain *chain, + struct symbol **parent) + +{ + u8 cpumode = PERF_RECORD_MISC_USER; + unsigned int i; + int err; + + callchain_cursor_reset(&callchain_cursor); + + if (chain->nr > PERF_MAX_STACK_DEPTH) { + pr_warning("corrupted callchain. skipping...\n"); + return 0; + } + + for (i = 0; i < chain->nr; i++) { + u64 ip; + struct addr_location al; + + if (callchain_param.order == ORDER_CALLEE) + ip = chain->ips[i]; + else + ip = chain->ips[chain->nr - i - 1]; + + if (ip >= PERF_CONTEXT_MAX) { + switch (ip) { + case PERF_CONTEXT_HV: + cpumode = PERF_RECORD_MISC_HYPERVISOR; + break; + case PERF_CONTEXT_KERNEL: + cpumode = PERF_RECORD_MISC_KERNEL; + break; + case PERF_CONTEXT_USER: + cpumode = PERF_RECORD_MISC_USER; + break; + default: + pr_debug("invalid callchain context: " + "%"PRId64"\n", (s64) ip); + /* + * It seems the callchain is corrupted. + * Discard all. + */ + callchain_cursor_reset(&callchain_cursor); + return 0; + } + continue; + } + + al.filtered = false; + thread__find_addr_location(thread, machine, cpumode, + MAP__FUNCTION, ip, &al, NULL); + if (al.sym != NULL) { + if (sort__has_parent && !*parent && + symbol__match_parent_regex(al.sym)) + *parent = al.sym; + if (!symbol_conf.use_callchain) + break; + } + + err = callchain_cursor_append(&callchain_cursor, + ip, al.map, al.sym); + if (err) + return err; + } + + return 0; +} + +static int unwind_entry(struct unwind_entry *entry, void *arg) +{ + struct callchain_cursor *cursor = arg; + return callchain_cursor_append(cursor, entry->ip, + entry->map, entry->sym); +} + +int machine__resolve_callchain(struct machine *machine, + struct perf_evsel *evsel, + struct thread *thread, + struct perf_sample *sample, + struct symbol **parent) + +{ + int ret; + + callchain_cursor_reset(&callchain_cursor); + + ret = machine__resolve_callchain_sample(machine, thread, + sample->callchain, parent); + if (ret) + return ret; + + /* Can we do dwarf post unwind? */ + if (!((evsel->attr.sample_type & PERF_SAMPLE_REGS_USER) && + (evsel->attr.sample_type & PERF_SAMPLE_STACK_USER))) + return 0; + + /* Bail out if nothing was captured. */ + if ((!sample->user_regs.regs) || + (!sample->user_stack.size)) + return 0; + + return unwind__get_entries(unwind_entry, &callchain_cursor, machine, + thread, evsel->attr.sample_regs_user, + sample); + +} diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h new file mode 100644 index 00000000..5ac5892f --- /dev/null +++ b/tools/perf/util/machine.h @@ -0,0 +1,157 @@ +#ifndef __PERF_MACHINE_H +#define __PERF_MACHINE_H + +#include <sys/types.h> +#include <linux/rbtree.h> +#include "map.h" + +struct branch_stack; +struct perf_evsel; +struct perf_sample; +struct symbol; +struct thread; +union perf_event; + +/* Native host kernel uses -1 as pid index in machine */ +#define HOST_KERNEL_ID (-1) +#define DEFAULT_GUEST_KERNEL_ID (0) + +struct machine { + struct rb_node rb_node; + pid_t pid; + u16 id_hdr_size; + char *root_dir; + struct rb_root threads; + struct list_head dead_threads; + struct thread *last_match; + struct list_head user_dsos; + struct list_head kernel_dsos; + struct map_groups kmaps; + struct map *vmlinux_maps[MAP__NR_TYPES]; +}; + +static inline +struct map *machine__kernel_map(struct machine *machine, enum map_type type) +{ + return machine->vmlinux_maps[type]; +} + +struct thread *machine__find_thread(struct machine *machine, pid_t pid); + +int machine__process_comm_event(struct machine *machine, union perf_event *event); +int machine__process_exit_event(struct machine *machine, union perf_event *event); +int machine__process_fork_event(struct machine *machine, union perf_event *event); +int machine__process_lost_event(struct machine *machine, union perf_event *event); +int machine__process_mmap_event(struct machine *machine, union perf_event *event); +int machine__process_event(struct machine *machine, union perf_event *event); + +typedef void (*machine__process_t)(struct machine *machine, void *data); + +struct machines { + struct machine host; + struct rb_root guests; +}; + +void machines__init(struct machines *machines); +void machines__exit(struct machines *machines); + +void machines__process_guests(struct machines *machines, + machine__process_t process, void *data); + +struct machine *machines__add(struct machines *machines, pid_t pid, + const char *root_dir); +struct machine *machines__find_host(struct machines *machines); +struct machine *machines__find(struct machines *machines, pid_t pid); +struct machine *machines__findnew(struct machines *machines, pid_t pid); + +void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size); +char *machine__mmap_name(struct machine *machine, char *bf, size_t size); + +int machine__init(struct machine *machine, const char *root_dir, pid_t pid); +void machine__exit(struct machine *machine); +void machine__delete_dead_threads(struct machine *machine); +void machine__delete_threads(struct machine *machine); +void machine__delete(struct machine *machine); + +struct branch_info *machine__resolve_bstack(struct machine *machine, + struct thread *thread, + struct branch_stack *bs); +int machine__resolve_callchain(struct machine *machine, + struct perf_evsel *evsel, + struct thread *thread, + struct perf_sample *sample, + struct symbol **parent); + +/* + * Default guest kernel is defined by parameter --guestkallsyms + * and --guestmodules + */ +static inline bool machine__is_default_guest(struct machine *machine) +{ + return machine ? machine->pid == DEFAULT_GUEST_KERNEL_ID : false; +} + +static inline bool machine__is_host(struct machine *machine) +{ + return machine ? machine->pid == HOST_KERNEL_ID : false; +} + +struct thread *machine__findnew_thread(struct machine *machine, pid_t pid); +void machine__remove_thread(struct machine *machine, struct thread *th); + +size_t machine__fprintf(struct machine *machine, FILE *fp); + +static inline +struct symbol *machine__find_kernel_symbol(struct machine *machine, + enum map_type type, u64 addr, + struct map **mapp, + symbol_filter_t filter) +{ + return map_groups__find_symbol(&machine->kmaps, type, addr, + mapp, filter); +} + +static inline +struct symbol *machine__find_kernel_function(struct machine *machine, u64 addr, + struct map **mapp, + symbol_filter_t filter) +{ + return machine__find_kernel_symbol(machine, MAP__FUNCTION, addr, + mapp, filter); +} + +static inline +struct symbol *machine__find_kernel_function_by_name(struct machine *machine, + const char *name, + struct map **mapp, + symbol_filter_t filter) +{ + return map_groups__find_function_by_name(&machine->kmaps, name, mapp, + filter); +} + +struct map *machine__new_module(struct machine *machine, u64 start, + const char *filename); + +int machine__load_kallsyms(struct machine *machine, const char *filename, + enum map_type type, symbol_filter_t filter); +int machine__load_vmlinux_path(struct machine *machine, enum map_type type, + symbol_filter_t filter); + +size_t machine__fprintf_dsos_buildid(struct machine *machine, FILE *fp, + bool (skip)(struct dso *dso, int parm), int parm); +size_t machines__fprintf_dsos(struct machines *machines, FILE *fp); +size_t machines__fprintf_dsos_buildid(struct machines *machines, FILE *fp, + bool (skip)(struct dso *dso, int parm), int parm); + +void machine__destroy_kernel_maps(struct machine *machine); +int __machine__create_kernel_maps(struct machine *machine, struct dso *kernel); +int machine__create_kernel_maps(struct machine *machine); + +int machines__create_kernel_maps(struct machines *machines, pid_t pid); +int machines__create_guest_kernel_maps(struct machines *machines); +void machines__destroy_kernel_maps(struct machines *machines); + +size_t machine__fprintf_vmlinux_path(struct machine *machine, FILE *fp); + +#endif /* __PERF_MACHINE_H */ diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c new file mode 100644 index 00000000..6fcb9de6 --- /dev/null +++ b/tools/perf/util/map.c @@ -0,0 +1,587 @@ +#include "symbol.h" +#include <errno.h> +#include <inttypes.h> +#include <limits.h> +#include <stdlib.h> +#include <string.h> +#include <stdio.h> +#include <unistd.h> +#include "map.h" +#include "thread.h" +#include "strlist.h" +#include "vdso.h" +#include "build-id.h" +#include <linux/string.h> + +const char *map_type__name[MAP__NR_TYPES] = { + [MAP__FUNCTION] = "Functions", + [MAP__VARIABLE] = "Variables", +}; + +static inline int is_anon_memory(const char *filename) +{ + return !strcmp(filename, "//anon") || + !strcmp(filename, "/anon_hugepage (deleted)"); +} + +static inline int is_no_dso_memory(const char *filename) +{ + return !strncmp(filename, "[stack", 6) || + !strcmp(filename, "[heap]"); +} + +void map__init(struct map *map, enum map_type type, + u64 start, u64 end, u64 pgoff, struct dso *dso) +{ + map->type = type; + map->start = start; + map->end = end; + map->pgoff = pgoff; + map->dso = dso; + map->map_ip = map__map_ip; + map->unmap_ip = map__unmap_ip; + RB_CLEAR_NODE(&map->rb_node); + map->groups = NULL; + map->referenced = false; + map->erange_warned = false; +} + +struct map *map__new(struct list_head *dsos__list, u64 start, u64 len, + u64 pgoff, u32 pid, char *filename, + enum map_type type) +{ + struct map *map = malloc(sizeof(*map)); + + if (map != NULL) { + char newfilename[PATH_MAX]; + struct dso *dso; + int anon, no_dso, vdso; + + anon = is_anon_memory(filename); + vdso = is_vdso_map(filename); + no_dso = is_no_dso_memory(filename); + + if (anon) { + snprintf(newfilename, sizeof(newfilename), "/tmp/perf-%d.map", pid); + filename = newfilename; + } + + if (vdso) { + pgoff = 0; + dso = vdso__dso_findnew(dsos__list); + } else + dso = __dsos__findnew(dsos__list, filename); + + if (dso == NULL) + goto out_delete; + + map__init(map, type, start, start + len, pgoff, dso); + + if (anon || no_dso) { + map->map_ip = map->unmap_ip = identity__map_ip; + + /* + * Set memory without DSO as loaded. All map__find_* + * functions still return NULL, and we avoid the + * unnecessary map__load warning. + */ + if (no_dso) + dso__set_loaded(dso, map->type); + } + } + return map; +out_delete: + free(map); + return NULL; +} + +/* + * Constructor variant for modules (where we know from /proc/modules where + * they are loaded) and for vmlinux, where only after we load all the + * symbols we'll know where it starts and ends. + */ +struct map *map__new2(u64 start, struct dso *dso, enum map_type type) +{ + struct map *map = calloc(1, (sizeof(*map) + + (dso->kernel ? sizeof(struct kmap) : 0))); + if (map != NULL) { + /* + * ->end will be filled after we load all the symbols + */ + map__init(map, type, start, 0, 0, dso); + } + + return map; +} + +void map__delete(struct map *map) +{ + free(map); +} + +void map__fixup_start(struct map *map) +{ + struct rb_root *symbols = &map->dso->symbols[map->type]; + struct rb_node *nd = rb_first(symbols); + if (nd != NULL) { + struct symbol *sym = rb_entry(nd, struct symbol, rb_node); + map->start = sym->start; + } +} + +void map__fixup_end(struct map *map) +{ + struct rb_root *symbols = &map->dso->symbols[map->type]; + struct rb_node *nd = rb_last(symbols); + if (nd != NULL) { + struct symbol *sym = rb_entry(nd, struct symbol, rb_node); + map->end = sym->end; + } +} + +#define DSO__DELETED "(deleted)" + +int map__load(struct map *map, symbol_filter_t filter) +{ + const char *name = map->dso->long_name; + int nr; + + if (dso__loaded(map->dso, map->type)) + return 0; + + nr = dso__load(map->dso, map, filter); + if (nr < 0) { + if (map->dso->has_build_id) { + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + build_id__sprintf(map->dso->build_id, + sizeof(map->dso->build_id), + sbuild_id); + pr_warning("%s with build id %s not found", + name, sbuild_id); + } else + pr_warning("Failed to open %s", name); + + pr_warning(", continuing without symbols\n"); + return -1; + } else if (nr == 0) { +#ifdef LIBELF_SUPPORT + const size_t len = strlen(name); + const size_t real_len = len - sizeof(DSO__DELETED); + + if (len > sizeof(DSO__DELETED) && + strcmp(name + real_len + 1, DSO__DELETED) == 0) { + pr_warning("%.*s was updated (is prelink enabled?). " + "Restart the long running apps that use it!\n", + (int)real_len, name); + } else { + pr_warning("no symbols found in %s, maybe install " + "a debug package?\n", name); + } +#endif + return -1; + } + /* + * Only applies to the kernel, as its symtabs aren't relative like the + * module ones. + */ + if (map->dso->kernel) + map__reloc_vmlinux(map); + + return 0; +} + +struct symbol *map__find_symbol(struct map *map, u64 addr, + symbol_filter_t filter) +{ + if (map__load(map, filter) < 0) + return NULL; + + return dso__find_symbol(map->dso, map->type, addr); +} + +struct symbol *map__find_symbol_by_name(struct map *map, const char *name, + symbol_filter_t filter) +{ + if (map__load(map, filter) < 0) + return NULL; + + if (!dso__sorted_by_name(map->dso, map->type)) + dso__sort_by_name(map->dso, map->type); + + return dso__find_symbol_by_name(map->dso, map->type, name); +} + +struct map *map__clone(struct map *map) +{ + return memdup(map, sizeof(*map)); +} + +int map__overlap(struct map *l, struct map *r) +{ + if (l->start > r->start) { + struct map *t = l; + l = r; + r = t; + } + + if (l->end > r->start) + return 1; + + return 0; +} + +size_t map__fprintf(struct map *map, FILE *fp) +{ + return fprintf(fp, " %" PRIx64 "-%" PRIx64 " %" PRIx64 " %s\n", + map->start, map->end, map->pgoff, map->dso->name); +} + +size_t map__fprintf_dsoname(struct map *map, FILE *fp) +{ + const char *dsoname = "[unknown]"; + + if (map && map->dso && (map->dso->name || map->dso->long_name)) { + if (symbol_conf.show_kernel_path && map->dso->long_name) + dsoname = map->dso->long_name; + else if (map->dso->name) + dsoname = map->dso->name; + } + + return fprintf(fp, "%s", dsoname); +} + +/* + * objdump wants/reports absolute IPs for ET_EXEC, and RIPs for ET_DYN. + * map->dso->adjust_symbols==1 for ET_EXEC-like cases. + */ +u64 map__rip_2objdump(struct map *map, u64 rip) +{ + u64 addr = map->dso->adjust_symbols ? + map->unmap_ip(map, rip) : /* RIP -> IP */ + rip; + return addr; +} + +void map_groups__init(struct map_groups *mg) +{ + int i; + for (i = 0; i < MAP__NR_TYPES; ++i) { + mg->maps[i] = RB_ROOT; + INIT_LIST_HEAD(&mg->removed_maps[i]); + } + mg->machine = NULL; +} + +static void maps__delete(struct rb_root *maps) +{ + struct rb_node *next = rb_first(maps); + + while (next) { + struct map *pos = rb_entry(next, struct map, rb_node); + + next = rb_next(&pos->rb_node); + rb_erase(&pos->rb_node, maps); + map__delete(pos); + } +} + +static void maps__delete_removed(struct list_head *maps) +{ + struct map *pos, *n; + + list_for_each_entry_safe(pos, n, maps, node) { + list_del(&pos->node); + map__delete(pos); + } +} + +void map_groups__exit(struct map_groups *mg) +{ + int i; + + for (i = 0; i < MAP__NR_TYPES; ++i) { + maps__delete(&mg->maps[i]); + maps__delete_removed(&mg->removed_maps[i]); + } +} + +void map_groups__flush(struct map_groups *mg) +{ + int type; + + for (type = 0; type < MAP__NR_TYPES; type++) { + struct rb_root *root = &mg->maps[type]; + struct rb_node *next = rb_first(root); + + while (next) { + struct map *pos = rb_entry(next, struct map, rb_node); + next = rb_next(&pos->rb_node); + rb_erase(&pos->rb_node, root); + /* + * We may have references to this map, for + * instance in some hist_entry instances, so + * just move them to a separate list. + */ + list_add_tail(&pos->node, &mg->removed_maps[pos->type]); + } + } +} + +struct symbol *map_groups__find_symbol(struct map_groups *mg, + enum map_type type, u64 addr, + struct map **mapp, + symbol_filter_t filter) +{ + struct map *map = map_groups__find(mg, type, addr); + + if (map != NULL) { + if (mapp != NULL) + *mapp = map; + return map__find_symbol(map, map->map_ip(map, addr), filter); + } + + return NULL; +} + +struct symbol *map_groups__find_symbol_by_name(struct map_groups *mg, + enum map_type type, + const char *name, + struct map **mapp, + symbol_filter_t filter) +{ + struct rb_node *nd; + + for (nd = rb_first(&mg->maps[type]); nd; nd = rb_next(nd)) { + struct map *pos = rb_entry(nd, struct map, rb_node); + struct symbol *sym = map__find_symbol_by_name(pos, name, filter); + + if (sym == NULL) + continue; + if (mapp != NULL) + *mapp = pos; + return sym; + } + + return NULL; +} + +size_t __map_groups__fprintf_maps(struct map_groups *mg, + enum map_type type, int verbose, FILE *fp) +{ + size_t printed = fprintf(fp, "%s:\n", map_type__name[type]); + struct rb_node *nd; + + for (nd = rb_first(&mg->maps[type]); nd; nd = rb_next(nd)) { + struct map *pos = rb_entry(nd, struct map, rb_node); + printed += fprintf(fp, "Map:"); + printed += map__fprintf(pos, fp); + if (verbose > 2) { + printed += dso__fprintf(pos->dso, type, fp); + printed += fprintf(fp, "--\n"); + } + } + + return printed; +} + +size_t map_groups__fprintf_maps(struct map_groups *mg, int verbose, FILE *fp) +{ + size_t printed = 0, i; + for (i = 0; i < MAP__NR_TYPES; ++i) + printed += __map_groups__fprintf_maps(mg, i, verbose, fp); + return printed; +} + +static size_t __map_groups__fprintf_removed_maps(struct map_groups *mg, + enum map_type type, + int verbose, FILE *fp) +{ + struct map *pos; + size_t printed = 0; + + list_for_each_entry(pos, &mg->removed_maps[type], node) { + printed += fprintf(fp, "Map:"); + printed += map__fprintf(pos, fp); + if (verbose > 1) { + printed += dso__fprintf(pos->dso, type, fp); + printed += fprintf(fp, "--\n"); + } + } + return printed; +} + +static size_t map_groups__fprintf_removed_maps(struct map_groups *mg, + int verbose, FILE *fp) +{ + size_t printed = 0, i; + for (i = 0; i < MAP__NR_TYPES; ++i) + printed += __map_groups__fprintf_removed_maps(mg, i, verbose, fp); + return printed; +} + +size_t map_groups__fprintf(struct map_groups *mg, int verbose, FILE *fp) +{ + size_t printed = map_groups__fprintf_maps(mg, verbose, fp); + printed += fprintf(fp, "Removed maps:\n"); + return printed + map_groups__fprintf_removed_maps(mg, verbose, fp); +} + +int map_groups__fixup_overlappings(struct map_groups *mg, struct map *map, + int verbose, FILE *fp) +{ + struct rb_root *root = &mg->maps[map->type]; + struct rb_node *next = rb_first(root); + int err = 0; + + while (next) { + struct map *pos = rb_entry(next, struct map, rb_node); + next = rb_next(&pos->rb_node); + + if (!map__overlap(pos, map)) + continue; + + if (verbose >= 2) { + fputs("overlapping maps:\n", fp); + map__fprintf(map, fp); + map__fprintf(pos, fp); + } + + rb_erase(&pos->rb_node, root); + /* + * Now check if we need to create new maps for areas not + * overlapped by the new map: + */ + if (map->start > pos->start) { + struct map *before = map__clone(pos); + + if (before == NULL) { + err = -ENOMEM; + goto move_map; + } + + before->end = map->start - 1; + map_groups__insert(mg, before); + if (verbose >= 2) + map__fprintf(before, fp); + } + + if (map->end < pos->end) { + struct map *after = map__clone(pos); + + if (after == NULL) { + err = -ENOMEM; + goto move_map; + } + + after->start = map->end + 1; + map_groups__insert(mg, after); + if (verbose >= 2) + map__fprintf(after, fp); + } +move_map: + /* + * If we have references, just move them to a separate list. + */ + if (pos->referenced) + list_add_tail(&pos->node, &mg->removed_maps[map->type]); + else + map__delete(pos); + + if (err) + return err; + } + + return 0; +} + +/* + * XXX This should not really _copy_ te maps, but refcount them. + */ +int map_groups__clone(struct map_groups *mg, + struct map_groups *parent, enum map_type type) +{ + struct rb_node *nd; + for (nd = rb_first(&parent->maps[type]); nd; nd = rb_next(nd)) { + struct map *map = rb_entry(nd, struct map, rb_node); + struct map *new = map__clone(map); + if (new == NULL) + return -ENOMEM; + map_groups__insert(mg, new); + } + return 0; +} + +static u64 map__reloc_map_ip(struct map *map, u64 ip) +{ + return ip + (s64)map->pgoff; +} + +static u64 map__reloc_unmap_ip(struct map *map, u64 ip) +{ + return ip - (s64)map->pgoff; +} + +void map__reloc_vmlinux(struct map *map) +{ + struct kmap *kmap = map__kmap(map); + s64 reloc; + + if (!kmap->ref_reloc_sym || !kmap->ref_reloc_sym->unrelocated_addr) + return; + + reloc = (kmap->ref_reloc_sym->unrelocated_addr - + kmap->ref_reloc_sym->addr); + + if (!reloc) + return; + + map->map_ip = map__reloc_map_ip; + map->unmap_ip = map__reloc_unmap_ip; + map->pgoff = reloc; +} + +void maps__insert(struct rb_root *maps, struct map *map) +{ + struct rb_node **p = &maps->rb_node; + struct rb_node *parent = NULL; + const u64 ip = map->start; + struct map *m; + + while (*p != NULL) { + parent = *p; + m = rb_entry(parent, struct map, rb_node); + if (ip < m->start) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + + rb_link_node(&map->rb_node, parent, p); + rb_insert_color(&map->rb_node, maps); +} + +void maps__remove(struct rb_root *maps, struct map *map) +{ + rb_erase(&map->rb_node, maps); +} + +struct map *maps__find(struct rb_root *maps, u64 ip) +{ + struct rb_node **p = &maps->rb_node; + struct rb_node *parent = NULL; + struct map *m; + + while (*p != NULL) { + parent = *p; + m = rb_entry(parent, struct map, rb_node); + if (ip < m->start) + p = &(*p)->rb_left; + else if (ip > m->end) + p = &(*p)->rb_right; + else + return m; + } + + return NULL; +} diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h new file mode 100644 index 00000000..a887f2c9 --- /dev/null +++ b/tools/perf/util/map.h @@ -0,0 +1,169 @@ +#ifndef __PERF_MAP_H +#define __PERF_MAP_H + +#include <linux/compiler.h> +#include <linux/list.h> +#include <linux/rbtree.h> +#include <stdio.h> +#include <stdbool.h> +#include "types.h" + +enum map_type { + MAP__FUNCTION = 0, + MAP__VARIABLE, +}; + +#define MAP__NR_TYPES (MAP__VARIABLE + 1) + +extern const char *map_type__name[MAP__NR_TYPES]; + +struct dso; +struct ip_callchain; +struct ref_reloc_sym; +struct map_groups; +struct machine; +struct perf_evsel; + +struct map { + union { + struct rb_node rb_node; + struct list_head node; + }; + u64 start; + u64 end; + u8 /* enum map_type */ type; + bool referenced; + bool erange_warned; + u32 priv; + u64 pgoff; + + /* ip -> dso rip */ + u64 (*map_ip)(struct map *, u64); + /* dso rip -> ip */ + u64 (*unmap_ip)(struct map *, u64); + + struct dso *dso; + struct map_groups *groups; +}; + +struct kmap { + struct ref_reloc_sym *ref_reloc_sym; + struct map_groups *kmaps; +}; + +struct map_groups { + struct rb_root maps[MAP__NR_TYPES]; + struct list_head removed_maps[MAP__NR_TYPES]; + struct machine *machine; +}; + +static inline struct kmap *map__kmap(struct map *map) +{ + return (struct kmap *)(map + 1); +} + +static inline u64 map__map_ip(struct map *map, u64 ip) +{ + return ip - map->start + map->pgoff; +} + +static inline u64 map__unmap_ip(struct map *map, u64 ip) +{ + return ip + map->start - map->pgoff; +} + +static inline u64 identity__map_ip(struct map *map __maybe_unused, u64 ip) +{ + return ip; +} + + +/* rip/ip <-> addr suitable for passing to `objdump --start-address=` */ +u64 map__rip_2objdump(struct map *map, u64 rip); + +struct symbol; + +typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym); + +void map__init(struct map *map, enum map_type type, + u64 start, u64 end, u64 pgoff, struct dso *dso); +struct map *map__new(struct list_head *dsos__list, u64 start, u64 len, + u64 pgoff, u32 pid, char *filename, + enum map_type type); +struct map *map__new2(u64 start, struct dso *dso, enum map_type type); +void map__delete(struct map *map); +struct map *map__clone(struct map *map); +int map__overlap(struct map *l, struct map *r); +size_t map__fprintf(struct map *map, FILE *fp); +size_t map__fprintf_dsoname(struct map *map, FILE *fp); + +int map__load(struct map *map, symbol_filter_t filter); +struct symbol *map__find_symbol(struct map *map, + u64 addr, symbol_filter_t filter); +struct symbol *map__find_symbol_by_name(struct map *map, const char *name, + symbol_filter_t filter); +void map__fixup_start(struct map *map); +void map__fixup_end(struct map *map); + +void map__reloc_vmlinux(struct map *map); + +size_t __map_groups__fprintf_maps(struct map_groups *mg, + enum map_type type, int verbose, FILE *fp); +void maps__insert(struct rb_root *maps, struct map *map); +void maps__remove(struct rb_root *maps, struct map *map); +struct map *maps__find(struct rb_root *maps, u64 addr); +void map_groups__init(struct map_groups *mg); +void map_groups__exit(struct map_groups *mg); +int map_groups__clone(struct map_groups *mg, + struct map_groups *parent, enum map_type type); +size_t map_groups__fprintf(struct map_groups *mg, int verbose, FILE *fp); +size_t map_groups__fprintf_maps(struct map_groups *mg, int verbose, FILE *fp); + +int maps__set_kallsyms_ref_reloc_sym(struct map **maps, const char *symbol_name, + u64 addr); + +static inline void map_groups__insert(struct map_groups *mg, struct map *map) +{ + maps__insert(&mg->maps[map->type], map); + map->groups = mg; +} + +static inline void map_groups__remove(struct map_groups *mg, struct map *map) +{ + maps__remove(&mg->maps[map->type], map); +} + +static inline struct map *map_groups__find(struct map_groups *mg, + enum map_type type, u64 addr) +{ + return maps__find(&mg->maps[type], addr); +} + +struct symbol *map_groups__find_symbol(struct map_groups *mg, + enum map_type type, u64 addr, + struct map **mapp, + symbol_filter_t filter); + +struct symbol *map_groups__find_symbol_by_name(struct map_groups *mg, + enum map_type type, + const char *name, + struct map **mapp, + symbol_filter_t filter); + +static inline +struct symbol *map_groups__find_function_by_name(struct map_groups *mg, + const char *name, struct map **mapp, + symbol_filter_t filter) +{ + return map_groups__find_symbol_by_name(mg, MAP__FUNCTION, name, mapp, filter); +} + +int map_groups__fixup_overlappings(struct map_groups *mg, struct map *map, + int verbose, FILE *fp); + +struct map *map_groups__find_by_name(struct map_groups *mg, + enum map_type type, const char *name); + +void map_groups__flush(struct map_groups *mg); + +#endif /* __PERF_MAP_H */ diff --git a/tools/perf/util/pager.c b/tools/perf/util/pager.c new file mode 100644 index 00000000..3322b844 --- /dev/null +++ b/tools/perf/util/pager.c @@ -0,0 +1,100 @@ +#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; + +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); +} + +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) { + if (!access("/usr/bin/pager", X_OK)) + pager = "/usr/bin/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; + pager_process.preexec_cb = pager_preexec; + + 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 00000000..c84f48cf --- /dev/null +++ b/tools/perf/util/parse-events.c @@ -0,0 +1,1240 @@ +#include <linux/hw_breakpoint.h> +#include "util.h" +#include "../perf.h" +#include "evlist.h" +#include "evsel.h" +#include "parse-options.h" +#include "parse-events.h" +#include "exec_cmd.h" +#include "string.h" +#include "symbol.h" +#include "cache.h" +#include "header.h" +#include "debugfs.h" +#include "parse-events-bison.h" +#define YY_EXTRA_TYPE int +#include "parse-events-flex.h" +#include "pmu.h" + +#define MAX_NAME_LEN 100 + +struct event_symbol { + const char *symbol; + const char *alias; +}; + +#ifdef PARSER_DEBUG +extern int parse_events_debug; +#endif +int parse_events_parse(void *data, void *scanner); + +static struct event_symbol event_symbols_hw[PERF_COUNT_HW_MAX] = { + [PERF_COUNT_HW_CPU_CYCLES] = { + .symbol = "cpu-cycles", + .alias = "cycles", + }, + [PERF_COUNT_HW_INSTRUCTIONS] = { + .symbol = "instructions", + .alias = "", + }, + [PERF_COUNT_HW_CACHE_REFERENCES] = { + .symbol = "cache-references", + .alias = "", + }, + [PERF_COUNT_HW_CACHE_MISSES] = { + .symbol = "cache-misses", + .alias = "", + }, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { + .symbol = "branch-instructions", + .alias = "branches", + }, + [PERF_COUNT_HW_BRANCH_MISSES] = { + .symbol = "branch-misses", + .alias = "", + }, + [PERF_COUNT_HW_BUS_CYCLES] = { + .symbol = "bus-cycles", + .alias = "", + }, + [PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = { + .symbol = "stalled-cycles-frontend", + .alias = "idle-cycles-frontend", + }, + [PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = { + .symbol = "stalled-cycles-backend", + .alias = "idle-cycles-backend", + }, + [PERF_COUNT_HW_REF_CPU_CYCLES] = { + .symbol = "ref-cycles", + .alias = "", + }, +}; + +static struct event_symbol event_symbols_sw[PERF_COUNT_SW_MAX] = { + [PERF_COUNT_SW_CPU_CLOCK] = { + .symbol = "cpu-clock", + .alias = "", + }, + [PERF_COUNT_SW_TASK_CLOCK] = { + .symbol = "task-clock", + .alias = "", + }, + [PERF_COUNT_SW_PAGE_FAULTS] = { + .symbol = "page-faults", + .alias = "faults", + }, + [PERF_COUNT_SW_CONTEXT_SWITCHES] = { + .symbol = "context-switches", + .alias = "cs", + }, + [PERF_COUNT_SW_CPU_MIGRATIONS] = { + .symbol = "cpu-migrations", + .alias = "migrations", + }, + [PERF_COUNT_SW_PAGE_FAULTS_MIN] = { + .symbol = "minor-faults", + .alias = "", + }, + [PERF_COUNT_SW_PAGE_FAULTS_MAJ] = { + .symbol = "major-faults", + .alias = "", + }, + [PERF_COUNT_SW_ALIGNMENT_FAULTS] = { + .symbol = "alignment-faults", + .alias = "", + }, + [PERF_COUNT_SW_EMULATION_FAULTS] = { + .symbol = "emulation-faults", + .alias = "", + }, +}; + +#define __PERF_EVENT_FIELD(config, name) \ + ((config & PERF_EVENT_##name##_MASK) >> PERF_EVENT_##name##_SHIFT) + +#define PERF_EVENT_RAW(config) __PERF_EVENT_FIELD(config, RAW) +#define PERF_EVENT_CONFIG(config) __PERF_EVENT_FIELD(config, CONFIG) +#define PERF_EVENT_TYPE(config) __PERF_EVENT_FIELD(config, TYPE) +#define PERF_EVENT_ID(config) __PERF_EVENT_FIELD(config, EVENT) + +#define for_each_subsystem(sys_dir, sys_dirent, sys_next) \ + while (!readdir_r(sys_dir, &sys_dirent, &sys_next) && sys_next) \ + if (sys_dirent.d_type == DT_DIR && \ + (strcmp(sys_dirent.d_name, ".")) && \ + (strcmp(sys_dirent.d_name, ".."))) + +static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir) +{ + char evt_path[MAXPATHLEN]; + int fd; + + snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", tracing_events_path, + sys_dir->d_name, evt_dir->d_name); + fd = open(evt_path, O_RDONLY); + if (fd < 0) + return -EINVAL; + close(fd); + + return 0; +} + +#define for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) \ + while (!readdir_r(evt_dir, &evt_dirent, &evt_next) && evt_next) \ + if (evt_dirent.d_type == DT_DIR && \ + (strcmp(evt_dirent.d_name, ".")) && \ + (strcmp(evt_dirent.d_name, "..")) && \ + (!tp_event_has_id(&sys_dirent, &evt_dirent))) + +#define MAX_EVENT_LENGTH 512 + + +struct tracepoint_path *tracepoint_id_to_path(u64 config) +{ + struct tracepoint_path *path = NULL; + DIR *sys_dir, *evt_dir; + struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent; + char id_buf[24]; + int fd; + u64 id; + char evt_path[MAXPATHLEN]; + char dir_path[MAXPATHLEN]; + + if (debugfs_valid_mountpoint(tracing_events_path)) + return NULL; + + sys_dir = opendir(tracing_events_path); + if (!sys_dir) + return NULL; + + for_each_subsystem(sys_dir, sys_dirent, sys_next) { + + snprintf(dir_path, MAXPATHLEN, "%s/%s", tracing_events_path, + sys_dirent.d_name); + evt_dir = opendir(dir_path); + if (!evt_dir) + continue; + + for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) { + + snprintf(evt_path, MAXPATHLEN, "%s/%s/id", dir_path, + evt_dirent.d_name); + fd = open(evt_path, O_RDONLY); + if (fd < 0) + continue; + if (read(fd, id_buf, sizeof(id_buf)) < 0) { + close(fd); + continue; + } + close(fd); + id = atoll(id_buf); + if (id == config) { + closedir(evt_dir); + closedir(sys_dir); + path = zalloc(sizeof(*path)); + path->system = malloc(MAX_EVENT_LENGTH); + if (!path->system) { + free(path); + return NULL; + } + path->name = malloc(MAX_EVENT_LENGTH); + if (!path->name) { + free(path->system); + free(path); + return NULL; + } + strncpy(path->system, sys_dirent.d_name, + MAX_EVENT_LENGTH); + strncpy(path->name, evt_dirent.d_name, + MAX_EVENT_LENGTH); + return path; + } + } + closedir(evt_dir); + } + + closedir(sys_dir); + return NULL; +} + +const char *event_type(int type) +{ + switch (type) { + case PERF_TYPE_HARDWARE: + return "hardware"; + + case PERF_TYPE_SOFTWARE: + return "software"; + + case PERF_TYPE_TRACEPOINT: + return "tracepoint"; + + case PERF_TYPE_HW_CACHE: + return "hardware-cache"; + + default: + break; + } + + return "unknown"; +} + + + +static int __add_event(struct list_head **_list, int *idx, + struct perf_event_attr *attr, + char *name, struct cpu_map *cpus) +{ + struct perf_evsel *evsel; + struct list_head *list = *_list; + + if (!list) { + list = malloc(sizeof(*list)); + if (!list) + return -ENOMEM; + INIT_LIST_HEAD(list); + } + + event_attr_init(attr); + + evsel = perf_evsel__new(attr, (*idx)++); + if (!evsel) { + free(list); + return -ENOMEM; + } + + evsel->cpus = cpus; + if (name) + evsel->name = strdup(name); + list_add_tail(&evsel->node, list); + *_list = list; + return 0; +} + +static int add_event(struct list_head **_list, int *idx, + struct perf_event_attr *attr, char *name) +{ + return __add_event(_list, idx, attr, name, NULL); +} + +static int parse_aliases(char *str, const char *names[][PERF_EVSEL__MAX_ALIASES], int size) +{ + int i, j; + int n, longest = -1; + + for (i = 0; i < size; i++) { + for (j = 0; j < PERF_EVSEL__MAX_ALIASES && names[i][j]; j++) { + n = strlen(names[i][j]); + if (n > longest && !strncasecmp(str, names[i][j], n)) + longest = n; + } + if (longest > 0) + return i; + } + + return -1; +} + +int parse_events_add_cache(struct list_head **list, int *idx, + char *type, char *op_result1, char *op_result2) +{ + struct perf_event_attr attr; + char name[MAX_NAME_LEN]; + int cache_type = -1, cache_op = -1, cache_result = -1; + char *op_result[2] = { op_result1, op_result2 }; + int i, n; + + /* + * No fallback - if we cannot get a clear cache type + * then bail out: + */ + cache_type = parse_aliases(type, perf_evsel__hw_cache, + PERF_COUNT_HW_CACHE_MAX); + if (cache_type == -1) + return -EINVAL; + + n = snprintf(name, MAX_NAME_LEN, "%s", type); + + for (i = 0; (i < 2) && (op_result[i]); i++) { + char *str = op_result[i]; + + n += snprintf(name + n, MAX_NAME_LEN - n, "-%s", str); + + if (cache_op == -1) { + cache_op = parse_aliases(str, perf_evsel__hw_cache_op, + PERF_COUNT_HW_CACHE_OP_MAX); + if (cache_op >= 0) { + if (!perf_evsel__is_cache_op_valid(cache_type, cache_op)) + return -EINVAL; + continue; + } + } + + if (cache_result == -1) { + cache_result = parse_aliases(str, perf_evsel__hw_cache_result, + PERF_COUNT_HW_CACHE_RESULT_MAX); + if (cache_result >= 0) + continue; + } + } + + /* + * Fall back to reads: + */ + if (cache_op == -1) + cache_op = PERF_COUNT_HW_CACHE_OP_READ; + + /* + * Fall back to accesses: + */ + if (cache_result == -1) + cache_result = PERF_COUNT_HW_CACHE_RESULT_ACCESS; + + memset(&attr, 0, sizeof(attr)); + attr.config = cache_type | (cache_op << 8) | (cache_result << 16); + attr.type = PERF_TYPE_HW_CACHE; + return add_event(list, idx, &attr, name); +} + +static int add_tracepoint(struct list_head **listp, int *idx, + char *sys_name, char *evt_name) +{ + struct perf_evsel *evsel; + struct list_head *list = *listp; + + if (!list) { + list = malloc(sizeof(*list)); + if (!list) + return -ENOMEM; + INIT_LIST_HEAD(list); + } + + evsel = perf_evsel__newtp(sys_name, evt_name, (*idx)++); + if (!evsel) { + free(list); + return -ENOMEM; + } + + list_add_tail(&evsel->node, list); + *listp = list; + return 0; +} + +static int add_tracepoint_multi_event(struct list_head **list, int *idx, + char *sys_name, char *evt_name) +{ + char evt_path[MAXPATHLEN]; + struct dirent *evt_ent; + DIR *evt_dir; + int ret = 0; + + snprintf(evt_path, MAXPATHLEN, "%s/%s", tracing_events_path, sys_name); + evt_dir = opendir(evt_path); + if (!evt_dir) { + perror("Can't open event dir"); + return -1; + } + + while (!ret && (evt_ent = readdir(evt_dir))) { + if (!strcmp(evt_ent->d_name, ".") + || !strcmp(evt_ent->d_name, "..") + || !strcmp(evt_ent->d_name, "enable") + || !strcmp(evt_ent->d_name, "filter")) + continue; + + if (!strglobmatch(evt_ent->d_name, evt_name)) + continue; + + ret = add_tracepoint(list, idx, sys_name, evt_ent->d_name); + } + + closedir(evt_dir); + return ret; +} + +static int add_tracepoint_event(struct list_head **list, int *idx, + char *sys_name, char *evt_name) +{ + return strpbrk(evt_name, "*?") ? + add_tracepoint_multi_event(list, idx, sys_name, evt_name) : + add_tracepoint(list, idx, sys_name, evt_name); +} + +static int add_tracepoint_multi_sys(struct list_head **list, int *idx, + char *sys_name, char *evt_name) +{ + struct dirent *events_ent; + DIR *events_dir; + int ret = 0; + + events_dir = opendir(tracing_events_path); + if (!events_dir) { + perror("Can't open event dir"); + return -1; + } + + while (!ret && (events_ent = readdir(events_dir))) { + if (!strcmp(events_ent->d_name, ".") + || !strcmp(events_ent->d_name, "..") + || !strcmp(events_ent->d_name, "enable") + || !strcmp(events_ent->d_name, "header_event") + || !strcmp(events_ent->d_name, "header_page")) + continue; + + if (!strglobmatch(events_ent->d_name, sys_name)) + continue; + + ret = add_tracepoint_event(list, idx, events_ent->d_name, + evt_name); + } + + closedir(events_dir); + return ret; +} + +int parse_events_add_tracepoint(struct list_head **list, int *idx, + char *sys, char *event) +{ + int ret; + + ret = debugfs_valid_mountpoint(tracing_events_path); + if (ret) + return ret; + + if (strpbrk(sys, "*?")) + return add_tracepoint_multi_sys(list, idx, sys, event); + else + return add_tracepoint_event(list, idx, sys, event); +} + +static int +parse_breakpoint_type(const char *type, struct perf_event_attr *attr) +{ + int i; + + for (i = 0; i < 3; i++) { + if (!type || !type[i]) + break; + +#define CHECK_SET_TYPE(bit) \ +do { \ + if (attr->bp_type & bit) \ + return -EINVAL; \ + else \ + attr->bp_type |= bit; \ +} while (0) + + switch (type[i]) { + case 'r': + CHECK_SET_TYPE(HW_BREAKPOINT_R); + break; + case 'w': + CHECK_SET_TYPE(HW_BREAKPOINT_W); + break; + case 'x': + CHECK_SET_TYPE(HW_BREAKPOINT_X); + break; + default: + return -EINVAL; + } + } + +#undef CHECK_SET_TYPE + + if (!attr->bp_type) /* Default */ + attr->bp_type = HW_BREAKPOINT_R | HW_BREAKPOINT_W; + + return 0; +} + +int parse_events_add_breakpoint(struct list_head **list, int *idx, + void *ptr, char *type) +{ + struct perf_event_attr attr; + + memset(&attr, 0, sizeof(attr)); + attr.bp_addr = (unsigned long) ptr; + + if (parse_breakpoint_type(type, &attr)) + return -EINVAL; + + /* + * We should find a nice way to override the access length + * Provide some defaults for now + */ + if (attr.bp_type == HW_BREAKPOINT_X) + attr.bp_len = sizeof(long); + else + attr.bp_len = HW_BREAKPOINT_LEN_4; + + attr.type = PERF_TYPE_BREAKPOINT; + attr.sample_period = 1; + + return add_event(list, idx, &attr, NULL); +} + +static int config_term(struct perf_event_attr *attr, + struct parse_events_term *term) +{ +#define CHECK_TYPE_VAL(type) \ +do { \ + if (PARSE_EVENTS__TERM_TYPE_ ## type != term->type_val) \ + return -EINVAL; \ +} while (0) + + switch (term->type_term) { + case PARSE_EVENTS__TERM_TYPE_CONFIG: + CHECK_TYPE_VAL(NUM); + attr->config = term->val.num; + break; + case PARSE_EVENTS__TERM_TYPE_CONFIG1: + CHECK_TYPE_VAL(NUM); + attr->config1 = term->val.num; + break; + case PARSE_EVENTS__TERM_TYPE_CONFIG2: + CHECK_TYPE_VAL(NUM); + attr->config2 = term->val.num; + break; + case PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD: + CHECK_TYPE_VAL(NUM); + attr->sample_period = term->val.num; + break; + case PARSE_EVENTS__TERM_TYPE_BRANCH_SAMPLE_TYPE: + /* + * TODO uncomment when the field is available + * attr->branch_sample_type = term->val.num; + */ + break; + case PARSE_EVENTS__TERM_TYPE_NAME: + CHECK_TYPE_VAL(STR); + break; + default: + return -EINVAL; + } + + return 0; +#undef CHECK_TYPE_VAL +} + +static int config_attr(struct perf_event_attr *attr, + struct list_head *head, int fail) +{ + struct parse_events_term *term; + + list_for_each_entry(term, head, list) + if (config_term(attr, term) && fail) + return -EINVAL; + + return 0; +} + +int parse_events_add_numeric(struct list_head **list, int *idx, + u32 type, u64 config, + struct list_head *head_config) +{ + struct perf_event_attr attr; + + memset(&attr, 0, sizeof(attr)); + attr.type = type; + attr.config = config; + + if (head_config && + config_attr(&attr, head_config, 1)) + return -EINVAL; + + return add_event(list, idx, &attr, NULL); +} + +static int parse_events__is_name_term(struct parse_events_term *term) +{ + return term->type_term == PARSE_EVENTS__TERM_TYPE_NAME; +} + +static char *pmu_event_name(struct list_head *head_terms) +{ + struct parse_events_term *term; + + list_for_each_entry(term, head_terms, list) + if (parse_events__is_name_term(term)) + return term->val.str; + + return NULL; +} + +int parse_events_add_pmu(struct list_head **list, int *idx, + char *name, struct list_head *head_config) +{ + struct perf_event_attr attr; + struct perf_pmu *pmu; + + pmu = perf_pmu__find(name); + if (!pmu) + return -EINVAL; + + memset(&attr, 0, sizeof(attr)); + + if (perf_pmu__check_alias(pmu, head_config)) + return -EINVAL; + + /* + * Configure hardcoded terms first, no need to check + * return value when called with fail == 0 ;) + */ + config_attr(&attr, head_config, 0); + + if (perf_pmu__config(pmu, &attr, head_config)) + return -EINVAL; + + return __add_event(list, idx, &attr, pmu_event_name(head_config), + pmu->cpus); +} + +int parse_events__modifier_group(struct list_head *list, + char *event_mod) +{ + return parse_events__modifier_event(list, event_mod, true); +} + +void parse_events__set_leader(char *name, struct list_head *list) +{ + struct perf_evsel *leader; + + __perf_evlist__set_leader(list); + leader = list_entry(list->next, struct perf_evsel, node); + leader->group_name = name ? strdup(name) : NULL; +} + +void parse_events_update_lists(struct list_head *list_event, + struct list_head *list_all) +{ + /* + * Called for single event definition. Update the + * 'all event' list, and reinit the 'single event' + * list, for next event definition. + */ + list_splice_tail(list_event, list_all); + free(list_event); +} + +struct event_modifier { + int eu; + int ek; + int eh; + int eH; + int eG; + int precise; + int exclude_GH; +}; + +static int get_event_modifier(struct event_modifier *mod, char *str, + struct perf_evsel *evsel) +{ + int eu = evsel ? evsel->attr.exclude_user : 0; + int ek = evsel ? evsel->attr.exclude_kernel : 0; + int eh = evsel ? evsel->attr.exclude_hv : 0; + int eH = evsel ? evsel->attr.exclude_host : 0; + int eG = evsel ? evsel->attr.exclude_guest : 0; + int precise = evsel ? evsel->attr.precise_ip : 0; + + int exclude = eu | ek | eh; + int exclude_GH = evsel ? evsel->exclude_GH : 0; + + memset(mod, 0, sizeof(*mod)); + + while (*str) { + if (*str == 'u') { + if (!exclude) + exclude = eu = ek = eh = 1; + eu = 0; + } else if (*str == 'k') { + if (!exclude) + exclude = eu = ek = eh = 1; + ek = 0; + } else if (*str == 'h') { + if (!exclude) + exclude = eu = ek = eh = 1; + eh = 0; + } else if (*str == 'G') { + if (!exclude_GH) + exclude_GH = eG = eH = 1; + eG = 0; + } else if (*str == 'H') { + if (!exclude_GH) + exclude_GH = eG = eH = 1; + eH = 0; + } else if (*str == 'p') { + precise++; + /* use of precise requires exclude_guest */ + if (!exclude_GH) + eG = 1; + } else + break; + + ++str; + } + + /* + * precise ip: + * + * 0 - SAMPLE_IP can have arbitrary skid + * 1 - SAMPLE_IP must have constant skid + * 2 - SAMPLE_IP requested to have 0 skid + * 3 - SAMPLE_IP must have 0 skid + * + * See also PERF_RECORD_MISC_EXACT_IP + */ + if (precise > 3) + return -EINVAL; + + mod->eu = eu; + mod->ek = ek; + mod->eh = eh; + mod->eH = eH; + mod->eG = eG; + mod->precise = precise; + mod->exclude_GH = exclude_GH; + return 0; +} + +/* + * Basic modifier sanity check to validate it contains only one + * instance of any modifier (apart from 'p') present. + */ +static int check_modifier(char *str) +{ + char *p = str; + + /* The sizeof includes 0 byte as well. */ + if (strlen(str) > (sizeof("ukhGHppp") - 1)) + return -1; + + while (*p) { + if (*p != 'p' && strchr(p + 1, *p)) + return -1; + p++; + } + + return 0; +} + +int parse_events__modifier_event(struct list_head *list, char *str, bool add) +{ + struct perf_evsel *evsel; + struct event_modifier mod; + + if (str == NULL) + return 0; + + if (check_modifier(str)) + return -EINVAL; + + if (!add && get_event_modifier(&mod, str, NULL)) + return -EINVAL; + + list_for_each_entry(evsel, list, node) { + + if (add && get_event_modifier(&mod, str, evsel)) + return -EINVAL; + + evsel->attr.exclude_user = mod.eu; + evsel->attr.exclude_kernel = mod.ek; + evsel->attr.exclude_hv = mod.eh; + evsel->attr.precise_ip = mod.precise; + evsel->attr.exclude_host = mod.eH; + evsel->attr.exclude_guest = mod.eG; + evsel->exclude_GH = mod.exclude_GH; + } + + return 0; +} + +int parse_events_name(struct list_head *list, char *name) +{ + struct perf_evsel *evsel; + + list_for_each_entry(evsel, list, node) { + if (!evsel->name) + evsel->name = strdup(name); + } + + return 0; +} + +static int parse_events__scanner(const char *str, void *data, int start_token) +{ + YY_BUFFER_STATE buffer; + void *scanner; + int ret; + + ret = parse_events_lex_init_extra(start_token, &scanner); + if (ret) + return ret; + + buffer = parse_events__scan_string(str, scanner); + +#ifdef PARSER_DEBUG + parse_events_debug = 1; +#endif + ret = parse_events_parse(data, scanner); + + parse_events__flush_buffer(buffer, scanner); + parse_events__delete_buffer(buffer, scanner); + parse_events_lex_destroy(scanner); + return ret; +} + +/* + * parse event config string, return a list of event terms. + */ +int parse_events_terms(struct list_head *terms, const char *str) +{ + struct parse_events_terms data = { + .terms = NULL, + }; + int ret; + + ret = parse_events__scanner(str, &data, PE_START_TERMS); + if (!ret) { + list_splice(data.terms, terms); + free(data.terms); + return 0; + } + + parse_events__free_terms(data.terms); + return ret; +} + +int parse_events(struct perf_evlist *evlist, const char *str) +{ + struct parse_events_evlist data = { + .list = LIST_HEAD_INIT(data.list), + .idx = evlist->nr_entries, + }; + int ret; + + ret = parse_events__scanner(str, &data, PE_START_EVENTS); + if (!ret) { + int entries = data.idx - evlist->nr_entries; + perf_evlist__splice_list_tail(evlist, &data.list, entries); + evlist->nr_groups += data.nr_groups; + return 0; + } + + /* + * There are 2 users - builtin-record and builtin-test objects. + * Both call perf_evlist__delete in case of error, so we dont + * need to bother. + */ + return ret; +} + +int parse_events_option(const struct option *opt, const char *str, + int unset __maybe_unused) +{ + struct perf_evlist *evlist = *(struct perf_evlist **)opt->value; + int ret = parse_events(evlist, str); + + if (ret) { + fprintf(stderr, "invalid or unsupported event: '%s'\n", str); + fprintf(stderr, "Run 'perf list' for a list of valid events\n"); + } + return ret; +} + +int parse_filter(const struct option *opt, const char *str, + int unset __maybe_unused) +{ + struct perf_evlist *evlist = *(struct perf_evlist **)opt->value; + struct perf_evsel *last = NULL; + + if (evlist->nr_entries > 0) + last = perf_evlist__last(evlist); + + if (last == NULL || last->attr.type != PERF_TYPE_TRACEPOINT) { + fprintf(stderr, + "-F option should follow a -e tracepoint option\n"); + return -1; + } + + last->filter = strdup(str); + if (last->filter == NULL) { + fprintf(stderr, "not enough memory to hold filter string\n"); + return -1; + } + + return 0; +} + +static const char * const event_type_descriptors[] = { + "Hardware event", + "Software event", + "Tracepoint event", + "Hardware cache event", + "Raw hardware event descriptor", + "Hardware breakpoint", +}; + +/* + * Print the events from <debugfs_mount_point>/tracing/events + */ + +void print_tracepoint_events(const char *subsys_glob, const char *event_glob, + bool name_only) +{ + DIR *sys_dir, *evt_dir; + struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent; + char evt_path[MAXPATHLEN]; + char dir_path[MAXPATHLEN]; + + if (debugfs_valid_mountpoint(tracing_events_path)) + return; + + sys_dir = opendir(tracing_events_path); + if (!sys_dir) + return; + + for_each_subsystem(sys_dir, sys_dirent, sys_next) { + if (subsys_glob != NULL && + !strglobmatch(sys_dirent.d_name, subsys_glob)) + continue; + + snprintf(dir_path, MAXPATHLEN, "%s/%s", tracing_events_path, + sys_dirent.d_name); + evt_dir = opendir(dir_path); + if (!evt_dir) + continue; + + for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) { + if (event_glob != NULL && + !strglobmatch(evt_dirent.d_name, event_glob)) + continue; + + if (name_only) { + printf("%s:%s ", sys_dirent.d_name, evt_dirent.d_name); + continue; + } + + snprintf(evt_path, MAXPATHLEN, "%s:%s", + sys_dirent.d_name, evt_dirent.d_name); + printf(" %-50s [%s]\n", evt_path, + event_type_descriptors[PERF_TYPE_TRACEPOINT]); + } + closedir(evt_dir); + } + closedir(sys_dir); +} + +/* + * Check whether event is in <debugfs_mount_point>/tracing/events + */ + +int is_valid_tracepoint(const char *event_string) +{ + DIR *sys_dir, *evt_dir; + struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent; + char evt_path[MAXPATHLEN]; + char dir_path[MAXPATHLEN]; + + if (debugfs_valid_mountpoint(tracing_events_path)) + return 0; + + sys_dir = opendir(tracing_events_path); + if (!sys_dir) + return 0; + + for_each_subsystem(sys_dir, sys_dirent, sys_next) { + + snprintf(dir_path, MAXPATHLEN, "%s/%s", tracing_events_path, + sys_dirent.d_name); + evt_dir = opendir(dir_path); + if (!evt_dir) + continue; + + for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) { + snprintf(evt_path, MAXPATHLEN, "%s:%s", + sys_dirent.d_name, evt_dirent.d_name); + if (!strcmp(evt_path, event_string)) { + closedir(evt_dir); + closedir(sys_dir); + return 1; + } + } + closedir(evt_dir); + } + closedir(sys_dir); + return 0; +} + +static void __print_events_type(u8 type, struct event_symbol *syms, + unsigned max) +{ + char name[64]; + unsigned i; + + for (i = 0; i < max ; i++, syms++) { + if (strlen(syms->alias)) + snprintf(name, sizeof(name), "%s OR %s", + syms->symbol, syms->alias); + else + snprintf(name, sizeof(name), "%s", syms->symbol); + + printf(" %-50s [%s]\n", name, + event_type_descriptors[type]); + } +} + +void print_events_type(u8 type) +{ + if (type == PERF_TYPE_SOFTWARE) + __print_events_type(type, event_symbols_sw, PERF_COUNT_SW_MAX); + else + __print_events_type(type, event_symbols_hw, PERF_COUNT_HW_MAX); +} + +int print_hwcache_events(const char *event_glob, bool name_only) +{ + unsigned int type, op, i, printed = 0; + char name[64]; + + for (type = 0; type < PERF_COUNT_HW_CACHE_MAX; type++) { + for (op = 0; op < PERF_COUNT_HW_CACHE_OP_MAX; op++) { + /* skip invalid cache type */ + if (!perf_evsel__is_cache_op_valid(type, op)) + continue; + + for (i = 0; i < PERF_COUNT_HW_CACHE_RESULT_MAX; i++) { + __perf_evsel__hw_cache_type_op_res_name(type, op, i, + name, sizeof(name)); + if (event_glob != NULL && !strglobmatch(name, event_glob)) + continue; + + if (name_only) + printf("%s ", name); + else + printf(" %-50s [%s]\n", name, + event_type_descriptors[PERF_TYPE_HW_CACHE]); + ++printed; + } + } + } + + return printed; +} + +static void print_symbol_events(const char *event_glob, unsigned type, + struct event_symbol *syms, unsigned max, + bool name_only) +{ + unsigned i, printed = 0; + char name[MAX_NAME_LEN]; + + for (i = 0; i < max; i++, syms++) { + + if (event_glob != NULL && + !(strglobmatch(syms->symbol, event_glob) || + (syms->alias && strglobmatch(syms->alias, event_glob)))) + continue; + + if (name_only) { + printf("%s ", syms->symbol); + continue; + } + + if (strlen(syms->alias)) + snprintf(name, MAX_NAME_LEN, "%s OR %s", syms->symbol, syms->alias); + else + strncpy(name, syms->symbol, MAX_NAME_LEN); + + printf(" %-50s [%s]\n", name, event_type_descriptors[type]); + + printed++; + } + + if (printed) + printf("\n"); +} + +/* + * Print the help text for the event symbols: + */ +void print_events(const char *event_glob, bool name_only) +{ + if (!name_only) { + printf("\n"); + printf("List of pre-defined events (to be used in -e):\n"); + } + + print_symbol_events(event_glob, PERF_TYPE_HARDWARE, + event_symbols_hw, PERF_COUNT_HW_MAX, name_only); + + print_symbol_events(event_glob, PERF_TYPE_SOFTWARE, + event_symbols_sw, PERF_COUNT_SW_MAX, name_only); + + print_hwcache_events(event_glob, name_only); + + if (event_glob != NULL) + return; + + if (!name_only) { + printf("\n"); + printf(" %-50s [%s]\n", + "rNNN", + event_type_descriptors[PERF_TYPE_RAW]); + printf(" %-50s [%s]\n", + "cpu/t1=v1[,t2=v2,t3 ...]/modifier", + event_type_descriptors[PERF_TYPE_RAW]); + printf(" (see 'man perf-list' on how to encode it)\n"); + printf("\n"); + + printf(" %-50s [%s]\n", + "mem:<addr>[:access]", + event_type_descriptors[PERF_TYPE_BREAKPOINT]); + printf("\n"); + } + + print_tracepoint_events(NULL, NULL, name_only); +} + +int parse_events__is_hardcoded_term(struct parse_events_term *term) +{ + return term->type_term != PARSE_EVENTS__TERM_TYPE_USER; +} + +static int new_term(struct parse_events_term **_term, int type_val, + int type_term, char *config, + char *str, u64 num) +{ + struct parse_events_term *term; + + term = zalloc(sizeof(*term)); + if (!term) + return -ENOMEM; + + INIT_LIST_HEAD(&term->list); + term->type_val = type_val; + term->type_term = type_term; + term->config = config; + + switch (type_val) { + case PARSE_EVENTS__TERM_TYPE_NUM: + term->val.num = num; + break; + case PARSE_EVENTS__TERM_TYPE_STR: + term->val.str = str; + break; + default: + return -EINVAL; + } + + *_term = term; + return 0; +} + +int parse_events_term__num(struct parse_events_term **term, + int type_term, char *config, u64 num) +{ + return new_term(term, PARSE_EVENTS__TERM_TYPE_NUM, type_term, + config, NULL, num); +} + +int parse_events_term__str(struct parse_events_term **term, + int type_term, char *config, char *str) +{ + return new_term(term, PARSE_EVENTS__TERM_TYPE_STR, type_term, + config, str, 0); +} + +int parse_events_term__sym_hw(struct parse_events_term **term, + char *config, unsigned idx) +{ + struct event_symbol *sym; + + BUG_ON(idx >= PERF_COUNT_HW_MAX); + sym = &event_symbols_hw[idx]; + + if (config) + return new_term(term, PARSE_EVENTS__TERM_TYPE_STR, + PARSE_EVENTS__TERM_TYPE_USER, config, + (char *) sym->symbol, 0); + else + return new_term(term, PARSE_EVENTS__TERM_TYPE_STR, + PARSE_EVENTS__TERM_TYPE_USER, + (char *) "event", (char *) sym->symbol, 0); +} + +int parse_events_term__clone(struct parse_events_term **new, + struct parse_events_term *term) +{ + return new_term(new, term->type_val, term->type_term, term->config, + term->val.str, term->val.num); +} + +void parse_events__free_terms(struct list_head *terms) +{ + struct parse_events_term *term, *h; + + list_for_each_entry_safe(term, h, terms, list) + free(term); + + free(terms); +} diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h new file mode 100644 index 00000000..8a485931 --- /dev/null +++ b/tools/perf/util/parse-events.h @@ -0,0 +1,112 @@ +#ifndef __PERF_PARSE_EVENTS_H +#define __PERF_PARSE_EVENTS_H +/* + * Parse symbolic events/counts passed in as options: + */ + +#include <linux/list.h> +#include <stdbool.h> +#include "types.h" +#include <linux/perf_event.h> +#include "types.h" + +struct list_head; +struct perf_evsel; +struct perf_evlist; + +struct option; + +struct tracepoint_path { + char *system; + char *name; + struct tracepoint_path *next; +}; + +extern struct tracepoint_path *tracepoint_id_to_path(u64 config); +extern bool have_tracepoints(struct list_head *evlist); + +const char *event_type(int type); + +extern int parse_events_option(const struct option *opt, const char *str, + int unset); +extern int parse_events(struct perf_evlist *evlist, const char *str); +extern int parse_events_terms(struct list_head *terms, const char *str); +extern int parse_filter(const struct option *opt, const char *str, int unset); + +#define EVENTS_HELP_MAX (128*1024) + +enum { + PARSE_EVENTS__TERM_TYPE_NUM, + PARSE_EVENTS__TERM_TYPE_STR, +}; + +enum { + PARSE_EVENTS__TERM_TYPE_USER, + PARSE_EVENTS__TERM_TYPE_CONFIG, + PARSE_EVENTS__TERM_TYPE_CONFIG1, + PARSE_EVENTS__TERM_TYPE_CONFIG2, + PARSE_EVENTS__TERM_TYPE_NAME, + PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD, + PARSE_EVENTS__TERM_TYPE_BRANCH_SAMPLE_TYPE, +}; + +struct parse_events_term { + char *config; + union { + char *str; + u64 num; + } val; + int type_val; + int type_term; + struct list_head list; +}; + +struct parse_events_evlist { + struct list_head list; + int idx; + int nr_groups; +}; + +struct parse_events_terms { + struct list_head *terms; +}; + +int parse_events__is_hardcoded_term(struct parse_events_term *term); +int parse_events_term__num(struct parse_events_term **_term, + int type_term, char *config, u64 num); +int parse_events_term__str(struct parse_events_term **_term, + int type_term, char *config, char *str); +int parse_events_term__sym_hw(struct parse_events_term **term, + char *config, unsigned idx); +int parse_events_term__clone(struct parse_events_term **new, + struct parse_events_term *term); +void parse_events__free_terms(struct list_head *terms); +int parse_events__modifier_event(struct list_head *list, char *str, bool add); +int parse_events__modifier_group(struct list_head *list, char *event_mod); +int parse_events_name(struct list_head *list, char *name); +int parse_events_add_tracepoint(struct list_head **list, int *idx, + char *sys, char *event); +int parse_events_add_numeric(struct list_head **list, int *idx, + u32 type, u64 config, + struct list_head *head_config); +int parse_events_add_cache(struct list_head **list, int *idx, + char *type, char *op_result1, char *op_result2); +int parse_events_add_breakpoint(struct list_head **list, int *idx, + void *ptr, char *type); +int parse_events_add_pmu(struct list_head **list, int *idx, + char *pmu , struct list_head *head_config); +void parse_events__set_leader(char *name, struct list_head *list); +void parse_events_update_lists(struct list_head *list_event, + struct list_head *list_all); +void parse_events_error(void *data, void *scanner, char const *msg); + +void print_events(const char *event_glob, bool name_only); +void print_events_type(u8 type); +void print_tracepoint_events(const char *subsys_glob, const char *event_glob, + bool name_only); +int print_hwcache_events(const char *event_glob, bool name_only); +extern int is_valid_tracepoint(const char *event_string); + +extern int valid_debugfs_mount(const char *debugfs); + +#endif /* __PERF_PARSE_EVENTS_H */ diff --git a/tools/perf/util/parse-events.l b/tools/perf/util/parse-events.l new file mode 100644 index 00000000..e9d1134c --- /dev/null +++ b/tools/perf/util/parse-events.l @@ -0,0 +1,215 @@ + +%option reentrant +%option bison-bridge +%option prefix="parse_events_" +%option stack + +%{ +#include <errno.h> +#include "../perf.h" +#include "parse-events-bison.h" +#include "parse-events.h" + +char *parse_events_get_text(yyscan_t yyscanner); +YYSTYPE *parse_events_get_lval(yyscan_t yyscanner); + +static int __value(YYSTYPE *yylval, char *str, int base, int token) +{ + u64 num; + + errno = 0; + num = strtoull(str, NULL, base); + if (errno) + return PE_ERROR; + + yylval->num = num; + return token; +} + +static int value(yyscan_t scanner, int base) +{ + YYSTYPE *yylval = parse_events_get_lval(scanner); + char *text = parse_events_get_text(scanner); + + return __value(yylval, text, base, PE_VALUE); +} + +static int raw(yyscan_t scanner) +{ + YYSTYPE *yylval = parse_events_get_lval(scanner); + char *text = parse_events_get_text(scanner); + + return __value(yylval, text + 1, 16, PE_RAW); +} + +static int str(yyscan_t scanner, int token) +{ + YYSTYPE *yylval = parse_events_get_lval(scanner); + char *text = parse_events_get_text(scanner); + + yylval->str = strdup(text); + return token; +} + +static int sym(yyscan_t scanner, int type, int config) +{ + YYSTYPE *yylval = parse_events_get_lval(scanner); + + yylval->num = (type << 16) + config; + return type == PERF_TYPE_HARDWARE ? PE_VALUE_SYM_HW : PE_VALUE_SYM_SW; +} + +static int term(yyscan_t scanner, int type) +{ + YYSTYPE *yylval = parse_events_get_lval(scanner); + + yylval->num = type; + return PE_TERM; +} + +%} + +%x mem +%s config +%x event + +group [^,{}/]*[{][^}]*[}][^,{}/]* +event_pmu [^,{}/]+[/][^/]*[/][^,{}/]* +event [^,{}/]+ + +num_dec [0-9]+ +num_hex 0x[a-fA-F0-9]+ +num_raw_hex [a-fA-F0-9]+ +name [a-zA-Z_*?][a-zA-Z0-9_*?]* +name_minus [a-zA-Z_*?][a-zA-Z0-9\-_*?]* +modifier_event [ukhpGH]+ +modifier_bp [rwx]{1,3} + +%% + +%{ + { + int start_token; + + start_token = parse_events_get_extra(yyscanner); + + if (start_token == PE_START_TERMS) + BEGIN(config); + else if (start_token == PE_START_EVENTS) + BEGIN(event); + + if (start_token) { + parse_events_set_extra(NULL, yyscanner); + return start_token; + } + } +%} + +<event>{ + +{group} { + BEGIN(INITIAL); yyless(0); + } + +{event_pmu} | +{event} { + str(yyscanner, PE_EVENT_NAME); + BEGIN(INITIAL); yyless(0); + return PE_EVENT_NAME; + } + +. | +<<EOF>> { + BEGIN(INITIAL); yyless(0); + } + +} + +cpu-cycles|cycles { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_CPU_CYCLES); } +stalled-cycles-frontend|idle-cycles-frontend { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_STALLED_CYCLES_FRONTEND); } +stalled-cycles-backend|idle-cycles-backend { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_STALLED_CYCLES_BACKEND); } +instructions { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_INSTRUCTIONS); } +cache-references { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_CACHE_REFERENCES); } +cache-misses { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_CACHE_MISSES); } +branch-instructions|branches { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_BRANCH_INSTRUCTIONS); } +branch-misses { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_BRANCH_MISSES); } +bus-cycles { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_BUS_CYCLES); } +ref-cycles { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_REF_CPU_CYCLES); } +cpu-clock { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CPU_CLOCK); } +task-clock { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_TASK_CLOCK); } +page-faults|faults { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_PAGE_FAULTS); } +minor-faults { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_PAGE_FAULTS_MIN); } +major-faults { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_PAGE_FAULTS_MAJ); } +context-switches|cs { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CONTEXT_SWITCHES); } +cpu-migrations|migrations { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CPU_MIGRATIONS); } +alignment-faults { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_ALIGNMENT_FAULTS); } +emulation-faults { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_EMULATION_FAULTS); } + +L1-dcache|l1-d|l1d|L1-data | +L1-icache|l1-i|l1i|L1-instruction | +LLC|L2 | +dTLB|d-tlb|Data-TLB | +iTLB|i-tlb|Instruction-TLB | +branch|branches|bpu|btb|bpc | +node { return str(yyscanner, PE_NAME_CACHE_TYPE); } + +load|loads|read | +store|stores|write | +prefetch|prefetches | +speculative-read|speculative-load | +refs|Reference|ops|access | +misses|miss { return str(yyscanner, PE_NAME_CACHE_OP_RESULT); } + +<config>{ +config { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_CONFIG); } +config1 { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_CONFIG1); } +config2 { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_CONFIG2); } +name { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_NAME); } +period { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD); } +branch_type { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_BRANCH_SAMPLE_TYPE); } +, { return ','; } +"/" { BEGIN(INITIAL); return '/'; } +{name_minus} { return str(yyscanner, PE_NAME); } +} + +mem: { BEGIN(mem); return PE_PREFIX_MEM; } +r{num_raw_hex} { return raw(yyscanner); } +{num_dec} { return value(yyscanner, 10); } +{num_hex} { return value(yyscanner, 16); } + +{modifier_event} { return str(yyscanner, PE_MODIFIER_EVENT); } +{name} { return str(yyscanner, PE_NAME); } +"/" { BEGIN(config); return '/'; } +- { return '-'; } +, { BEGIN(event); return ','; } +: { return ':'; } +"{" { BEGIN(event); return '{'; } +"}" { return '}'; } += { return '='; } +\n { } + +<mem>{ +{modifier_bp} { return str(yyscanner, PE_MODIFIER_BP); } +: { return ':'; } +{num_dec} { return value(yyscanner, 10); } +{num_hex} { return value(yyscanner, 16); } + /* + * We need to separate 'mem:' scanner part, in order to get specific + * modifier bits parsed out. Otherwise we would need to handle PE_NAME + * and we'd need to parse it manually. During the escape from <mem> + * state we need to put the escaping char back, so we dont miss it. + */ +. { unput(*yytext); BEGIN(INITIAL); } + /* + * We destroy the scanner after reaching EOF, + * but anyway just to be sure get back to INIT state. + */ +<<EOF>> { BEGIN(INITIAL); } +} + +%% + +int parse_events_wrap(void *scanner __maybe_unused) +{ + return 1; +} diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y new file mode 100644 index 00000000..afc44c18 --- /dev/null +++ b/tools/perf/util/parse-events.y @@ -0,0 +1,424 @@ +%pure-parser +%parse-param {void *_data} +%parse-param {void *scanner} +%lex-param {void* scanner} + +%{ + +#define YYDEBUG 1 + +#include <linux/compiler.h> +#include <linux/list.h> +#include "types.h" +#include "util.h" +#include "parse-events.h" +#include "parse-events-bison.h" + +extern int parse_events_lex (YYSTYPE* lvalp, void* scanner); + +#define ABORT_ON(val) \ +do { \ + if (val) \ + YYABORT; \ +} while (0) + +static inc_group_count(struct list_head *list, + struct parse_events_evlist *data) +{ + /* Count groups only have more than 1 members */ + if (!list_is_last(list->next, list)) + data->nr_groups++; +} + +%} + +%token PE_START_EVENTS PE_START_TERMS +%token PE_VALUE PE_VALUE_SYM_HW PE_VALUE_SYM_SW PE_RAW PE_TERM +%token PE_EVENT_NAME +%token PE_NAME +%token PE_MODIFIER_EVENT PE_MODIFIER_BP +%token PE_NAME_CACHE_TYPE PE_NAME_CACHE_OP_RESULT +%token PE_PREFIX_MEM PE_PREFIX_RAW PE_PREFIX_GROUP +%token PE_ERROR +%type <num> PE_VALUE +%type <num> PE_VALUE_SYM_HW +%type <num> PE_VALUE_SYM_SW +%type <num> PE_RAW +%type <num> PE_TERM +%type <str> PE_NAME +%type <str> PE_NAME_CACHE_TYPE +%type <str> PE_NAME_CACHE_OP_RESULT +%type <str> PE_MODIFIER_EVENT +%type <str> PE_MODIFIER_BP +%type <str> PE_EVENT_NAME +%type <num> value_sym +%type <head> event_config +%type <term> event_term +%type <head> event_pmu +%type <head> event_legacy_symbol +%type <head> event_legacy_cache +%type <head> event_legacy_mem +%type <head> event_legacy_tracepoint +%type <head> event_legacy_numeric +%type <head> event_legacy_raw +%type <head> event_def +%type <head> event_mod +%type <head> event_name +%type <head> event +%type <head> events +%type <head> group_def +%type <head> group +%type <head> groups + +%union +{ + char *str; + u64 num; + struct list_head *head; + struct parse_events_term *term; +} +%% + +start: +PE_START_EVENTS start_events +| +PE_START_TERMS start_terms + +start_events: groups +{ + struct parse_events_evlist *data = _data; + + parse_events_update_lists($1, &data->list); +} + +groups: +groups ',' group +{ + struct list_head *list = $1; + struct list_head *group = $3; + + parse_events_update_lists(group, list); + $$ = list; +} +| +groups ',' event +{ + struct list_head *list = $1; + struct list_head *event = $3; + + parse_events_update_lists(event, list); + $$ = list; +} +| +group +| +event + +group: +group_def ':' PE_MODIFIER_EVENT +{ + struct list_head *list = $1; + + ABORT_ON(parse_events__modifier_group(list, $3)); + $$ = list; +} +| +group_def + +group_def: +PE_NAME '{' events '}' +{ + struct list_head *list = $3; + + inc_group_count(list, _data); + parse_events__set_leader($1, list); + $$ = list; +} +| +'{' events '}' +{ + struct list_head *list = $2; + + inc_group_count(list, _data); + parse_events__set_leader(NULL, list); + $$ = list; +} + +events: +events ',' event +{ + struct list_head *event = $3; + struct list_head *list = $1; + + parse_events_update_lists(event, list); + $$ = list; +} +| +event + +event: event_mod + +event_mod: +event_name PE_MODIFIER_EVENT +{ + struct list_head *list = $1; + + /* + * Apply modifier on all events added by single event definition + * (there could be more events added for multiple tracepoint + * definitions via '*?'. + */ + ABORT_ON(parse_events__modifier_event(list, $2, false)); + $$ = list; +} +| +event_name + +event_name: +PE_EVENT_NAME event_def +{ + ABORT_ON(parse_events_name($2, $1)); + free($1); + $$ = $2; +} +| +event_def + +event_def: event_pmu | + event_legacy_symbol | + event_legacy_cache sep_dc | + event_legacy_mem | + event_legacy_tracepoint sep_dc | + event_legacy_numeric sep_dc | + event_legacy_raw sep_dc + +event_pmu: +PE_NAME '/' event_config '/' +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + + ABORT_ON(parse_events_add_pmu(&list, &data->idx, $1, $3)); + parse_events__free_terms($3); + $$ = list; +} + +value_sym: +PE_VALUE_SYM_HW +| +PE_VALUE_SYM_SW + +event_legacy_symbol: +value_sym '/' event_config '/' +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + int type = $1 >> 16; + int config = $1 & 255; + + ABORT_ON(parse_events_add_numeric(&list, &data->idx, + type, config, $3)); + parse_events__free_terms($3); + $$ = list; +} +| +value_sym sep_slash_dc +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + int type = $1 >> 16; + int config = $1 & 255; + + ABORT_ON(parse_events_add_numeric(&list, &data->idx, + type, config, NULL)); + $$ = list; +} + +event_legacy_cache: +PE_NAME_CACHE_TYPE '-' PE_NAME_CACHE_OP_RESULT '-' PE_NAME_CACHE_OP_RESULT +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + + ABORT_ON(parse_events_add_cache(&list, &data->idx, $1, $3, $5)); + $$ = list; +} +| +PE_NAME_CACHE_TYPE '-' PE_NAME_CACHE_OP_RESULT +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + + ABORT_ON(parse_events_add_cache(&list, &data->idx, $1, $3, NULL)); + $$ = list; +} +| +PE_NAME_CACHE_TYPE +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + + ABORT_ON(parse_events_add_cache(&list, &data->idx, $1, NULL, NULL)); + $$ = list; +} + +event_legacy_mem: +PE_PREFIX_MEM PE_VALUE ':' PE_MODIFIER_BP sep_dc +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + + ABORT_ON(parse_events_add_breakpoint(&list, &data->idx, + (void *) $2, $4)); + $$ = list; +} +| +PE_PREFIX_MEM PE_VALUE sep_dc +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + + ABORT_ON(parse_events_add_breakpoint(&list, &data->idx, + (void *) $2, NULL)); + $$ = list; +} + +event_legacy_tracepoint: +PE_NAME ':' PE_NAME +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + + ABORT_ON(parse_events_add_tracepoint(&list, &data->idx, $1, $3)); + $$ = list; +} + +event_legacy_numeric: +PE_VALUE ':' PE_VALUE +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + + ABORT_ON(parse_events_add_numeric(&list, &data->idx, (u32)$1, $3, NULL)); + $$ = list; +} + +event_legacy_raw: +PE_RAW +{ + struct parse_events_evlist *data = _data; + struct list_head *list = NULL; + + ABORT_ON(parse_events_add_numeric(&list, &data->idx, + PERF_TYPE_RAW, $1, NULL)); + $$ = list; +} + +start_terms: event_config +{ + struct parse_events_terms *data = _data; + data->terms = $1; +} + +event_config: +event_config ',' event_term +{ + struct list_head *head = $1; + struct parse_events_term *term = $3; + + ABORT_ON(!head); + list_add_tail(&term->list, head); + $$ = $1; +} +| +event_term +{ + struct list_head *head = malloc(sizeof(*head)); + struct parse_events_term *term = $1; + + ABORT_ON(!head); + INIT_LIST_HEAD(head); + list_add_tail(&term->list, head); + $$ = head; +} + +event_term: +PE_NAME '=' PE_NAME +{ + struct parse_events_term *term; + + ABORT_ON(parse_events_term__str(&term, PARSE_EVENTS__TERM_TYPE_USER, + $1, $3)); + $$ = term; +} +| +PE_NAME '=' PE_VALUE +{ + struct parse_events_term *term; + + ABORT_ON(parse_events_term__num(&term, PARSE_EVENTS__TERM_TYPE_USER, + $1, $3)); + $$ = term; +} +| +PE_NAME '=' PE_VALUE_SYM_HW +{ + struct parse_events_term *term; + int config = $3 & 255; + + ABORT_ON(parse_events_term__sym_hw(&term, $1, config)); + $$ = term; +} +| +PE_NAME +{ + struct parse_events_term *term; + + ABORT_ON(parse_events_term__num(&term, PARSE_EVENTS__TERM_TYPE_USER, + $1, 1)); + $$ = term; +} +| +PE_VALUE_SYM_HW +{ + struct parse_events_term *term; + int config = $1 & 255; + + ABORT_ON(parse_events_term__sym_hw(&term, NULL, config)); + $$ = term; +} +| +PE_TERM '=' PE_NAME +{ + struct parse_events_term *term; + + ABORT_ON(parse_events_term__str(&term, (int)$1, NULL, $3)); + $$ = term; +} +| +PE_TERM '=' PE_VALUE +{ + struct parse_events_term *term; + + ABORT_ON(parse_events_term__num(&term, (int)$1, NULL, $3)); + $$ = term; +} +| +PE_TERM +{ + struct parse_events_term *term; + + ABORT_ON(parse_events_term__num(&term, (int)$1, NULL, 1)); + $$ = term; +} + +sep_dc: ':' | + +sep_slash_dc: '/' | ':' | + +%% + +void parse_events_error(void *data __maybe_unused, void *scanner __maybe_unused, + char const *msg __maybe_unused) +{ +} diff --git a/tools/perf/util/parse-options.c b/tools/perf/util/parse-options.c new file mode 100644 index 00000000..2bc9e70d --- /dev/null +++ b/tools/perf/util/parse-options.c @@ -0,0 +1,589 @@ +#include "util.h" +#include "parse-options.h" +#include "cache.h" +#include "header.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 ((opt->flags & PARSE_OPT_LASTARG_DEFAULT) && (p->argc == 1 || + **(p->argv + 1) == '-')) { + *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_INCR: + case OPTION_BIT: + case OPTION_SET_UINT: + case OPTION_SET_PTR: + return opterror(opt, "takes no value", flags); + case OPTION_END: + case OPTION_ARGUMENT: + case OPTION_GROUP: + case OPTION_STRING: + case OPTION_INTEGER: + case OPTION_UINTEGER: + case OPTION_LONG: + case OPTION_U64: + default: + break; + } + } + + switch (opt->type) { + case OPTION_BIT: + if (unset) + *(int *)opt->value &= ~opt->defval; + else + *(int *)opt->value |= opt->defval; + return 0; + + case OPTION_BOOLEAN: + *(bool *)opt->value = unset ? false : true; + return 0; + + case OPTION_INCR: + *(int *)opt->value = unset ? 0 : *(int *)opt->value + 1; + return 0; + + case OPTION_SET_UINT: + *(unsigned 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_UINTEGER: + if (unset) { + *(unsigned int *)opt->value = 0; + return 0; + } + if (opt->flags & PARSE_OPT_OPTARG && !p->opt) { + *(unsigned int *)opt->value = opt->defval; + return 0; + } + if (get_arg(p, opt, flags, &arg)) + return -1; + *(unsigned int *)opt->value = strtol(arg, (char **)&s, 10); + if (*s) + return opterror(opt, "expects a numerical value", flags); + return 0; + + case OPTION_LONG: + if (unset) { + *(long *)opt->value = 0; + 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; + + case OPTION_U64: + if (unset) { + *(u64 *)opt->value = 0; + return 0; + } + if (opt->flags & PARSE_OPT_OPTARG && !p->opt) { + *(u64 *)opt->value = opt->defval; + return 0; + } + if (get_arg(p, opt, flags, &arg)) + return -1; + *(u64 *)opt->value = strtoull(arg, (char **)&s, 10); + if (*s) + return opterror(opt, "expects a numerical value", flags); + return 0; + + case OPTION_END: + case OPTION_ARGUMENT: + case OPTION_GROUP: + default: + die("should not happen, someone must be hit on the forehead"); + } +} + +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; + default: + break; + } + 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; + default: + break; + } + } + 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); + if (!strcmp(arg + 2, "list-opts")) + return PARSE_OPT_LIST; + switch (parse_long_opt(ctx, arg + 2, options)) { + case -1: + return parse_options_usage(usagestr, options); + case -2: + goto unknown; + default: + break; + } + 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; + + perf_header__set_cmdline(argc, argv); + + 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; + case PARSE_OPT_LIST: + while (options->type != OPTION_END) { + printf("--%s ", options->long_name); + options++; + } + exit(130); + 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); + else + pos += fprintf(stderr, " "); + + 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_LONG: + case OPTION_U64: + case OPTION_INTEGER: + case OPTION_UINTEGER: + 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_UINT,SET_PTR} */ + case OPTION_END: + case OPTION_GROUP: + case OPTION_BIT: + case OPTION_BOOLEAN: + case OPTION_INCR: + case OPTION_SET_UINT: + case OPTION_SET_PTR: + break; + } + + 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) +{ + exit_browser(false); + 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 __maybe_unused, + 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 00000000..7bb59999 --- /dev/null +++ b/tools/perf/util/parse-options.h @@ -0,0 +1,193 @@ +#ifndef __PERF_PARSE_OPTIONS_H +#define __PERF_PARSE_OPTIONS_H + +#include <linux/kernel.h> +#include <stdbool.h> + +enum parse_opt_type { + /* special types */ + OPTION_END, + OPTION_ARGUMENT, + OPTION_GROUP, + /* options with no arguments */ + OPTION_BIT, + OPTION_BOOLEAN, + OPTION_INCR, + OPTION_SET_UINT, + OPTION_SET_PTR, + /* options with arguments (usually) */ + OPTION_STRING, + OPTION_INTEGER, + OPTION_LONG, + OPTION_CALLBACK, + OPTION_U64, + OPTION_UINTEGER, +}; + +enum parse_opt_flags { + 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_UINT,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 check_vtype(v, type) ( BUILD_BUG_ON_ZERO(!__builtin_types_compatible_p(typeof(v), type)) + v ) + +#define OPT_END() { .type = OPTION_END } +#define OPT_ARGUMENT(l, h) { .type = OPTION_ARGUMENT, .long_name = (l), .help = (h) } +#define OPT_GROUP(h) { .type = OPTION_GROUP, .help = (h) } +#define OPT_BIT(s, l, v, h, b) { .type = OPTION_BIT, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h), .defval = (b) } +#define OPT_BOOLEAN(s, l, v, h) { .type = OPTION_BOOLEAN, .short_name = (s), .long_name = (l), .value = check_vtype(v, bool *), .help = (h) } +#define OPT_INCR(s, l, v, h) { .type = OPTION_INCR, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h) } +#define OPT_SET_UINT(s, l, v, h, i) { .type = OPTION_SET_UINT, .short_name = (s), .long_name = (l), .value = check_vtype(v, unsigned int *), .help = (h), .defval = (i) } +#define OPT_SET_PTR(s, l, v, h, p) { .type = OPTION_SET_PTR, .short_name = (s), .long_name = (l), .value = (v), .help = (h), .defval = (p) } +#define OPT_INTEGER(s, l, v, h) { .type = OPTION_INTEGER, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h) } +#define OPT_UINTEGER(s, l, v, h) { .type = OPTION_UINTEGER, .short_name = (s), .long_name = (l), .value = check_vtype(v, unsigned int *), .help = (h) } +#define OPT_LONG(s, l, v, h) { .type = OPTION_LONG, .short_name = (s), .long_name = (l), .value = check_vtype(v, long *), .help = (h) } +#define OPT_U64(s, l, v, h) { .type = OPTION_U64, .short_name = (s), .long_name = (l), .value = check_vtype(v, u64 *), .help = (h) } +#define OPT_STRING(s, l, v, a, h) { .type = OPTION_STRING, .short_name = (s), .long_name = (l), .value = check_vtype(v, const char **), (a), .help = (h) } +#define OPT_DATE(s, l, v, h) \ + { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), .argh = "time", .help = (h), .callback = parse_opt_approxidate_cb } +#define OPT_CALLBACK(s, l, v, a, h, f) \ + { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f) } +#define OPT_CALLBACK_NOOPT(s, l, v, a, h, f) \ + { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f), .flags = PARSE_OPT_NOARG } +#define OPT_CALLBACK_DEFAULT(s, l, v, a, h, f, d) \ + { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f), .defval = (intptr_t)d, .flags = PARSE_OPT_LASTARG_DEFAULT } +#define OPT_CALLBACK_DEFAULT_NOOPT(s, l, v, a, h, f, d) \ + { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l),\ + .value = (v), (a), .help = (h), .callback = (f), .defval = (intptr_t)d,\ + .flags = PARSE_OPT_LASTARG_DEFAULT | PARSE_OPT_NOARG} + +/* parse_options() will filter out the processed options and leave the + * non-option argments in argv[]. + * 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_LIST, + 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 /* __PERF_PARSE_OPTIONS_H */ diff --git a/tools/perf/util/path.c b/tools/perf/util/path.c new file mode 100644 index 00000000..a8c49548 --- /dev/null +++ b/tools/perf/util/path.c @@ -0,0 +1,157 @@ +/* + * 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 const char *get_perf_dir(void) +{ + return "."; +} + +#ifndef HAVE_STRLCPY +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; +} +#endif + +static char *get_pathname(void) +{ + static char pathname_array[4][PATH_MAX]; + static int idx; + + return pathname_array[3 & ++idx]; +} + +static char *cleanup_path(char *path) +{ + /* Clean it up */ + if (!memcmp(path, "./", 2)) { + path += 2; + while (*path == '/') + path++; + } + return path; +} + +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_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); +} + +/* 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 strndup(path, chomp_trailing_dir_sep(path, path_len)); +} diff --git a/tools/perf/util/perf_regs.h b/tools/perf/util/perf_regs.h new file mode 100644 index 00000000..5a4f2b6f --- /dev/null +++ b/tools/perf/util/perf_regs.h @@ -0,0 +1,14 @@ +#ifndef __PERF_REGS_H +#define __PERF_REGS_H + +#ifdef HAVE_PERF_REGS +#include <perf_regs.h> +#else +#define PERF_REGS_MASK 0 + +static inline const char *perf_reg_name(int id __maybe_unused) +{ + return NULL; +} +#endif /* HAVE_PERF_REGS */ +#endif /* __PERF_REGS_H */ diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c new file mode 100644 index 00000000..4c6f9c49 --- /dev/null +++ b/tools/perf/util/pmu.c @@ -0,0 +1,566 @@ +#include <linux/list.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <unistd.h> +#include <stdio.h> +#include <dirent.h> +#include "sysfs.h" +#include "util.h" +#include "pmu.h" +#include "parse-events.h" +#include "cpumap.h" + +struct perf_pmu_alias { + char *name; + struct list_head terms; + struct list_head list; +}; + +struct perf_pmu_format { + char *name; + int value; + DECLARE_BITMAP(bits, PERF_PMU_FORMAT_BITS); + struct list_head list; +}; + +#define EVENT_SOURCE_DEVICE_PATH "/bus/event_source/devices/" + +int perf_pmu_parse(struct list_head *list, char *name); +extern FILE *perf_pmu_in; + +static LIST_HEAD(pmus); + +/* + * Parse & process all the sysfs attributes located under + * the directory specified in 'dir' parameter. + */ +int perf_pmu__format_parse(char *dir, struct list_head *head) +{ + struct dirent *evt_ent; + DIR *format_dir; + int ret = 0; + + format_dir = opendir(dir); + if (!format_dir) + return -EINVAL; + + while (!ret && (evt_ent = readdir(format_dir))) { + char path[PATH_MAX]; + char *name = evt_ent->d_name; + FILE *file; + + if (!strcmp(name, ".") || !strcmp(name, "..")) + continue; + + snprintf(path, PATH_MAX, "%s/%s", dir, name); + + ret = -EINVAL; + file = fopen(path, "r"); + if (!file) + break; + + perf_pmu_in = file; + ret = perf_pmu_parse(head, name); + fclose(file); + } + + closedir(format_dir); + return ret; +} + +/* + * Reading/parsing the default pmu format definition, which should be + * located at: + * /sys/bus/event_source/devices/<dev>/format as sysfs group attributes. + */ +static int pmu_format(char *name, struct list_head *format) +{ + struct stat st; + char path[PATH_MAX]; + const char *sysfs; + + sysfs = sysfs_find_mountpoint(); + if (!sysfs) + return -1; + + snprintf(path, PATH_MAX, + "%s" EVENT_SOURCE_DEVICE_PATH "%s/format", sysfs, name); + + if (stat(path, &st) < 0) + return 0; /* no error if format does not exist */ + + if (perf_pmu__format_parse(path, format)) + return -1; + + return 0; +} + +static int perf_pmu__new_alias(struct list_head *list, char *name, FILE *file) +{ + struct perf_pmu_alias *alias; + char buf[256]; + int ret; + + ret = fread(buf, 1, sizeof(buf), file); + if (ret == 0) + return -EINVAL; + buf[ret] = 0; + + alias = malloc(sizeof(*alias)); + if (!alias) + return -ENOMEM; + + INIT_LIST_HEAD(&alias->terms); + ret = parse_events_terms(&alias->terms, buf); + if (ret) { + free(alias); + return ret; + } + + alias->name = strdup(name); + list_add_tail(&alias->list, list); + return 0; +} + +/* + * Process all the sysfs attributes located under the directory + * specified in 'dir' parameter. + */ +static int pmu_aliases_parse(char *dir, struct list_head *head) +{ + struct dirent *evt_ent; + DIR *event_dir; + int ret = 0; + + event_dir = opendir(dir); + if (!event_dir) + return -EINVAL; + + while (!ret && (evt_ent = readdir(event_dir))) { + char path[PATH_MAX]; + char *name = evt_ent->d_name; + FILE *file; + + if (!strcmp(name, ".") || !strcmp(name, "..")) + continue; + + snprintf(path, PATH_MAX, "%s/%s", dir, name); + + ret = -EINVAL; + file = fopen(path, "r"); + if (!file) + break; + ret = perf_pmu__new_alias(head, name, file); + fclose(file); + } + + closedir(event_dir); + return ret; +} + +/* + * Reading the pmu event aliases definition, which should be located at: + * /sys/bus/event_source/devices/<dev>/events as sysfs group attributes. + */ +static int pmu_aliases(char *name, struct list_head *head) +{ + struct stat st; + char path[PATH_MAX]; + const char *sysfs; + + sysfs = sysfs_find_mountpoint(); + if (!sysfs) + return -1; + + snprintf(path, PATH_MAX, + "%s/bus/event_source/devices/%s/events", sysfs, name); + + if (stat(path, &st) < 0) + return 0; /* no error if 'events' does not exist */ + + if (pmu_aliases_parse(path, head)) + return -1; + + return 0; +} + +static int pmu_alias_terms(struct perf_pmu_alias *alias, + struct list_head *terms) +{ + struct parse_events_term *term, *clone; + LIST_HEAD(list); + int ret; + + list_for_each_entry(term, &alias->terms, list) { + ret = parse_events_term__clone(&clone, term); + if (ret) { + parse_events__free_terms(&list); + return ret; + } + list_add_tail(&clone->list, &list); + } + list_splice(&list, terms); + return 0; +} + +/* + * Reading/parsing the default pmu type value, which should be + * located at: + * /sys/bus/event_source/devices/<dev>/type as sysfs attribute. + */ +static int pmu_type(char *name, __u32 *type) +{ + struct stat st; + char path[PATH_MAX]; + const char *sysfs; + FILE *file; + int ret = 0; + + sysfs = sysfs_find_mountpoint(); + if (!sysfs) + return -1; + + snprintf(path, PATH_MAX, + "%s" EVENT_SOURCE_DEVICE_PATH "%s/type", sysfs, name); + + if (stat(path, &st) < 0) + return -1; + + file = fopen(path, "r"); + if (!file) + return -EINVAL; + + if (1 != fscanf(file, "%u", type)) + ret = -1; + + fclose(file); + return ret; +} + +/* Add all pmus in sysfs to pmu list: */ +static void pmu_read_sysfs(void) +{ + char path[PATH_MAX]; + const char *sysfs; + DIR *dir; + struct dirent *dent; + + sysfs = sysfs_find_mountpoint(); + if (!sysfs) + return; + + snprintf(path, PATH_MAX, + "%s" EVENT_SOURCE_DEVICE_PATH, sysfs); + + dir = opendir(path); + if (!dir) + return; + + while ((dent = readdir(dir))) { + if (!strcmp(dent->d_name, ".") || !strcmp(dent->d_name, "..")) + continue; + /* add to static LIST_HEAD(pmus): */ + perf_pmu__find(dent->d_name); + } + + closedir(dir); +} + +static struct cpu_map *pmu_cpumask(char *name) +{ + struct stat st; + char path[PATH_MAX]; + const char *sysfs; + FILE *file; + struct cpu_map *cpus; + + sysfs = sysfs_find_mountpoint(); + if (!sysfs) + return NULL; + + snprintf(path, PATH_MAX, + "%s/bus/event_source/devices/%s/cpumask", sysfs, name); + + if (stat(path, &st) < 0) + return NULL; + + file = fopen(path, "r"); + if (!file) + return NULL; + + cpus = cpu_map__read(file); + fclose(file); + return cpus; +} + +static struct perf_pmu *pmu_lookup(char *name) +{ + struct perf_pmu *pmu; + LIST_HEAD(format); + LIST_HEAD(aliases); + __u32 type; + + /* + * The pmu data we store & need consists of the pmu + * type value and format definitions. Load both right + * now. + */ + if (pmu_format(name, &format)) + return NULL; + + if (pmu_aliases(name, &aliases)) + return NULL; + + if (pmu_type(name, &type)) + return NULL; + + pmu = zalloc(sizeof(*pmu)); + if (!pmu) + return NULL; + + pmu->cpus = pmu_cpumask(name); + + INIT_LIST_HEAD(&pmu->format); + INIT_LIST_HEAD(&pmu->aliases); + list_splice(&format, &pmu->format); + list_splice(&aliases, &pmu->aliases); + pmu->name = strdup(name); + pmu->type = type; + list_add_tail(&pmu->list, &pmus); + return pmu; +} + +static struct perf_pmu *pmu_find(char *name) +{ + struct perf_pmu *pmu; + + list_for_each_entry(pmu, &pmus, list) + if (!strcmp(pmu->name, name)) + return pmu; + + return NULL; +} + +struct perf_pmu *perf_pmu__scan(struct perf_pmu *pmu) +{ + /* + * pmu iterator: If pmu is NULL, we start at the begin, + * otherwise return the next pmu. Returns NULL on end. + */ + if (!pmu) { + pmu_read_sysfs(); + pmu = list_prepare_entry(pmu, &pmus, list); + } + list_for_each_entry_continue(pmu, &pmus, list) + return pmu; + return NULL; +} + +struct perf_pmu *perf_pmu__find(char *name) +{ + struct perf_pmu *pmu; + + /* + * Once PMU is loaded it stays in the list, + * so we keep us from multiple reading/parsing + * the pmu format definitions. + */ + pmu = pmu_find(name); + if (pmu) + return pmu; + + return pmu_lookup(name); +} + +static struct perf_pmu_format * +pmu_find_format(struct list_head *formats, char *name) +{ + struct perf_pmu_format *format; + + list_for_each_entry(format, formats, list) + if (!strcmp(format->name, name)) + return format; + + return NULL; +} + +/* + * Returns value based on the format definition (format parameter) + * and unformated value (value parameter). + * + * TODO maybe optimize a little ;) + */ +static __u64 pmu_format_value(unsigned long *format, __u64 value) +{ + unsigned long fbit, vbit; + __u64 v = 0; + + for (fbit = 0, vbit = 0; fbit < PERF_PMU_FORMAT_BITS; fbit++) { + + if (!test_bit(fbit, format)) + continue; + + if (!(value & (1llu << vbit++))) + continue; + + v |= (1llu << fbit); + } + + return v; +} + +/* + * Setup one of config[12] attr members based on the + * user input data - temr parameter. + */ +static int pmu_config_term(struct list_head *formats, + struct perf_event_attr *attr, + struct parse_events_term *term) +{ + struct perf_pmu_format *format; + __u64 *vp; + + /* + * Support only for hardcoded and numnerial terms. + * Hardcoded terms should be already in, so nothing + * to be done for them. + */ + if (parse_events__is_hardcoded_term(term)) + return 0; + + if (term->type_val != PARSE_EVENTS__TERM_TYPE_NUM) + return -EINVAL; + + format = pmu_find_format(formats, term->config); + if (!format) + return -EINVAL; + + switch (format->value) { + case PERF_PMU_FORMAT_VALUE_CONFIG: + vp = &attr->config; + break; + case PERF_PMU_FORMAT_VALUE_CONFIG1: + vp = &attr->config1; + break; + case PERF_PMU_FORMAT_VALUE_CONFIG2: + vp = &attr->config2; + break; + default: + return -EINVAL; + } + + /* + * XXX If we ever decide to go with string values for + * non-hardcoded terms, here's the place to translate + * them into value. + */ + *vp |= pmu_format_value(format->bits, term->val.num); + return 0; +} + +int perf_pmu__config_terms(struct list_head *formats, + struct perf_event_attr *attr, + struct list_head *head_terms) +{ + struct parse_events_term *term; + + list_for_each_entry(term, head_terms, list) + if (pmu_config_term(formats, attr, term)) + return -EINVAL; + + return 0; +} + +/* + * Configures event's 'attr' parameter based on the: + * 1) users input - specified in terms parameter + * 2) pmu format definitions - specified by pmu parameter + */ +int perf_pmu__config(struct perf_pmu *pmu, struct perf_event_attr *attr, + struct list_head *head_terms) +{ + attr->type = pmu->type; + return perf_pmu__config_terms(&pmu->format, attr, head_terms); +} + +static struct perf_pmu_alias *pmu_find_alias(struct perf_pmu *pmu, + struct parse_events_term *term) +{ + struct perf_pmu_alias *alias; + char *name; + + if (parse_events__is_hardcoded_term(term)) + return NULL; + + if (term->type_val == PARSE_EVENTS__TERM_TYPE_NUM) { + if (term->val.num != 1) + return NULL; + if (pmu_find_format(&pmu->format, term->config)) + return NULL; + name = term->config; + } else if (term->type_val == PARSE_EVENTS__TERM_TYPE_STR) { + if (strcasecmp(term->config, "event")) + return NULL; + name = term->val.str; + } else { + return NULL; + } + + list_for_each_entry(alias, &pmu->aliases, list) { + if (!strcasecmp(alias->name, name)) + return alias; + } + return NULL; +} + +/* + * Find alias in the terms list and replace it with the terms + * defined for the alias + */ +int perf_pmu__check_alias(struct perf_pmu *pmu, struct list_head *head_terms) +{ + struct parse_events_term *term, *h; + struct perf_pmu_alias *alias; + int ret; + + list_for_each_entry_safe(term, h, head_terms, list) { + alias = pmu_find_alias(pmu, term); + if (!alias) + continue; + ret = pmu_alias_terms(alias, &term->list); + if (ret) + return ret; + list_del(&term->list); + free(term); + } + return 0; +} + +int perf_pmu__new_format(struct list_head *list, char *name, + int config, unsigned long *bits) +{ + struct perf_pmu_format *format; + + format = zalloc(sizeof(*format)); + if (!format) + return -ENOMEM; + + format->name = strdup(name); + format->value = config; + memcpy(format->bits, bits, sizeof(format->bits)); + + list_add_tail(&format->list, list); + return 0; +} + +void perf_pmu__set_format(unsigned long *bits, long from, long to) +{ + long b; + + if (!to) + to = from; + + memset(bits, 0, BITS_TO_BYTES(PERF_PMU_FORMAT_BITS)); + for (b = from; b <= to; b++) + set_bit(b, bits); +} diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h new file mode 100644 index 00000000..32fe55b6 --- /dev/null +++ b/tools/perf/util/pmu.h @@ -0,0 +1,44 @@ +#ifndef __PMU_H +#define __PMU_H + +#include <linux/bitops.h> +#include <linux/perf_event.h> + +enum { + PERF_PMU_FORMAT_VALUE_CONFIG, + PERF_PMU_FORMAT_VALUE_CONFIG1, + PERF_PMU_FORMAT_VALUE_CONFIG2, +}; + +#define PERF_PMU_FORMAT_BITS 64 + +struct perf_pmu { + char *name; + __u32 type; + struct cpu_map *cpus; + struct list_head format; + struct list_head aliases; + struct list_head list; +}; + +struct perf_pmu *perf_pmu__find(char *name); +int perf_pmu__config(struct perf_pmu *pmu, struct perf_event_attr *attr, + struct list_head *head_terms); +int perf_pmu__config_terms(struct list_head *formats, + struct perf_event_attr *attr, + struct list_head *head_terms); +int perf_pmu__check_alias(struct perf_pmu *pmu, struct list_head *head_terms); +struct list_head *perf_pmu__alias(struct perf_pmu *pmu, + struct list_head *head_terms); +int perf_pmu_wrap(void); +void perf_pmu_error(struct list_head *list, char *name, char const *msg); + +int perf_pmu__new_format(struct list_head *list, char *name, + int config, unsigned long *bits); +void perf_pmu__set_format(unsigned long *bits, long from, long to); +int perf_pmu__format_parse(char *dir, struct list_head *head); + +struct perf_pmu *perf_pmu__scan(struct perf_pmu *pmu); + +int perf_pmu__test(void); +#endif /* __PMU_H */ diff --git a/tools/perf/util/pmu.l b/tools/perf/util/pmu.l new file mode 100644 index 00000000..a15d9fbd --- /dev/null +++ b/tools/perf/util/pmu.l @@ -0,0 +1,43 @@ +%option prefix="perf_pmu_" + +%{ +#include <stdlib.h> +#include <linux/bitops.h> +#include "pmu.h" +#include "pmu-bison.h" + +static int value(int base) +{ + long num; + + errno = 0; + num = strtoul(perf_pmu_text, NULL, base); + if (errno) + return PP_ERROR; + + perf_pmu_lval.num = num; + return PP_VALUE; +} + +%} + +num_dec [0-9]+ + +%% + +{num_dec} { return value(10); } +config { return PP_CONFIG; } +config1 { return PP_CONFIG1; } +config2 { return PP_CONFIG2; } +- { return '-'; } +: { return ':'; } +, { return ','; } +. { ; } +\n { ; } + +%% + +int perf_pmu_wrap(void) +{ + return 1; +} diff --git a/tools/perf/util/pmu.y b/tools/perf/util/pmu.y new file mode 100644 index 00000000..bfd7e850 --- /dev/null +++ b/tools/perf/util/pmu.y @@ -0,0 +1,92 @@ + +%parse-param {struct list_head *format} +%parse-param {char *name} + +%{ + +#include <linux/compiler.h> +#include <linux/list.h> +#include <linux/bitmap.h> +#include <string.h> +#include "pmu.h" + +extern int perf_pmu_lex (void); + +#define ABORT_ON(val) \ +do { \ + if (val) \ + YYABORT; \ +} while (0) + +%} + +%token PP_CONFIG PP_CONFIG1 PP_CONFIG2 +%token PP_VALUE PP_ERROR +%type <num> PP_VALUE +%type <bits> bit_term +%type <bits> bits + +%union +{ + unsigned long num; + DECLARE_BITMAP(bits, PERF_PMU_FORMAT_BITS); +} + +%% + +format: +format format_term +| +format_term + +format_term: +PP_CONFIG ':' bits +{ + ABORT_ON(perf_pmu__new_format(format, name, + PERF_PMU_FORMAT_VALUE_CONFIG, + $3)); +} +| +PP_CONFIG1 ':' bits +{ + ABORT_ON(perf_pmu__new_format(format, name, + PERF_PMU_FORMAT_VALUE_CONFIG1, + $3)); +} +| +PP_CONFIG2 ':' bits +{ + ABORT_ON(perf_pmu__new_format(format, name, + PERF_PMU_FORMAT_VALUE_CONFIG2, + $3)); +} + +bits: +bits ',' bit_term +{ + bitmap_or($$, $1, $3, 64); +} +| +bit_term +{ + memcpy($$, $1, sizeof($1)); +} + +bit_term: +PP_VALUE '-' PP_VALUE +{ + perf_pmu__set_format($$, $1, $3); +} +| +PP_VALUE +{ + perf_pmu__set_format($$, $1, 0); +} + +%% + +void perf_pmu_error(struct list_head *list __maybe_unused, + char *name __maybe_unused, + char const *msg __maybe_unused) +{ +} diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c new file mode 100644 index 00000000..49a256e6 --- /dev/null +++ b/tools/perf/util/probe-event.c @@ -0,0 +1,2374 @@ +/* + * probe-event.c : perf-probe definition to probe_events format converter + * + * Written by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ + +#include <sys/utsname.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <errno.h> +#include <stdio.h> +#include <unistd.h> +#include <stdlib.h> +#include <string.h> +#include <stdarg.h> +#include <limits.h> +#include <elf.h> + +#include "util.h" +#include "event.h" +#include "strlist.h" +#include "debug.h" +#include "cache.h" +#include "color.h" +#include "symbol.h" +#include "thread.h" +#include "debugfs.h" +#include "trace-event.h" /* For __maybe_unused */ +#include "probe-event.h" +#include "probe-finder.h" +#include "session.h" + +#define MAX_CMDLEN 256 +#define MAX_PROBE_ARGS 128 +#define PERFPROBE_GROUP "probe" + +bool probe_event_dry_run; /* Dry run flag */ + +#define semantic_error(msg ...) pr_err("Semantic error :" msg) + +/* If there is no space to write, returns -E2BIG. */ +static int e_snprintf(char *str, size_t size, const char *format, ...) + __attribute__((format(printf, 3, 4))); + +static int e_snprintf(char *str, size_t size, const char *format, ...) +{ + int ret; + va_list ap; + va_start(ap, format); + ret = vsnprintf(str, size, format, ap); + va_end(ap); + if (ret >= (int)size) + ret = -E2BIG; + return ret; +} + +static char *synthesize_perf_probe_point(struct perf_probe_point *pp); +static int convert_name_to_addr(struct perf_probe_event *pev, + const char *exec); +static struct machine machine; + +/* Initialize symbol maps and path of vmlinux/modules */ +static int init_vmlinux(void) +{ + int ret; + + symbol_conf.sort_by_name = true; + if (symbol_conf.vmlinux_name == NULL) + symbol_conf.try_vmlinux_path = true; + else + pr_debug("Use vmlinux: %s\n", symbol_conf.vmlinux_name); + ret = symbol__init(); + if (ret < 0) { + pr_debug("Failed to init symbol map.\n"); + goto out; + } + + ret = machine__init(&machine, "", HOST_KERNEL_ID); + if (ret < 0) + goto out; + + if (machine__create_kernel_maps(&machine) < 0) { + pr_debug("machine__create_kernel_maps() failed.\n"); + goto out; + } +out: + if (ret < 0) + pr_warning("Failed to init vmlinux path.\n"); + return ret; +} + +static struct symbol *__find_kernel_function_by_name(const char *name, + struct map **mapp) +{ + return machine__find_kernel_function_by_name(&machine, name, mapp, + NULL); +} + +static struct map *kernel_get_module_map(const char *module) +{ + struct rb_node *nd; + struct map_groups *grp = &machine.kmaps; + + /* A file path -- this is an offline module */ + if (module && strchr(module, '/')) + return machine__new_module(&machine, 0, module); + + if (!module) + module = "kernel"; + + for (nd = rb_first(&grp->maps[MAP__FUNCTION]); nd; nd = rb_next(nd)) { + struct map *pos = rb_entry(nd, struct map, rb_node); + if (strncmp(pos->dso->short_name + 1, module, + pos->dso->short_name_len - 2) == 0) { + return pos; + } + } + return NULL; +} + +static struct dso *kernel_get_module_dso(const char *module) +{ + struct dso *dso; + struct map *map; + const char *vmlinux_name; + + if (module) { + list_for_each_entry(dso, &machine.kernel_dsos, node) { + if (strncmp(dso->short_name + 1, module, + dso->short_name_len - 2) == 0) + goto found; + } + pr_debug("Failed to find module %s.\n", module); + return NULL; + } + + map = machine.vmlinux_maps[MAP__FUNCTION]; + dso = map->dso; + + vmlinux_name = symbol_conf.vmlinux_name; + if (vmlinux_name) { + if (dso__load_vmlinux(dso, map, vmlinux_name, NULL) <= 0) + return NULL; + } else { + if (dso__load_vmlinux_path(dso, map, NULL) <= 0) { + pr_debug("Failed to load kernel map.\n"); + return NULL; + } + } +found: + return dso; +} + +const char *kernel_get_module_path(const char *module) +{ + struct dso *dso = kernel_get_module_dso(module); + return (dso) ? dso->long_name : NULL; +} + +static int init_user_exec(void) +{ + int ret = 0; + + symbol_conf.try_vmlinux_path = false; + symbol_conf.sort_by_name = true; + ret = symbol__init(); + + if (ret < 0) + pr_debug("Failed to init symbol map.\n"); + + return ret; +} + +static int convert_to_perf_probe_point(struct probe_trace_point *tp, + struct perf_probe_point *pp) +{ + pp->function = strdup(tp->symbol); + + if (pp->function == NULL) + return -ENOMEM; + + pp->offset = tp->offset; + pp->retprobe = tp->retprobe; + + return 0; +} + +#ifdef DWARF_SUPPORT +/* Open new debuginfo of given module */ +static struct debuginfo *open_debuginfo(const char *module) +{ + const char *path; + + /* A file path -- this is an offline module */ + if (module && strchr(module, '/')) + path = module; + else { + path = kernel_get_module_path(module); + + if (!path) { + pr_err("Failed to find path of %s module.\n", + module ?: "kernel"); + return NULL; + } + } + return debuginfo__new(path); +} + +/* + * Convert trace point to probe point with debuginfo + * Currently only handles kprobes. + */ +static int kprobe_convert_to_perf_probe(struct probe_trace_point *tp, + struct perf_probe_point *pp) +{ + struct symbol *sym; + struct map *map; + u64 addr; + int ret = -ENOENT; + struct debuginfo *dinfo; + + sym = __find_kernel_function_by_name(tp->symbol, &map); + if (sym) { + addr = map->unmap_ip(map, sym->start + tp->offset); + pr_debug("try to find %s+%ld@%" PRIx64 "\n", tp->symbol, + tp->offset, addr); + + dinfo = debuginfo__new_online_kernel(addr); + if (dinfo) { + ret = debuginfo__find_probe_point(dinfo, + (unsigned long)addr, pp); + debuginfo__delete(dinfo); + } else { + pr_debug("Failed to open debuginfo at 0x%" PRIx64 "\n", + addr); + ret = -ENOENT; + } + } + if (ret <= 0) { + pr_debug("Failed to find corresponding probes from " + "debuginfo. Use kprobe event information.\n"); + return convert_to_perf_probe_point(tp, pp); + } + pp->retprobe = tp->retprobe; + + return 0; +} + +static int add_module_to_probe_trace_events(struct probe_trace_event *tevs, + int ntevs, const char *module) +{ + int i, ret = 0; + char *tmp; + + if (!module) + return 0; + + tmp = strrchr(module, '/'); + if (tmp) { + /* This is a module path -- get the module name */ + module = strdup(tmp + 1); + if (!module) + return -ENOMEM; + tmp = strchr(module, '.'); + if (tmp) + *tmp = '\0'; + tmp = (char *)module; /* For free() */ + } + + for (i = 0; i < ntevs; i++) { + tevs[i].point.module = strdup(module); + if (!tevs[i].point.module) { + ret = -ENOMEM; + break; + } + } + + if (tmp) + free(tmp); + + return ret; +} + +/* Try to find perf_probe_event with debuginfo */ +static int try_to_find_probe_trace_events(struct perf_probe_event *pev, + struct probe_trace_event **tevs, + int max_tevs, const char *target) +{ + bool need_dwarf = perf_probe_event_need_dwarf(pev); + struct debuginfo *dinfo; + int ntevs, ret = 0; + + if (pev->uprobes) { + if (need_dwarf) { + pr_warning("Debuginfo-analysis is not yet supported" + " with -x/--exec option.\n"); + return -ENOSYS; + } + return convert_name_to_addr(pev, target); + } + + dinfo = open_debuginfo(target); + + if (!dinfo) { + if (need_dwarf) { + pr_warning("Failed to open debuginfo file.\n"); + return -ENOENT; + } + pr_debug("Could not open debuginfo. Try to use symbols.\n"); + return 0; + } + + /* Searching trace events corresponding to a probe event */ + ntevs = debuginfo__find_trace_events(dinfo, pev, tevs, max_tevs); + + debuginfo__delete(dinfo); + + if (ntevs > 0) { /* Succeeded to find trace events */ + pr_debug("find %d probe_trace_events.\n", ntevs); + if (target) + ret = add_module_to_probe_trace_events(*tevs, ntevs, + target); + return ret < 0 ? ret : ntevs; + } + + if (ntevs == 0) { /* No error but failed to find probe point. */ + pr_warning("Probe point '%s' not found.\n", + synthesize_perf_probe_point(&pev->point)); + return -ENOENT; + } + /* Error path : ntevs < 0 */ + pr_debug("An error occurred in debuginfo analysis (%d).\n", ntevs); + if (ntevs == -EBADF) { + pr_warning("Warning: No dwarf info found in the vmlinux - " + "please rebuild kernel with CONFIG_DEBUG_INFO=y.\n"); + if (!need_dwarf) { + pr_debug("Trying to use symbols.\n"); + return 0; + } + } + return ntevs; +} + +/* + * Find a src file from a DWARF tag path. Prepend optional source path prefix + * and chop off leading directories that do not exist. Result is passed back as + * a newly allocated path on success. + * Return 0 if file was found and readable, -errno otherwise. + */ +static int get_real_path(const char *raw_path, const char *comp_dir, + char **new_path) +{ + const char *prefix = symbol_conf.source_prefix; + + if (!prefix) { + if (raw_path[0] != '/' && comp_dir) + /* If not an absolute path, try to use comp_dir */ + prefix = comp_dir; + else { + if (access(raw_path, R_OK) == 0) { + *new_path = strdup(raw_path); + return 0; + } else + return -errno; + } + } + + *new_path = malloc((strlen(prefix) + strlen(raw_path) + 2)); + if (!*new_path) + return -ENOMEM; + + for (;;) { + sprintf(*new_path, "%s/%s", prefix, raw_path); + + if (access(*new_path, R_OK) == 0) + return 0; + + if (!symbol_conf.source_prefix) + /* In case of searching comp_dir, don't retry */ + return -errno; + + switch (errno) { + case ENAMETOOLONG: + case ENOENT: + case EROFS: + case EFAULT: + raw_path = strchr(++raw_path, '/'); + if (!raw_path) { + free(*new_path); + *new_path = NULL; + return -ENOENT; + } + continue; + + default: + free(*new_path); + *new_path = NULL; + return -errno; + } + } +} + +#define LINEBUF_SIZE 256 +#define NR_ADDITIONAL_LINES 2 + +static int __show_one_line(FILE *fp, int l, bool skip, bool show_num) +{ + char buf[LINEBUF_SIZE]; + const char *color = show_num ? "" : PERF_COLOR_BLUE; + const char *prefix = NULL; + + do { + if (fgets(buf, LINEBUF_SIZE, fp) == NULL) + goto error; + if (skip) + continue; + if (!prefix) { + prefix = show_num ? "%7d " : " "; + color_fprintf(stdout, color, prefix, l); + } + color_fprintf(stdout, color, "%s", buf); + + } while (strchr(buf, '\n') == NULL); + + return 1; +error: + if (ferror(fp)) { + pr_warning("File read error: %s\n", strerror(errno)); + return -1; + } + return 0; +} + +static int _show_one_line(FILE *fp, int l, bool skip, bool show_num) +{ + int rv = __show_one_line(fp, l, skip, show_num); + if (rv == 0) { + pr_warning("Source file is shorter than expected.\n"); + rv = -1; + } + return rv; +} + +#define show_one_line_with_num(f,l) _show_one_line(f,l,false,true) +#define show_one_line(f,l) _show_one_line(f,l,false,false) +#define skip_one_line(f,l) _show_one_line(f,l,true,false) +#define show_one_line_or_eof(f,l) __show_one_line(f,l,false,false) + +/* + * Show line-range always requires debuginfo to find source file and + * line number. + */ +int show_line_range(struct line_range *lr, const char *module) +{ + int l = 1; + struct line_node *ln; + struct debuginfo *dinfo; + FILE *fp; + int ret; + char *tmp; + + /* Search a line range */ + ret = init_vmlinux(); + if (ret < 0) + return ret; + + dinfo = open_debuginfo(module); + if (!dinfo) { + pr_warning("Failed to open debuginfo file.\n"); + return -ENOENT; + } + + ret = debuginfo__find_line_range(dinfo, lr); + debuginfo__delete(dinfo); + if (ret == 0) { + pr_warning("Specified source line is not found.\n"); + return -ENOENT; + } else if (ret < 0) { + pr_warning("Debuginfo analysis failed. (%d)\n", ret); + return ret; + } + + /* Convert source file path */ + tmp = lr->path; + ret = get_real_path(tmp, lr->comp_dir, &lr->path); + free(tmp); /* Free old path */ + if (ret < 0) { + pr_warning("Failed to find source file. (%d)\n", ret); + return ret; + } + + setup_pager(); + + if (lr->function) + fprintf(stdout, "<%s@%s:%d>\n", lr->function, lr->path, + lr->start - lr->offset); + else + fprintf(stdout, "<%s:%d>\n", lr->path, lr->start); + + fp = fopen(lr->path, "r"); + if (fp == NULL) { + pr_warning("Failed to open %s: %s\n", lr->path, + strerror(errno)); + return -errno; + } + /* Skip to starting line number */ + while (l < lr->start) { + ret = skip_one_line(fp, l++); + if (ret < 0) + goto end; + } + + list_for_each_entry(ln, &lr->line_list, list) { + for (; ln->line > l; l++) { + ret = show_one_line(fp, l - lr->offset); + if (ret < 0) + goto end; + } + ret = show_one_line_with_num(fp, l++ - lr->offset); + if (ret < 0) + goto end; + } + + if (lr->end == INT_MAX) + lr->end = l + NR_ADDITIONAL_LINES; + while (l <= lr->end) { + ret = show_one_line_or_eof(fp, l++ - lr->offset); + if (ret <= 0) + break; + } +end: + fclose(fp); + return ret; +} + +static int show_available_vars_at(struct debuginfo *dinfo, + struct perf_probe_event *pev, + int max_vls, struct strfilter *_filter, + bool externs) +{ + char *buf; + int ret, i, nvars; + struct str_node *node; + struct variable_list *vls = NULL, *vl; + const char *var; + + buf = synthesize_perf_probe_point(&pev->point); + if (!buf) + return -EINVAL; + pr_debug("Searching variables at %s\n", buf); + + ret = debuginfo__find_available_vars_at(dinfo, pev, &vls, + max_vls, externs); + if (ret <= 0) { + pr_err("Failed to find variables at %s (%d)\n", buf, ret); + goto end; + } + /* Some variables are found */ + fprintf(stdout, "Available variables at %s\n", buf); + for (i = 0; i < ret; i++) { + vl = &vls[i]; + /* + * A probe point might be converted to + * several trace points. + */ + fprintf(stdout, "\t@<%s+%lu>\n", vl->point.symbol, + vl->point.offset); + free(vl->point.symbol); + nvars = 0; + if (vl->vars) { + strlist__for_each(node, vl->vars) { + var = strchr(node->s, '\t') + 1; + if (strfilter__compare(_filter, var)) { + fprintf(stdout, "\t\t%s\n", node->s); + nvars++; + } + } + strlist__delete(vl->vars); + } + if (nvars == 0) + fprintf(stdout, "\t\t(No matched variables)\n"); + } + free(vls); +end: + free(buf); + return ret; +} + +/* Show available variables on given probe point */ +int show_available_vars(struct perf_probe_event *pevs, int npevs, + int max_vls, const char *module, + struct strfilter *_filter, bool externs) +{ + int i, ret = 0; + struct debuginfo *dinfo; + + ret = init_vmlinux(); + if (ret < 0) + return ret; + + dinfo = open_debuginfo(module); + if (!dinfo) { + pr_warning("Failed to open debuginfo file.\n"); + return -ENOENT; + } + + setup_pager(); + + for (i = 0; i < npevs && ret >= 0; i++) + ret = show_available_vars_at(dinfo, &pevs[i], max_vls, _filter, + externs); + + debuginfo__delete(dinfo); + return ret; +} + +#else /* !DWARF_SUPPORT */ + +static int kprobe_convert_to_perf_probe(struct probe_trace_point *tp, + struct perf_probe_point *pp) +{ + struct symbol *sym; + + sym = __find_kernel_function_by_name(tp->symbol, NULL); + if (!sym) { + pr_err("Failed to find symbol %s in kernel.\n", tp->symbol); + return -ENOENT; + } + + return convert_to_perf_probe_point(tp, pp); +} + +static int try_to_find_probe_trace_events(struct perf_probe_event *pev, + struct probe_trace_event **tevs __maybe_unused, + int max_tevs __maybe_unused, const char *target) +{ + if (perf_probe_event_need_dwarf(pev)) { + pr_warning("Debuginfo-analysis is not supported.\n"); + return -ENOSYS; + } + + if (pev->uprobes) + return convert_name_to_addr(pev, target); + + return 0; +} + +int show_line_range(struct line_range *lr __maybe_unused, + const char *module __maybe_unused) +{ + pr_warning("Debuginfo-analysis is not supported.\n"); + return -ENOSYS; +} + +int show_available_vars(struct perf_probe_event *pevs __maybe_unused, + int npevs __maybe_unused, int max_vls __maybe_unused, + const char *module __maybe_unused, + struct strfilter *filter __maybe_unused, + bool externs __maybe_unused) +{ + pr_warning("Debuginfo-analysis is not supported.\n"); + return -ENOSYS; +} +#endif + +static int parse_line_num(char **ptr, int *val, const char *what) +{ + const char *start = *ptr; + + errno = 0; + *val = strtol(*ptr, ptr, 0); + if (errno || *ptr == start) { + semantic_error("'%s' is not a valid number.\n", what); + return -EINVAL; + } + return 0; +} + +/* + * Stuff 'lr' according to the line range described by 'arg'. + * The line range syntax is described by: + * + * SRC[:SLN[+NUM|-ELN]] + * FNC[@SRC][:SLN[+NUM|-ELN]] + */ +int parse_line_range_desc(const char *arg, struct line_range *lr) +{ + char *range, *file, *name = strdup(arg); + int err; + + if (!name) + return -ENOMEM; + + lr->start = 0; + lr->end = INT_MAX; + + range = strchr(name, ':'); + if (range) { + *range++ = '\0'; + + err = parse_line_num(&range, &lr->start, "start line"); + if (err) + goto err; + + if (*range == '+' || *range == '-') { + const char c = *range++; + + err = parse_line_num(&range, &lr->end, "end line"); + if (err) + goto err; + + if (c == '+') { + lr->end += lr->start; + /* + * Adjust the number of lines here. + * If the number of lines == 1, the + * the end of line should be equal to + * the start of line. + */ + lr->end--; + } + } + + pr_debug("Line range is %d to %d\n", lr->start, lr->end); + + err = -EINVAL; + if (lr->start > lr->end) { + semantic_error("Start line must be smaller" + " than end line.\n"); + goto err; + } + if (*range != '\0') { + semantic_error("Tailing with invalid str '%s'.\n", range); + goto err; + } + } + + file = strchr(name, '@'); + if (file) { + *file = '\0'; + lr->file = strdup(++file); + if (lr->file == NULL) { + err = -ENOMEM; + goto err; + } + lr->function = name; + } else if (strchr(name, '.')) + lr->file = name; + else + lr->function = name; + + return 0; +err: + free(name); + return err; +} + +/* Check the name is good for event/group */ +static bool check_event_name(const char *name) +{ + if (!isalpha(*name) && *name != '_') + return false; + while (*++name != '\0') { + if (!isalpha(*name) && !isdigit(*name) && *name != '_') + return false; + } + return true; +} + +/* Parse probepoint definition. */ +static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev) +{ + struct perf_probe_point *pp = &pev->point; + char *ptr, *tmp; + char c, nc = 0; + /* + * <Syntax> + * perf probe [EVENT=]SRC[:LN|;PTN] + * perf probe [EVENT=]FUNC[@SRC][+OFFS|%return|:LN|;PAT] + * + * TODO:Group name support + */ + + ptr = strpbrk(arg, ";=@+%"); + if (ptr && *ptr == '=') { /* Event name */ + *ptr = '\0'; + tmp = ptr + 1; + if (strchr(arg, ':')) { + semantic_error("Group name is not supported yet.\n"); + return -ENOTSUP; + } + if (!check_event_name(arg)) { + semantic_error("%s is bad for event name -it must " + "follow C symbol-naming rule.\n", arg); + return -EINVAL; + } + pev->event = strdup(arg); + if (pev->event == NULL) + return -ENOMEM; + pev->group = NULL; + arg = tmp; + } + + ptr = strpbrk(arg, ";:+@%"); + if (ptr) { + nc = *ptr; + *ptr++ = '\0'; + } + + tmp = strdup(arg); + if (tmp == NULL) + return -ENOMEM; + + /* Check arg is function or file and copy it */ + if (strchr(tmp, '.')) /* File */ + pp->file = tmp; + else /* Function */ + pp->function = tmp; + + /* Parse other options */ + while (ptr) { + arg = ptr; + c = nc; + if (c == ';') { /* Lazy pattern must be the last part */ + pp->lazy_line = strdup(arg); + if (pp->lazy_line == NULL) + return -ENOMEM; + break; + } + ptr = strpbrk(arg, ";:+@%"); + if (ptr) { + nc = *ptr; + *ptr++ = '\0'; + } + switch (c) { + case ':': /* Line number */ + pp->line = strtoul(arg, &tmp, 0); + if (*tmp != '\0') { + semantic_error("There is non-digit char" + " in line number.\n"); + return -EINVAL; + } + break; + case '+': /* Byte offset from a symbol */ + pp->offset = strtoul(arg, &tmp, 0); + if (*tmp != '\0') { + semantic_error("There is non-digit character" + " in offset.\n"); + return -EINVAL; + } + break; + case '@': /* File name */ + if (pp->file) { + semantic_error("SRC@SRC is not allowed.\n"); + return -EINVAL; + } + pp->file = strdup(arg); + if (pp->file == NULL) + return -ENOMEM; + break; + case '%': /* Probe places */ + if (strcmp(arg, "return") == 0) { + pp->retprobe = 1; + } else { /* Others not supported yet */ + semantic_error("%%%s is not supported.\n", arg); + return -ENOTSUP; + } + break; + default: /* Buggy case */ + pr_err("This program has a bug at %s:%d.\n", + __FILE__, __LINE__); + return -ENOTSUP; + break; + } + } + + /* Exclusion check */ + if (pp->lazy_line && pp->line) { + semantic_error("Lazy pattern can't be used with" + " line number.\n"); + return -EINVAL; + } + + if (pp->lazy_line && pp->offset) { + semantic_error("Lazy pattern can't be used with offset.\n"); + return -EINVAL; + } + + if (pp->line && pp->offset) { + semantic_error("Offset can't be used with line number.\n"); + return -EINVAL; + } + + if (!pp->line && !pp->lazy_line && pp->file && !pp->function) { + semantic_error("File always requires line number or " + "lazy pattern.\n"); + return -EINVAL; + } + + if (pp->offset && !pp->function) { + semantic_error("Offset requires an entry function.\n"); + return -EINVAL; + } + + if (pp->retprobe && !pp->function) { + semantic_error("Return probe requires an entry function.\n"); + return -EINVAL; + } + + if ((pp->offset || pp->line || pp->lazy_line) && pp->retprobe) { + semantic_error("Offset/Line/Lazy pattern can't be used with " + "return probe.\n"); + return -EINVAL; + } + + pr_debug("symbol:%s file:%s line:%d offset:%lu return:%d lazy:%s\n", + pp->function, pp->file, pp->line, pp->offset, pp->retprobe, + pp->lazy_line); + return 0; +} + +/* Parse perf-probe event argument */ +static int parse_perf_probe_arg(char *str, struct perf_probe_arg *arg) +{ + char *tmp, *goodname; + struct perf_probe_arg_field **fieldp; + + pr_debug("parsing arg: %s into ", str); + + tmp = strchr(str, '='); + if (tmp) { + arg->name = strndup(str, tmp - str); + if (arg->name == NULL) + return -ENOMEM; + pr_debug("name:%s ", arg->name); + str = tmp + 1; + } + + tmp = strchr(str, ':'); + if (tmp) { /* Type setting */ + *tmp = '\0'; + arg->type = strdup(tmp + 1); + if (arg->type == NULL) + return -ENOMEM; + pr_debug("type:%s ", arg->type); + } + + tmp = strpbrk(str, "-.["); + if (!is_c_varname(str) || !tmp) { + /* A variable, register, symbol or special value */ + arg->var = strdup(str); + if (arg->var == NULL) + return -ENOMEM; + pr_debug("%s\n", arg->var); + return 0; + } + + /* Structure fields or array element */ + arg->var = strndup(str, tmp - str); + if (arg->var == NULL) + return -ENOMEM; + goodname = arg->var; + pr_debug("%s, ", arg->var); + fieldp = &arg->field; + + do { + *fieldp = zalloc(sizeof(struct perf_probe_arg_field)); + if (*fieldp == NULL) + return -ENOMEM; + if (*tmp == '[') { /* Array */ + str = tmp; + (*fieldp)->index = strtol(str + 1, &tmp, 0); + (*fieldp)->ref = true; + if (*tmp != ']' || tmp == str + 1) { + semantic_error("Array index must be a" + " number.\n"); + return -EINVAL; + } + tmp++; + if (*tmp == '\0') + tmp = NULL; + } else { /* Structure */ + if (*tmp == '.') { + str = tmp + 1; + (*fieldp)->ref = false; + } else if (tmp[1] == '>') { + str = tmp + 2; + (*fieldp)->ref = true; + } else { + semantic_error("Argument parse error: %s\n", + str); + return -EINVAL; + } + tmp = strpbrk(str, "-.["); + } + if (tmp) { + (*fieldp)->name = strndup(str, tmp - str); + if ((*fieldp)->name == NULL) + return -ENOMEM; + if (*str != '[') + goodname = (*fieldp)->name; + pr_debug("%s(%d), ", (*fieldp)->name, (*fieldp)->ref); + fieldp = &(*fieldp)->next; + } + } while (tmp); + (*fieldp)->name = strdup(str); + if ((*fieldp)->name == NULL) + return -ENOMEM; + if (*str != '[') + goodname = (*fieldp)->name; + pr_debug("%s(%d)\n", (*fieldp)->name, (*fieldp)->ref); + + /* If no name is specified, set the last field name (not array index)*/ + if (!arg->name) { + arg->name = strdup(goodname); + if (arg->name == NULL) + return -ENOMEM; + } + return 0; +} + +/* Parse perf-probe event command */ +int parse_perf_probe_command(const char *cmd, struct perf_probe_event *pev) +{ + char **argv; + int argc, i, ret = 0; + + argv = argv_split(cmd, &argc); + if (!argv) { + pr_debug("Failed to split arguments.\n"); + return -ENOMEM; + } + if (argc - 1 > MAX_PROBE_ARGS) { + semantic_error("Too many probe arguments (%d).\n", argc - 1); + ret = -ERANGE; + goto out; + } + /* Parse probe point */ + ret = parse_perf_probe_point(argv[0], pev); + if (ret < 0) + goto out; + + /* Copy arguments and ensure return probe has no C argument */ + pev->nargs = argc - 1; + pev->args = zalloc(sizeof(struct perf_probe_arg) * pev->nargs); + if (pev->args == NULL) { + ret = -ENOMEM; + goto out; + } + for (i = 0; i < pev->nargs && ret >= 0; i++) { + ret = parse_perf_probe_arg(argv[i + 1], &pev->args[i]); + if (ret >= 0 && + is_c_varname(pev->args[i].var) && pev->point.retprobe) { + semantic_error("You can't specify local variable for" + " kretprobe.\n"); + ret = -EINVAL; + } + } +out: + argv_free(argv); + + return ret; +} + +/* Return true if this perf_probe_event requires debuginfo */ +bool perf_probe_event_need_dwarf(struct perf_probe_event *pev) +{ + int i; + + if (pev->point.file || pev->point.line || pev->point.lazy_line) + return true; + + for (i = 0; i < pev->nargs; i++) + if (is_c_varname(pev->args[i].var)) + return true; + + return false; +} + +/* Parse probe_events event into struct probe_point */ +static int parse_probe_trace_command(const char *cmd, + struct probe_trace_event *tev) +{ + struct probe_trace_point *tp = &tev->point; + char pr; + char *p; + char *argv0_str = NULL, *fmt, *fmt1_str, *fmt2_str, *fmt3_str; + int ret, i, argc; + char **argv; + + pr_debug("Parsing probe_events: %s\n", cmd); + argv = argv_split(cmd, &argc); + if (!argv) { + pr_debug("Failed to split arguments.\n"); + return -ENOMEM; + } + if (argc < 2) { + semantic_error("Too few probe arguments.\n"); + ret = -ERANGE; + goto out; + } + + /* Scan event and group name. */ + argv0_str = strdup(argv[0]); + if (argv0_str == NULL) { + ret = -ENOMEM; + goto out; + } + fmt1_str = strtok_r(argv0_str, ":", &fmt); + fmt2_str = strtok_r(NULL, "/", &fmt); + fmt3_str = strtok_r(NULL, " \t", &fmt); + if (fmt1_str == NULL || strlen(fmt1_str) != 1 || fmt2_str == NULL + || fmt3_str == NULL) { + semantic_error("Failed to parse event name: %s\n", argv[0]); + ret = -EINVAL; + goto out; + } + pr = fmt1_str[0]; + tev->group = strdup(fmt2_str); + tev->event = strdup(fmt3_str); + if (tev->group == NULL || tev->event == NULL) { + ret = -ENOMEM; + goto out; + } + pr_debug("Group:%s Event:%s probe:%c\n", tev->group, tev->event, pr); + + tp->retprobe = (pr == 'r'); + + /* Scan module name(if there), function name and offset */ + p = strchr(argv[1], ':'); + if (p) { + tp->module = strndup(argv[1], p - argv[1]); + p++; + } else + p = argv[1]; + fmt1_str = strtok_r(p, "+", &fmt); + tp->symbol = strdup(fmt1_str); + if (tp->symbol == NULL) { + ret = -ENOMEM; + goto out; + } + fmt2_str = strtok_r(NULL, "", &fmt); + if (fmt2_str == NULL) + tp->offset = 0; + else + tp->offset = strtoul(fmt2_str, NULL, 10); + + tev->nargs = argc - 2; + tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs); + if (tev->args == NULL) { + ret = -ENOMEM; + goto out; + } + for (i = 0; i < tev->nargs; i++) { + p = strchr(argv[i + 2], '='); + if (p) /* We don't need which register is assigned. */ + *p++ = '\0'; + else + p = argv[i + 2]; + tev->args[i].name = strdup(argv[i + 2]); + /* TODO: parse regs and offset */ + tev->args[i].value = strdup(p); + if (tev->args[i].name == NULL || tev->args[i].value == NULL) { + ret = -ENOMEM; + goto out; + } + } + ret = 0; +out: + free(argv0_str); + argv_free(argv); + return ret; +} + +/* Compose only probe arg */ +int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf, size_t len) +{ + struct perf_probe_arg_field *field = pa->field; + int ret; + char *tmp = buf; + + if (pa->name && pa->var) + ret = e_snprintf(tmp, len, "%s=%s", pa->name, pa->var); + else + ret = e_snprintf(tmp, len, "%s", pa->name ? pa->name : pa->var); + if (ret <= 0) + goto error; + tmp += ret; + len -= ret; + + while (field) { + if (field->name[0] == '[') + ret = e_snprintf(tmp, len, "%s", field->name); + else + ret = e_snprintf(tmp, len, "%s%s", + field->ref ? "->" : ".", field->name); + if (ret <= 0) + goto error; + tmp += ret; + len -= ret; + field = field->next; + } + + if (pa->type) { + ret = e_snprintf(tmp, len, ":%s", pa->type); + if (ret <= 0) + goto error; + tmp += ret; + len -= ret; + } + + return tmp - buf; +error: + pr_debug("Failed to synthesize perf probe argument: %s\n", + strerror(-ret)); + return ret; +} + +/* Compose only probe point (not argument) */ +static char *synthesize_perf_probe_point(struct perf_probe_point *pp) +{ + char *buf, *tmp; + char offs[32] = "", line[32] = "", file[32] = ""; + int ret, len; + + buf = zalloc(MAX_CMDLEN); + if (buf == NULL) { + ret = -ENOMEM; + goto error; + } + if (pp->offset) { + ret = e_snprintf(offs, 32, "+%lu", pp->offset); + if (ret <= 0) + goto error; + } + if (pp->line) { + ret = e_snprintf(line, 32, ":%d", pp->line); + if (ret <= 0) + goto error; + } + if (pp->file) { + tmp = pp->file; + len = strlen(tmp); + if (len > 30) { + tmp = strchr(pp->file + len - 30, '/'); + tmp = tmp ? tmp + 1 : pp->file + len - 30; + } + ret = e_snprintf(file, 32, "@%s", tmp); + if (ret <= 0) + goto error; + } + + if (pp->function) + ret = e_snprintf(buf, MAX_CMDLEN, "%s%s%s%s%s", pp->function, + offs, pp->retprobe ? "%return" : "", line, + file); + else + ret = e_snprintf(buf, MAX_CMDLEN, "%s%s", file, line); + if (ret <= 0) + goto error; + + return buf; +error: + pr_debug("Failed to synthesize perf probe point: %s\n", + strerror(-ret)); + if (buf) + free(buf); + return NULL; +} + +#if 0 +char *synthesize_perf_probe_command(struct perf_probe_event *pev) +{ + char *buf; + int i, len, ret; + + buf = synthesize_perf_probe_point(&pev->point); + if (!buf) + return NULL; + + len = strlen(buf); + for (i = 0; i < pev->nargs; i++) { + ret = e_snprintf(&buf[len], MAX_CMDLEN - len, " %s", + pev->args[i].name); + if (ret <= 0) { + free(buf); + return NULL; + } + len += ret; + } + + return buf; +} +#endif + +static int __synthesize_probe_trace_arg_ref(struct probe_trace_arg_ref *ref, + char **buf, size_t *buflen, + int depth) +{ + int ret; + if (ref->next) { + depth = __synthesize_probe_trace_arg_ref(ref->next, buf, + buflen, depth + 1); + if (depth < 0) + goto out; + } + + ret = e_snprintf(*buf, *buflen, "%+ld(", ref->offset); + if (ret < 0) + depth = ret; + else { + *buf += ret; + *buflen -= ret; + } +out: + return depth; + +} + +static int synthesize_probe_trace_arg(struct probe_trace_arg *arg, + char *buf, size_t buflen) +{ + struct probe_trace_arg_ref *ref = arg->ref; + int ret, depth = 0; + char *tmp = buf; + + /* Argument name or separator */ + if (arg->name) + ret = e_snprintf(buf, buflen, " %s=", arg->name); + else + ret = e_snprintf(buf, buflen, " "); + if (ret < 0) + return ret; + buf += ret; + buflen -= ret; + + /* Special case: @XXX */ + if (arg->value[0] == '@' && arg->ref) + ref = ref->next; + + /* Dereferencing arguments */ + if (ref) { + depth = __synthesize_probe_trace_arg_ref(ref, &buf, + &buflen, 1); + if (depth < 0) + return depth; + } + + /* Print argument value */ + if (arg->value[0] == '@' && arg->ref) + ret = e_snprintf(buf, buflen, "%s%+ld", arg->value, + arg->ref->offset); + else + ret = e_snprintf(buf, buflen, "%s", arg->value); + if (ret < 0) + return ret; + buf += ret; + buflen -= ret; + + /* Closing */ + while (depth--) { + ret = e_snprintf(buf, buflen, ")"); + if (ret < 0) + return ret; + buf += ret; + buflen -= ret; + } + /* Print argument type */ + if (arg->type) { + ret = e_snprintf(buf, buflen, ":%s", arg->type); + if (ret <= 0) + return ret; + buf += ret; + } + + return buf - tmp; +} + +char *synthesize_probe_trace_command(struct probe_trace_event *tev) +{ + struct probe_trace_point *tp = &tev->point; + char *buf; + int i, len, ret; + + buf = zalloc(MAX_CMDLEN); + if (buf == NULL) + return NULL; + + if (tev->uprobes) + len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s:%s", + tp->retprobe ? 'r' : 'p', + tev->group, tev->event, + tp->module, tp->symbol); + else + len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s %s%s%s+%lu", + tp->retprobe ? 'r' : 'p', + tev->group, tev->event, + tp->module ?: "", tp->module ? ":" : "", + tp->symbol, tp->offset); + + if (len <= 0) + goto error; + + for (i = 0; i < tev->nargs; i++) { + ret = synthesize_probe_trace_arg(&tev->args[i], buf + len, + MAX_CMDLEN - len); + if (ret <= 0) + goto error; + len += ret; + } + + return buf; +error: + free(buf); + return NULL; +} + +static int convert_to_perf_probe_event(struct probe_trace_event *tev, + struct perf_probe_event *pev, bool is_kprobe) +{ + char buf[64] = ""; + int i, ret; + + /* Convert event/group name */ + pev->event = strdup(tev->event); + pev->group = strdup(tev->group); + if (pev->event == NULL || pev->group == NULL) + return -ENOMEM; + + /* Convert trace_point to probe_point */ + if (is_kprobe) + ret = kprobe_convert_to_perf_probe(&tev->point, &pev->point); + else + ret = convert_to_perf_probe_point(&tev->point, &pev->point); + + if (ret < 0) + return ret; + + /* Convert trace_arg to probe_arg */ + pev->nargs = tev->nargs; + pev->args = zalloc(sizeof(struct perf_probe_arg) * pev->nargs); + if (pev->args == NULL) + return -ENOMEM; + for (i = 0; i < tev->nargs && ret >= 0; i++) { + if (tev->args[i].name) + pev->args[i].name = strdup(tev->args[i].name); + else { + ret = synthesize_probe_trace_arg(&tev->args[i], + buf, 64); + pev->args[i].name = strdup(buf); + } + if (pev->args[i].name == NULL && ret >= 0) + ret = -ENOMEM; + } + + if (ret < 0) + clear_perf_probe_event(pev); + + return ret; +} + +void clear_perf_probe_event(struct perf_probe_event *pev) +{ + struct perf_probe_point *pp = &pev->point; + struct perf_probe_arg_field *field, *next; + int i; + + if (pev->event) + free(pev->event); + if (pev->group) + free(pev->group); + if (pp->file) + free(pp->file); + if (pp->function) + free(pp->function); + if (pp->lazy_line) + free(pp->lazy_line); + for (i = 0; i < pev->nargs; i++) { + if (pev->args[i].name) + free(pev->args[i].name); + if (pev->args[i].var) + free(pev->args[i].var); + if (pev->args[i].type) + free(pev->args[i].type); + field = pev->args[i].field; + while (field) { + next = field->next; + if (field->name) + free(field->name); + free(field); + field = next; + } + } + if (pev->args) + free(pev->args); + memset(pev, 0, sizeof(*pev)); +} + +static void clear_probe_trace_event(struct probe_trace_event *tev) +{ + struct probe_trace_arg_ref *ref, *next; + int i; + + if (tev->event) + free(tev->event); + if (tev->group) + free(tev->group); + if (tev->point.symbol) + free(tev->point.symbol); + if (tev->point.module) + free(tev->point.module); + for (i = 0; i < tev->nargs; i++) { + if (tev->args[i].name) + free(tev->args[i].name); + if (tev->args[i].value) + free(tev->args[i].value); + if (tev->args[i].type) + free(tev->args[i].type); + ref = tev->args[i].ref; + while (ref) { + next = ref->next; + free(ref); + ref = next; + } + } + if (tev->args) + free(tev->args); + memset(tev, 0, sizeof(*tev)); +} + +static void print_warn_msg(const char *file, bool is_kprobe) +{ + + if (errno == ENOENT) { + const char *config; + + if (!is_kprobe) + config = "CONFIG_UPROBE_EVENTS"; + else + config = "CONFIG_KPROBE_EVENTS"; + + pr_warning("%s file does not exist - please rebuild kernel" + " with %s.\n", file, config); + } else + pr_warning("Failed to open %s file: %s\n", file, + strerror(errno)); +} + +static int open_probe_events(const char *trace_file, bool readwrite, + bool is_kprobe) +{ + char buf[PATH_MAX]; + const char *__debugfs; + int ret; + + __debugfs = debugfs_find_mountpoint(); + if (__debugfs == NULL) { + pr_warning("Debugfs is not mounted.\n"); + return -ENOENT; + } + + ret = e_snprintf(buf, PATH_MAX, "%s/%s", __debugfs, trace_file); + if (ret >= 0) { + pr_debug("Opening %s write=%d\n", buf, readwrite); + if (readwrite && !probe_event_dry_run) + ret = open(buf, O_RDWR, O_APPEND); + else + ret = open(buf, O_RDONLY, 0); + + if (ret < 0) + print_warn_msg(buf, is_kprobe); + } + return ret; +} + +static int open_kprobe_events(bool readwrite) +{ + return open_probe_events("tracing/kprobe_events", readwrite, true); +} + +static int open_uprobe_events(bool readwrite) +{ + return open_probe_events("tracing/uprobe_events", readwrite, false); +} + +/* Get raw string list of current kprobe_events or uprobe_events */ +static struct strlist *get_probe_trace_command_rawlist(int fd) +{ + int ret, idx; + FILE *fp; + char buf[MAX_CMDLEN]; + char *p; + struct strlist *sl; + + sl = strlist__new(true, NULL); + + fp = fdopen(dup(fd), "r"); + while (!feof(fp)) { + p = fgets(buf, MAX_CMDLEN, fp); + if (!p) + break; + + idx = strlen(p) - 1; + if (p[idx] == '\n') + p[idx] = '\0'; + ret = strlist__add(sl, buf); + if (ret < 0) { + pr_debug("strlist__add failed: %s\n", strerror(-ret)); + strlist__delete(sl); + return NULL; + } + } + fclose(fp); + + return sl; +} + +/* Show an event */ +static int show_perf_probe_event(struct perf_probe_event *pev) +{ + int i, ret; + char buf[128]; + char *place; + + /* Synthesize only event probe point */ + place = synthesize_perf_probe_point(&pev->point); + if (!place) + return -EINVAL; + + ret = e_snprintf(buf, 128, "%s:%s", pev->group, pev->event); + if (ret < 0) + return ret; + + printf(" %-20s (on %s", buf, place); + + if (pev->nargs > 0) { + printf(" with"); + for (i = 0; i < pev->nargs; i++) { + ret = synthesize_perf_probe_arg(&pev->args[i], + buf, 128); + if (ret < 0) + break; + printf(" %s", buf); + } + } + printf(")\n"); + free(place); + return ret; +} + +static int __show_perf_probe_events(int fd, bool is_kprobe) +{ + int ret = 0; + struct probe_trace_event tev; + struct perf_probe_event pev; + struct strlist *rawlist; + struct str_node *ent; + + memset(&tev, 0, sizeof(tev)); + memset(&pev, 0, sizeof(pev)); + + rawlist = get_probe_trace_command_rawlist(fd); + if (!rawlist) + return -ENOENT; + + strlist__for_each(ent, rawlist) { + ret = parse_probe_trace_command(ent->s, &tev); + if (ret >= 0) { + ret = convert_to_perf_probe_event(&tev, &pev, + is_kprobe); + if (ret >= 0) + ret = show_perf_probe_event(&pev); + } + clear_perf_probe_event(&pev); + clear_probe_trace_event(&tev); + if (ret < 0) + break; + } + strlist__delete(rawlist); + + return ret; +} + +/* List up current perf-probe events */ +int show_perf_probe_events(void) +{ + int fd, ret; + + setup_pager(); + fd = open_kprobe_events(false); + + if (fd < 0) + return fd; + + ret = init_vmlinux(); + if (ret < 0) + return ret; + + ret = __show_perf_probe_events(fd, true); + close(fd); + + fd = open_uprobe_events(false); + if (fd >= 0) { + ret = __show_perf_probe_events(fd, false); + close(fd); + } + + return ret; +} + +/* Get current perf-probe event names */ +static struct strlist *get_probe_trace_event_names(int fd, bool include_group) +{ + char buf[128]; + struct strlist *sl, *rawlist; + struct str_node *ent; + struct probe_trace_event tev; + int ret = 0; + + memset(&tev, 0, sizeof(tev)); + rawlist = get_probe_trace_command_rawlist(fd); + sl = strlist__new(true, NULL); + strlist__for_each(ent, rawlist) { + ret = parse_probe_trace_command(ent->s, &tev); + if (ret < 0) + break; + if (include_group) { + ret = e_snprintf(buf, 128, "%s:%s", tev.group, + tev.event); + if (ret >= 0) + ret = strlist__add(sl, buf); + } else + ret = strlist__add(sl, tev.event); + clear_probe_trace_event(&tev); + if (ret < 0) + break; + } + strlist__delete(rawlist); + + if (ret < 0) { + strlist__delete(sl); + return NULL; + } + return sl; +} + +static int write_probe_trace_event(int fd, struct probe_trace_event *tev) +{ + int ret = 0; + char *buf = synthesize_probe_trace_command(tev); + + if (!buf) { + pr_debug("Failed to synthesize probe trace event.\n"); + return -EINVAL; + } + + pr_debug("Writing event: %s\n", buf); + if (!probe_event_dry_run) { + ret = write(fd, buf, strlen(buf)); + if (ret <= 0) + pr_warning("Failed to write event: %s\n", + strerror(errno)); + } + free(buf); + return ret; +} + +static int get_new_event_name(char *buf, size_t len, const char *base, + struct strlist *namelist, bool allow_suffix) +{ + int i, ret; + + /* Try no suffix */ + ret = e_snprintf(buf, len, "%s", base); + if (ret < 0) { + pr_debug("snprintf() failed: %s\n", strerror(-ret)); + return ret; + } + if (!strlist__has_entry(namelist, buf)) + return 0; + + if (!allow_suffix) { + pr_warning("Error: event \"%s\" already exists. " + "(Use -f to force duplicates.)\n", base); + return -EEXIST; + } + + /* Try to add suffix */ + for (i = 1; i < MAX_EVENT_INDEX; i++) { + ret = e_snprintf(buf, len, "%s_%d", base, i); + if (ret < 0) { + pr_debug("snprintf() failed: %s\n", strerror(-ret)); + return ret; + } + if (!strlist__has_entry(namelist, buf)) + break; + } + if (i == MAX_EVENT_INDEX) { + pr_warning("Too many events are on the same function.\n"); + ret = -ERANGE; + } + + return ret; +} + +static int __add_probe_trace_events(struct perf_probe_event *pev, + struct probe_trace_event *tevs, + int ntevs, bool allow_suffix) +{ + int i, fd, ret; + struct probe_trace_event *tev = NULL; + char buf[64]; + const char *event, *group; + struct strlist *namelist; + + if (pev->uprobes) + fd = open_uprobe_events(true); + else + fd = open_kprobe_events(true); + + if (fd < 0) + return fd; + /* Get current event names */ + namelist = get_probe_trace_event_names(fd, false); + if (!namelist) { + pr_debug("Failed to get current event list.\n"); + return -EIO; + } + + ret = 0; + printf("Added new event%s\n", (ntevs > 1) ? "s:" : ":"); + for (i = 0; i < ntevs; i++) { + tev = &tevs[i]; + if (pev->event) + event = pev->event; + else + if (pev->point.function) + event = pev->point.function; + else + event = tev->point.symbol; + if (pev->group) + group = pev->group; + else + group = PERFPROBE_GROUP; + + /* Get an unused new event name */ + ret = get_new_event_name(buf, 64, event, + namelist, allow_suffix); + if (ret < 0) + break; + event = buf; + + tev->event = strdup(event); + tev->group = strdup(group); + if (tev->event == NULL || tev->group == NULL) { + ret = -ENOMEM; + break; + } + ret = write_probe_trace_event(fd, tev); + if (ret < 0) + break; + /* Add added event name to namelist */ + strlist__add(namelist, event); + + /* Trick here - save current event/group */ + event = pev->event; + group = pev->group; + pev->event = tev->event; + pev->group = tev->group; + show_perf_probe_event(pev); + /* Trick here - restore current event/group */ + pev->event = (char *)event; + pev->group = (char *)group; + + /* + * Probes after the first probe which comes from same + * user input are always allowed to add suffix, because + * there might be several addresses corresponding to + * one code line. + */ + allow_suffix = true; + } + + if (ret >= 0) { + /* Show how to use the event. */ + printf("\nYou can now use it in all perf tools, such as:\n\n"); + printf("\tperf record -e %s:%s -aR sleep 1\n\n", tev->group, + tev->event); + } + + strlist__delete(namelist); + close(fd); + return ret; +} + +static int convert_to_probe_trace_events(struct perf_probe_event *pev, + struct probe_trace_event **tevs, + int max_tevs, const char *target) +{ + struct symbol *sym; + int ret = 0, i; + struct probe_trace_event *tev; + + /* Convert perf_probe_event with debuginfo */ + ret = try_to_find_probe_trace_events(pev, tevs, max_tevs, target); + if (ret != 0) + return ret; /* Found in debuginfo or got an error */ + + /* Allocate trace event buffer */ + tev = *tevs = zalloc(sizeof(struct probe_trace_event)); + if (tev == NULL) + return -ENOMEM; + + /* Copy parameters */ + tev->point.symbol = strdup(pev->point.function); + if (tev->point.symbol == NULL) { + ret = -ENOMEM; + goto error; + } + + if (target) { + tev->point.module = strdup(target); + if (tev->point.module == NULL) { + ret = -ENOMEM; + goto error; + } + } + + tev->point.offset = pev->point.offset; + tev->point.retprobe = pev->point.retprobe; + tev->nargs = pev->nargs; + tev->uprobes = pev->uprobes; + + if (tev->nargs) { + tev->args = zalloc(sizeof(struct probe_trace_arg) + * tev->nargs); + if (tev->args == NULL) { + ret = -ENOMEM; + goto error; + } + for (i = 0; i < tev->nargs; i++) { + if (pev->args[i].name) { + tev->args[i].name = strdup(pev->args[i].name); + if (tev->args[i].name == NULL) { + ret = -ENOMEM; + goto error; + } + } + tev->args[i].value = strdup(pev->args[i].var); + if (tev->args[i].value == NULL) { + ret = -ENOMEM; + goto error; + } + if (pev->args[i].type) { + tev->args[i].type = strdup(pev->args[i].type); + if (tev->args[i].type == NULL) { + ret = -ENOMEM; + goto error; + } + } + } + } + + if (pev->uprobes) + return 1; + + /* Currently just checking function name from symbol map */ + sym = __find_kernel_function_by_name(tev->point.symbol, NULL); + if (!sym) { + pr_warning("Kernel symbol \'%s\' not found.\n", + tev->point.symbol); + ret = -ENOENT; + goto error; + } else if (tev->point.offset > sym->end - sym->start) { + pr_warning("Offset specified is greater than size of %s\n", + tev->point.symbol); + ret = -ENOENT; + goto error; + + } + + return 1; +error: + clear_probe_trace_event(tev); + free(tev); + *tevs = NULL; + return ret; +} + +struct __event_package { + struct perf_probe_event *pev; + struct probe_trace_event *tevs; + int ntevs; +}; + +int add_perf_probe_events(struct perf_probe_event *pevs, int npevs, + int max_tevs, const char *target, bool force_add) +{ + int i, j, ret; + struct __event_package *pkgs; + + ret = 0; + pkgs = zalloc(sizeof(struct __event_package) * npevs); + + if (pkgs == NULL) + return -ENOMEM; + + if (!pevs->uprobes) + /* Init vmlinux path */ + ret = init_vmlinux(); + else + ret = init_user_exec(); + + if (ret < 0) { + free(pkgs); + return ret; + } + + /* Loop 1: convert all events */ + for (i = 0; i < npevs; i++) { + pkgs[i].pev = &pevs[i]; + /* Convert with or without debuginfo */ + ret = convert_to_probe_trace_events(pkgs[i].pev, + &pkgs[i].tevs, + max_tevs, + target); + if (ret < 0) + goto end; + pkgs[i].ntevs = ret; + } + + /* Loop 2: add all events */ + for (i = 0; i < npevs; i++) { + ret = __add_probe_trace_events(pkgs[i].pev, pkgs[i].tevs, + pkgs[i].ntevs, force_add); + if (ret < 0) + break; + } +end: + /* Loop 3: cleanup and free trace events */ + for (i = 0; i < npevs; i++) { + for (j = 0; j < pkgs[i].ntevs; j++) + clear_probe_trace_event(&pkgs[i].tevs[j]); + free(pkgs[i].tevs); + } + free(pkgs); + + return ret; +} + +static int __del_trace_probe_event(int fd, struct str_node *ent) +{ + char *p; + char buf[128]; + int ret; + + /* Convert from perf-probe event to trace-probe event */ + ret = e_snprintf(buf, 128, "-:%s", ent->s); + if (ret < 0) + goto error; + + p = strchr(buf + 2, ':'); + if (!p) { + pr_debug("Internal error: %s should have ':' but not.\n", + ent->s); + ret = -ENOTSUP; + goto error; + } + *p = '/'; + + pr_debug("Writing event: %s\n", buf); + ret = write(fd, buf, strlen(buf)); + if (ret < 0) { + ret = -errno; + goto error; + } + + printf("Removed event: %s\n", ent->s); + return 0; +error: + pr_warning("Failed to delete event: %s\n", strerror(-ret)); + return ret; +} + +static int del_trace_probe_event(int fd, const char *buf, + struct strlist *namelist) +{ + struct str_node *ent, *n; + int ret = -1; + + if (strpbrk(buf, "*?")) { /* Glob-exp */ + strlist__for_each_safe(ent, n, namelist) + if (strglobmatch(ent->s, buf)) { + ret = __del_trace_probe_event(fd, ent); + if (ret < 0) + break; + strlist__remove(namelist, ent); + } + } else { + ent = strlist__find(namelist, buf); + if (ent) { + ret = __del_trace_probe_event(fd, ent); + if (ret >= 0) + strlist__remove(namelist, ent); + } + } + + return ret; +} + +int del_perf_probe_events(struct strlist *dellist) +{ + int ret = -1, ufd = -1, kfd = -1; + char buf[128]; + const char *group, *event; + char *p, *str; + struct str_node *ent; + struct strlist *namelist = NULL, *unamelist = NULL; + + /* Get current event names */ + kfd = open_kprobe_events(true); + if (kfd < 0) + return kfd; + + namelist = get_probe_trace_event_names(kfd, true); + ufd = open_uprobe_events(true); + + if (ufd >= 0) + unamelist = get_probe_trace_event_names(ufd, true); + + if (namelist == NULL && unamelist == NULL) + goto error; + + strlist__for_each(ent, dellist) { + str = strdup(ent->s); + if (str == NULL) { + ret = -ENOMEM; + goto error; + } + pr_debug("Parsing: %s\n", str); + p = strchr(str, ':'); + if (p) { + group = str; + *p = '\0'; + event = p + 1; + } else { + group = "*"; + event = str; + } + + ret = e_snprintf(buf, 128, "%s:%s", group, event); + if (ret < 0) { + pr_err("Failed to copy event."); + free(str); + goto error; + } + + pr_debug("Group: %s, Event: %s\n", group, event); + + if (namelist) + ret = del_trace_probe_event(kfd, buf, namelist); + + if (unamelist && ret != 0) + ret = del_trace_probe_event(ufd, buf, unamelist); + + if (ret != 0) + pr_info("Info: Event \"%s\" does not exist.\n", buf); + + free(str); + } + +error: + if (kfd >= 0) { + strlist__delete(namelist); + close(kfd); + } + + if (ufd >= 0) { + strlist__delete(unamelist); + close(ufd); + } + + return ret; +} + +/* TODO: don't use a global variable for filter ... */ +static struct strfilter *available_func_filter; + +/* + * If a symbol corresponds to a function with global binding and + * matches filter return 0. For all others return 1. + */ +static int filter_available_functions(struct map *map __maybe_unused, + struct symbol *sym) +{ + if (sym->binding == STB_GLOBAL && + strfilter__compare(available_func_filter, sym->name)) + return 0; + return 1; +} + +static int __show_available_funcs(struct map *map) +{ + if (map__load(map, filter_available_functions)) { + pr_err("Failed to load map.\n"); + return -EINVAL; + } + if (!dso__sorted_by_name(map->dso, map->type)) + dso__sort_by_name(map->dso, map->type); + + dso__fprintf_symbols_by_name(map->dso, map->type, stdout); + return 0; +} + +static int available_kernel_funcs(const char *module) +{ + struct map *map; + int ret; + + ret = init_vmlinux(); + if (ret < 0) + return ret; + + map = kernel_get_module_map(module); + if (!map) { + pr_err("Failed to find %s map.\n", (module) ? : "kernel"); + return -EINVAL; + } + return __show_available_funcs(map); +} + +static int available_user_funcs(const char *target) +{ + struct map *map; + int ret; + + ret = init_user_exec(); + if (ret < 0) + return ret; + + map = dso__new_map(target); + ret = __show_available_funcs(map); + dso__delete(map->dso); + map__delete(map); + return ret; +} + +int show_available_funcs(const char *target, struct strfilter *_filter, + bool user) +{ + setup_pager(); + available_func_filter = _filter; + + if (!user) + return available_kernel_funcs(target); + + return available_user_funcs(target); +} + +/* + * uprobe_events only accepts address: + * Convert function and any offset to address + */ +static int convert_name_to_addr(struct perf_probe_event *pev, const char *exec) +{ + struct perf_probe_point *pp = &pev->point; + struct symbol *sym; + struct map *map = NULL; + char *function = NULL, *name = NULL; + int ret = -EINVAL; + unsigned long long vaddr = 0; + + if (!pp->function) { + pr_warning("No function specified for uprobes"); + goto out; + } + + function = strdup(pp->function); + if (!function) { + pr_warning("Failed to allocate memory by strdup.\n"); + ret = -ENOMEM; + goto out; + } + + name = realpath(exec, NULL); + if (!name) { + pr_warning("Cannot find realpath for %s.\n", exec); + goto out; + } + map = dso__new_map(name); + if (!map) { + pr_warning("Cannot find appropriate DSO for %s.\n", exec); + goto out; + } + available_func_filter = strfilter__new(function, NULL); + if (map__load(map, filter_available_functions)) { + pr_err("Failed to load map.\n"); + goto out; + } + + sym = map__find_symbol_by_name(map, function, NULL); + if (!sym) { + pr_warning("Cannot find %s in DSO %s\n", function, exec); + goto out; + } + + if (map->start > sym->start) + vaddr = map->start; + vaddr += sym->start + pp->offset + map->pgoff; + pp->offset = 0; + + if (!pev->event) { + pev->event = function; + function = NULL; + } + if (!pev->group) { + char *ptr1, *ptr2, *exec_copy; + + pev->group = zalloc(sizeof(char *) * 64); + exec_copy = strdup(exec); + if (!exec_copy) { + ret = -ENOMEM; + pr_warning("Failed to copy exec string.\n"); + goto out; + } + + ptr1 = strdup(basename(exec_copy)); + if (ptr1) { + ptr2 = strpbrk(ptr1, "-._"); + if (ptr2) + *ptr2 = '\0'; + e_snprintf(pev->group, 64, "%s_%s", PERFPROBE_GROUP, + ptr1); + free(ptr1); + } + free(exec_copy); + } + free(pp->function); + pp->function = zalloc(sizeof(char *) * MAX_PROBE_ARGS); + if (!pp->function) { + ret = -ENOMEM; + pr_warning("Failed to allocate memory by zalloc.\n"); + goto out; + } + e_snprintf(pp->function, MAX_PROBE_ARGS, "0x%llx", vaddr); + ret = 0; + +out: + if (map) { + dso__delete(map->dso); + map__delete(map); + } + if (function) + free(function); + if (name) + free(name); + return ret; +} diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h new file mode 100644 index 00000000..f9f3de8b --- /dev/null +++ b/tools/perf/util/probe-event.h @@ -0,0 +1,140 @@ +#ifndef _PROBE_EVENT_H +#define _PROBE_EVENT_H + +#include <stdbool.h> +#include "strlist.h" +#include "strfilter.h" + +extern bool probe_event_dry_run; + +/* kprobe-tracer and uprobe-tracer tracing point */ +struct probe_trace_point { + char *symbol; /* Base symbol */ + char *module; /* Module name */ + unsigned long offset; /* Offset from symbol */ + bool retprobe; /* Return probe flag */ +}; + +/* probe-tracer tracing argument referencing offset */ +struct probe_trace_arg_ref { + struct probe_trace_arg_ref *next; /* Next reference */ + long offset; /* Offset value */ +}; + +/* kprobe-tracer and uprobe-tracer tracing argument */ +struct probe_trace_arg { + char *name; /* Argument name */ + char *value; /* Base value */ + char *type; /* Type name */ + struct probe_trace_arg_ref *ref; /* Referencing offset */ +}; + +/* kprobe-tracer and uprobe-tracer tracing event (point + arg) */ +struct probe_trace_event { + char *event; /* Event name */ + char *group; /* Group name */ + struct probe_trace_point point; /* Trace point */ + int nargs; /* Number of args */ + bool uprobes; /* uprobes only */ + struct probe_trace_arg *args; /* Arguments */ +}; + +/* Perf probe probing point */ +struct perf_probe_point { + char *file; /* File path */ + char *function; /* Function name */ + int line; /* Line number */ + bool retprobe; /* Return probe flag */ + char *lazy_line; /* Lazy matching pattern */ + unsigned long offset; /* Offset from function entry */ +}; + +/* Perf probe probing argument field chain */ +struct perf_probe_arg_field { + struct perf_probe_arg_field *next; /* Next field */ + char *name; /* Name of the field */ + long index; /* Array index number */ + bool ref; /* Referencing flag */ +}; + +/* Perf probe probing argument */ +struct perf_probe_arg { + char *name; /* Argument name */ + char *var; /* Variable name */ + char *type; /* Type name */ + struct perf_probe_arg_field *field; /* Structure fields */ +}; + +/* Perf probe probing event (point + arg) */ +struct perf_probe_event { + char *event; /* Event name */ + char *group; /* Group name */ + struct perf_probe_point point; /* Probe point */ + int nargs; /* Number of arguments */ + bool uprobes; + struct perf_probe_arg *args; /* Arguments */ +}; + + +/* Line number container */ +struct line_node { + struct list_head list; + int line; +}; + +/* Line range */ +struct line_range { + char *file; /* File name */ + char *function; /* Function name */ + int start; /* Start line number */ + int end; /* End line number */ + int offset; /* Start line offset */ + char *path; /* Real path name */ + char *comp_dir; /* Compile directory */ + struct list_head line_list; /* Visible lines */ +}; + +/* List of variables */ +struct variable_list { + struct probe_trace_point point; /* Actual probepoint */ + struct strlist *vars; /* Available variables */ +}; + +/* Command string to events */ +extern int parse_perf_probe_command(const char *cmd, + struct perf_probe_event *pev); + +/* Events to command string */ +extern char *synthesize_perf_probe_command(struct perf_probe_event *pev); +extern char *synthesize_probe_trace_command(struct probe_trace_event *tev); +extern int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf, + size_t len); + +/* Check the perf_probe_event needs debuginfo */ +extern bool perf_probe_event_need_dwarf(struct perf_probe_event *pev); + +/* Release event contents */ +extern void clear_perf_probe_event(struct perf_probe_event *pev); + +/* Command string to line-range */ +extern int parse_line_range_desc(const char *cmd, struct line_range *lr); + +/* Internal use: Return kernel/module path */ +extern const char *kernel_get_module_path(const char *module); + +extern int add_perf_probe_events(struct perf_probe_event *pevs, int npevs, + int max_probe_points, const char *module, + bool force_add); +extern int del_perf_probe_events(struct strlist *dellist); +extern int show_perf_probe_events(void); +extern int show_line_range(struct line_range *lr, const char *module); +extern int show_available_vars(struct perf_probe_event *pevs, int npevs, + int max_probe_points, const char *module, + struct strfilter *filter, bool externs); +extern int show_available_funcs(const char *module, struct strfilter *filter, + bool user); + +/* Maximum index number of event-name postfix */ +#define MAX_EVENT_INDEX 1024 + +#endif /*_PROBE_EVENT_H */ diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c new file mode 100644 index 00000000..be032939 --- /dev/null +++ b/tools/perf/util/probe-finder.c @@ -0,0 +1,1588 @@ +/* + * probe-finder.c : C expression to kprobe event converter + * + * Written by Masami Hiramatsu <mhiramat@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + */ + +#include <sys/utsname.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <errno.h> +#include <stdio.h> +#include <unistd.h> +#include <getopt.h> +#include <stdlib.h> +#include <string.h> +#include <stdarg.h> +#include <dwarf-regs.h> + +#include <linux/bitops.h> +#include "event.h" +#include "debug.h" +#include "util.h" +#include "symbol.h" +#include "probe-finder.h" + +/* Kprobe tracer basic type is up to u64 */ +#define MAX_BASIC_TYPE_BITS 64 + +/* Line number list operations */ + +/* Add a line to line number list */ +static int line_list__add_line(struct list_head *head, int line) +{ + struct line_node *ln; + struct list_head *p; + + /* Reverse search, because new line will be the last one */ + list_for_each_entry_reverse(ln, head, list) { + if (ln->line < line) { + p = &ln->list; + goto found; + } else if (ln->line == line) /* Already exist */ + return 1; + } + /* List is empty, or the smallest entry */ + p = head; +found: + pr_debug("line list: add a line %u\n", line); + ln = zalloc(sizeof(struct line_node)); + if (ln == NULL) + return -ENOMEM; + ln->line = line; + INIT_LIST_HEAD(&ln->list); + list_add(&ln->list, p); + return 0; +} + +/* Check if the line in line number list */ +static int line_list__has_line(struct list_head *head, int line) +{ + struct line_node *ln; + + /* Reverse search, because new line will be the last one */ + list_for_each_entry(ln, head, list) + if (ln->line == line) + return 1; + + return 0; +} + +/* Init line number list */ +static void line_list__init(struct list_head *head) +{ + INIT_LIST_HEAD(head); +} + +/* Free line number list */ +static void line_list__free(struct list_head *head) +{ + struct line_node *ln; + while (!list_empty(head)) { + ln = list_first_entry(head, struct line_node, list); + list_del(&ln->list); + free(ln); + } +} + +/* Dwarf FL wrappers */ +static char *debuginfo_path; /* Currently dummy */ + +static const Dwfl_Callbacks offline_callbacks = { + .find_debuginfo = dwfl_standard_find_debuginfo, + .debuginfo_path = &debuginfo_path, + + .section_address = dwfl_offline_section_address, + + /* We use this table for core files too. */ + .find_elf = dwfl_build_id_find_elf, +}; + +/* Get a Dwarf from offline image */ +static int debuginfo__init_offline_dwarf(struct debuginfo *self, + const char *path) +{ + Dwfl_Module *mod; + int fd; + + fd = open(path, O_RDONLY); + if (fd < 0) + return fd; + + self->dwfl = dwfl_begin(&offline_callbacks); + if (!self->dwfl) + goto error; + + mod = dwfl_report_offline(self->dwfl, "", "", fd); + if (!mod) + goto error; + + self->dbg = dwfl_module_getdwarf(mod, &self->bias); + if (!self->dbg) + goto error; + + return 0; +error: + if (self->dwfl) + dwfl_end(self->dwfl); + else + close(fd); + memset(self, 0, sizeof(*self)); + + return -ENOENT; +} + +#if _ELFUTILS_PREREQ(0, 148) +/* This method is buggy if elfutils is older than 0.148 */ +static int __linux_kernel_find_elf(Dwfl_Module *mod, + void **userdata, + const char *module_name, + Dwarf_Addr base, + char **file_name, Elf **elfp) +{ + int fd; + const char *path = kernel_get_module_path(module_name); + + pr_debug2("Use file %s for %s\n", path, module_name); + if (path) { + fd = open(path, O_RDONLY); + if (fd >= 0) { + *file_name = strdup(path); + return fd; + } + } + /* If failed, try to call standard method */ + return dwfl_linux_kernel_find_elf(mod, userdata, module_name, base, + file_name, elfp); +} + +static const Dwfl_Callbacks kernel_callbacks = { + .find_debuginfo = dwfl_standard_find_debuginfo, + .debuginfo_path = &debuginfo_path, + + .find_elf = __linux_kernel_find_elf, + .section_address = dwfl_linux_kernel_module_section_address, +}; + +/* Get a Dwarf from live kernel image */ +static int debuginfo__init_online_kernel_dwarf(struct debuginfo *self, + Dwarf_Addr addr) +{ + self->dwfl = dwfl_begin(&kernel_callbacks); + if (!self->dwfl) + return -EINVAL; + + /* Load the kernel dwarves: Don't care the result here */ + dwfl_linux_kernel_report_kernel(self->dwfl); + dwfl_linux_kernel_report_modules(self->dwfl); + + self->dbg = dwfl_addrdwarf(self->dwfl, addr, &self->bias); + /* Here, check whether we could get a real dwarf */ + if (!self->dbg) { + pr_debug("Failed to find kernel dwarf at %lx\n", + (unsigned long)addr); + dwfl_end(self->dwfl); + memset(self, 0, sizeof(*self)); + return -ENOENT; + } + + return 0; +} +#else +/* With older elfutils, this just support kernel module... */ +static int debuginfo__init_online_kernel_dwarf(struct debuginfo *self, + Dwarf_Addr addr __maybe_unused) +{ + const char *path = kernel_get_module_path("kernel"); + + if (!path) { + pr_err("Failed to find vmlinux path\n"); + return -ENOENT; + } + + pr_debug2("Use file %s for debuginfo\n", path); + return debuginfo__init_offline_dwarf(self, path); +} +#endif + +struct debuginfo *debuginfo__new(const char *path) +{ + struct debuginfo *self = zalloc(sizeof(struct debuginfo)); + if (!self) + return NULL; + + if (debuginfo__init_offline_dwarf(self, path) < 0) { + free(self); + self = NULL; + } + + return self; +} + +struct debuginfo *debuginfo__new_online_kernel(unsigned long addr) +{ + struct debuginfo *self = zalloc(sizeof(struct debuginfo)); + if (!self) + return NULL; + + if (debuginfo__init_online_kernel_dwarf(self, (Dwarf_Addr)addr) < 0) { + free(self); + self = NULL; + } + + return self; +} + +void debuginfo__delete(struct debuginfo *self) +{ + if (self) { + if (self->dwfl) + dwfl_end(self->dwfl); + free(self); + } +} + +/* + * Probe finder related functions + */ + +static struct probe_trace_arg_ref *alloc_trace_arg_ref(long offs) +{ + struct probe_trace_arg_ref *ref; + ref = zalloc(sizeof(struct probe_trace_arg_ref)); + if (ref != NULL) + ref->offset = offs; + return ref; +} + +/* + * Convert a location into trace_arg. + * If tvar == NULL, this just checks variable can be converted. + */ +static int convert_variable_location(Dwarf_Die *vr_die, Dwarf_Addr addr, + Dwarf_Op *fb_ops, + struct probe_trace_arg *tvar) +{ + Dwarf_Attribute attr; + Dwarf_Op *op; + size_t nops; + unsigned int regn; + Dwarf_Word offs = 0; + bool ref = false; + const char *regs; + int ret; + + if (dwarf_attr(vr_die, DW_AT_external, &attr) != NULL) + goto static_var; + + /* TODO: handle more than 1 exprs */ + if (dwarf_attr(vr_die, DW_AT_location, &attr) == NULL || + dwarf_getlocation_addr(&attr, addr, &op, &nops, 1) <= 0 || + nops == 0) { + /* TODO: Support const_value */ + return -ENOENT; + } + + if (op->atom == DW_OP_addr) { +static_var: + if (!tvar) + return 0; + /* Static variables on memory (not stack), make @varname */ + ret = strlen(dwarf_diename(vr_die)); + tvar->value = zalloc(ret + 2); + if (tvar->value == NULL) + return -ENOMEM; + snprintf(tvar->value, ret + 2, "@%s", dwarf_diename(vr_die)); + tvar->ref = alloc_trace_arg_ref((long)offs); + if (tvar->ref == NULL) + return -ENOMEM; + return 0; + } + + /* If this is based on frame buffer, set the offset */ + if (op->atom == DW_OP_fbreg) { + if (fb_ops == NULL) + return -ENOTSUP; + ref = true; + offs = op->number; + op = &fb_ops[0]; + } + + if (op->atom >= DW_OP_breg0 && op->atom <= DW_OP_breg31) { + regn = op->atom - DW_OP_breg0; + offs += op->number; + ref = true; + } else if (op->atom >= DW_OP_reg0 && op->atom <= DW_OP_reg31) { + regn = op->atom - DW_OP_reg0; + } else if (op->atom == DW_OP_bregx) { + regn = op->number; + offs += op->number2; + ref = true; + } else if (op->atom == DW_OP_regx) { + regn = op->number; + } else { + pr_debug("DW_OP %x is not supported.\n", op->atom); + return -ENOTSUP; + } + + if (!tvar) + return 0; + + regs = get_arch_regstr(regn); + if (!regs) { + /* This should be a bug in DWARF or this tool */ + pr_warning("Mapping for the register number %u " + "missing on this architecture.\n", regn); + return -ERANGE; + } + + tvar->value = strdup(regs); + if (tvar->value == NULL) + return -ENOMEM; + + if (ref) { + tvar->ref = alloc_trace_arg_ref((long)offs); + if (tvar->ref == NULL) + return -ENOMEM; + } + return 0; +} + +#define BYTES_TO_BITS(nb) ((nb) * BITS_PER_LONG / sizeof(long)) + +static int convert_variable_type(Dwarf_Die *vr_die, + struct probe_trace_arg *tvar, + const char *cast) +{ + struct probe_trace_arg_ref **ref_ptr = &tvar->ref; + Dwarf_Die type; + char buf[16]; + int bsize, boffs, total; + int ret; + + /* TODO: check all types */ + if (cast && strcmp(cast, "string") != 0) { + /* Non string type is OK */ + tvar->type = strdup(cast); + return (tvar->type == NULL) ? -ENOMEM : 0; + } + + bsize = dwarf_bitsize(vr_die); + if (bsize > 0) { + /* This is a bitfield */ + boffs = dwarf_bitoffset(vr_die); + total = dwarf_bytesize(vr_die); + if (boffs < 0 || total < 0) + return -ENOENT; + ret = snprintf(buf, 16, "b%d@%d/%zd", bsize, boffs, + BYTES_TO_BITS(total)); + goto formatted; + } + + if (die_get_real_type(vr_die, &type) == NULL) { + pr_warning("Failed to get a type information of %s.\n", + dwarf_diename(vr_die)); + return -ENOENT; + } + + pr_debug("%s type is %s.\n", + dwarf_diename(vr_die), dwarf_diename(&type)); + + if (cast && strcmp(cast, "string") == 0) { /* String type */ + ret = dwarf_tag(&type); + if (ret != DW_TAG_pointer_type && + ret != DW_TAG_array_type) { + pr_warning("Failed to cast into string: " + "%s(%s) is not a pointer nor array.\n", + dwarf_diename(vr_die), dwarf_diename(&type)); + return -EINVAL; + } + if (die_get_real_type(&type, &type) == NULL) { + pr_warning("Failed to get a type" + " information.\n"); + return -ENOENT; + } + if (ret == DW_TAG_pointer_type) { + while (*ref_ptr) + ref_ptr = &(*ref_ptr)->next; + /* Add new reference with offset +0 */ + *ref_ptr = zalloc(sizeof(struct probe_trace_arg_ref)); + if (*ref_ptr == NULL) { + pr_warning("Out of memory error\n"); + return -ENOMEM; + } + } + if (!die_compare_name(&type, "char") && + !die_compare_name(&type, "unsigned char")) { + pr_warning("Failed to cast into string: " + "%s is not (unsigned) char *.\n", + dwarf_diename(vr_die)); + return -EINVAL; + } + tvar->type = strdup(cast); + return (tvar->type == NULL) ? -ENOMEM : 0; + } + + ret = dwarf_bytesize(&type); + if (ret <= 0) + /* No size ... try to use default type */ + return 0; + ret = BYTES_TO_BITS(ret); + + /* Check the bitwidth */ + if (ret > MAX_BASIC_TYPE_BITS) { + pr_info("%s exceeds max-bitwidth. Cut down to %d bits.\n", + dwarf_diename(&type), MAX_BASIC_TYPE_BITS); + ret = MAX_BASIC_TYPE_BITS; + } + ret = snprintf(buf, 16, "%c%d", + die_is_signed_type(&type) ? 's' : 'u', ret); + +formatted: + if (ret < 0 || ret >= 16) { + if (ret >= 16) + ret = -E2BIG; + pr_warning("Failed to convert variable type: %s\n", + strerror(-ret)); + return ret; + } + tvar->type = strdup(buf); + if (tvar->type == NULL) + return -ENOMEM; + return 0; +} + +static int convert_variable_fields(Dwarf_Die *vr_die, const char *varname, + struct perf_probe_arg_field *field, + struct probe_trace_arg_ref **ref_ptr, + Dwarf_Die *die_mem) +{ + struct probe_trace_arg_ref *ref = *ref_ptr; + Dwarf_Die type; + Dwarf_Word offs; + int ret, tag; + + pr_debug("converting %s in %s\n", field->name, varname); + if (die_get_real_type(vr_die, &type) == NULL) { + pr_warning("Failed to get the type of %s.\n", varname); + return -ENOENT; + } + pr_debug2("Var real type: (%x)\n", (unsigned)dwarf_dieoffset(&type)); + tag = dwarf_tag(&type); + + if (field->name[0] == '[' && + (tag == DW_TAG_array_type || tag == DW_TAG_pointer_type)) { + if (field->next) + /* Save original type for next field */ + memcpy(die_mem, &type, sizeof(*die_mem)); + /* Get the type of this array */ + if (die_get_real_type(&type, &type) == NULL) { + pr_warning("Failed to get the type of %s.\n", varname); + return -ENOENT; + } + pr_debug2("Array real type: (%x)\n", + (unsigned)dwarf_dieoffset(&type)); + if (tag == DW_TAG_pointer_type) { + ref = zalloc(sizeof(struct probe_trace_arg_ref)); + if (ref == NULL) + return -ENOMEM; + if (*ref_ptr) + (*ref_ptr)->next = ref; + else + *ref_ptr = ref; + } + ref->offset += dwarf_bytesize(&type) * field->index; + if (!field->next) + /* Save vr_die for converting types */ + memcpy(die_mem, vr_die, sizeof(*die_mem)); + goto next; + } else if (tag == DW_TAG_pointer_type) { + /* Check the pointer and dereference */ + if (!field->ref) { + pr_err("Semantic error: %s must be referred by '->'\n", + field->name); + return -EINVAL; + } + /* Get the type pointed by this pointer */ + if (die_get_real_type(&type, &type) == NULL) { + pr_warning("Failed to get the type of %s.\n", varname); + return -ENOENT; + } + /* Verify it is a data structure */ + tag = dwarf_tag(&type); + if (tag != DW_TAG_structure_type && tag != DW_TAG_union_type) { + pr_warning("%s is not a data structure nor an union.\n", + varname); + return -EINVAL; + } + + ref = zalloc(sizeof(struct probe_trace_arg_ref)); + if (ref == NULL) + return -ENOMEM; + if (*ref_ptr) + (*ref_ptr)->next = ref; + else + *ref_ptr = ref; + } else { + /* Verify it is a data structure */ + if (tag != DW_TAG_structure_type && tag != DW_TAG_union_type) { + pr_warning("%s is not a data structure nor an union.\n", + varname); + return -EINVAL; + } + if (field->name[0] == '[') { + pr_err("Semantic error: %s is not a pointor" + " nor array.\n", varname); + return -EINVAL; + } + if (field->ref) { + pr_err("Semantic error: %s must be referred by '.'\n", + field->name); + return -EINVAL; + } + if (!ref) { + pr_warning("Structure on a register is not " + "supported yet.\n"); + return -ENOTSUP; + } + } + + if (die_find_member(&type, field->name, die_mem) == NULL) { + pr_warning("%s(tyep:%s) has no member %s.\n", varname, + dwarf_diename(&type), field->name); + return -EINVAL; + } + + /* Get the offset of the field */ + if (tag == DW_TAG_union_type) { + offs = 0; + } else { + ret = die_get_data_member_location(die_mem, &offs); + if (ret < 0) { + pr_warning("Failed to get the offset of %s.\n", + field->name); + return ret; + } + } + ref->offset += (long)offs; + +next: + /* Converting next field */ + if (field->next) + return convert_variable_fields(die_mem, field->name, + field->next, &ref, die_mem); + else + return 0; +} + +/* Show a variables in kprobe event format */ +static int convert_variable(Dwarf_Die *vr_die, struct probe_finder *pf) +{ + Dwarf_Die die_mem; + int ret; + + pr_debug("Converting variable %s into trace event.\n", + dwarf_diename(vr_die)); + + ret = convert_variable_location(vr_die, pf->addr, pf->fb_ops, + pf->tvar); + if (ret == -ENOENT) + pr_err("Failed to find the location of %s at this address.\n" + " Perhaps, it has been optimized out.\n", pf->pvar->var); + else if (ret == -ENOTSUP) + pr_err("Sorry, we don't support this variable location yet.\n"); + else if (pf->pvar->field) { + ret = convert_variable_fields(vr_die, pf->pvar->var, + pf->pvar->field, &pf->tvar->ref, + &die_mem); + vr_die = &die_mem; + } + if (ret == 0) + ret = convert_variable_type(vr_die, pf->tvar, pf->pvar->type); + /* *expr will be cached in libdw. Don't free it. */ + return ret; +} + +/* Find a variable in a scope DIE */ +static int find_variable(Dwarf_Die *sc_die, struct probe_finder *pf) +{ + Dwarf_Die vr_die; + char buf[32], *ptr; + int ret = 0; + + if (!is_c_varname(pf->pvar->var)) { + /* Copy raw parameters */ + pf->tvar->value = strdup(pf->pvar->var); + if (pf->tvar->value == NULL) + return -ENOMEM; + if (pf->pvar->type) { + pf->tvar->type = strdup(pf->pvar->type); + if (pf->tvar->type == NULL) + return -ENOMEM; + } + if (pf->pvar->name) { + pf->tvar->name = strdup(pf->pvar->name); + if (pf->tvar->name == NULL) + return -ENOMEM; + } else + pf->tvar->name = NULL; + return 0; + } + + if (pf->pvar->name) + pf->tvar->name = strdup(pf->pvar->name); + else { + ret = synthesize_perf_probe_arg(pf->pvar, buf, 32); + if (ret < 0) + return ret; + ptr = strchr(buf, ':'); /* Change type separator to _ */ + if (ptr) + *ptr = '_'; + pf->tvar->name = strdup(buf); + } + if (pf->tvar->name == NULL) + return -ENOMEM; + + pr_debug("Searching '%s' variable in context.\n", pf->pvar->var); + /* Search child die for local variables and parameters. */ + if (!die_find_variable_at(sc_die, pf->pvar->var, pf->addr, &vr_die)) { + /* Search again in global variables */ + if (!die_find_variable_at(&pf->cu_die, pf->pvar->var, 0, &vr_die)) + ret = -ENOENT; + } + if (ret >= 0) + ret = convert_variable(&vr_die, pf); + + if (ret < 0) + pr_warning("Failed to find '%s' in this function.\n", + pf->pvar->var); + return ret; +} + +/* Convert subprogram DIE to trace point */ +static int convert_to_trace_point(Dwarf_Die *sp_die, Dwarf_Addr paddr, + bool retprobe, struct probe_trace_point *tp) +{ + Dwarf_Addr eaddr, highaddr; + const char *name; + + /* Copy the name of probe point */ + name = dwarf_diename(sp_die); + if (name) { + if (dwarf_entrypc(sp_die, &eaddr) != 0) { + pr_warning("Failed to get entry address of %s\n", + dwarf_diename(sp_die)); + return -ENOENT; + } + if (dwarf_highpc(sp_die, &highaddr) != 0) { + pr_warning("Failed to get end address of %s\n", + dwarf_diename(sp_die)); + return -ENOENT; + } + if (paddr > highaddr) { + pr_warning("Offset specified is greater than size of %s\n", + dwarf_diename(sp_die)); + return -EINVAL; + } + tp->symbol = strdup(name); + if (tp->symbol == NULL) + return -ENOMEM; + tp->offset = (unsigned long)(paddr - eaddr); + } else + /* This function has no name. */ + tp->offset = (unsigned long)paddr; + + /* Return probe must be on the head of a subprogram */ + if (retprobe) { + if (eaddr != paddr) { + pr_warning("Return probe must be on the head of" + " a real function.\n"); + return -EINVAL; + } + tp->retprobe = true; + } + + return 0; +} + +/* Call probe_finder callback with scope DIE */ +static int call_probe_finder(Dwarf_Die *sc_die, struct probe_finder *pf) +{ + Dwarf_Attribute fb_attr; + size_t nops; + int ret; + + if (!sc_die) { + pr_err("Caller must pass a scope DIE. Program error.\n"); + return -EINVAL; + } + + /* If not a real subprogram, find a real one */ + if (dwarf_tag(sc_die) != DW_TAG_subprogram) { + if (!die_find_realfunc(&pf->cu_die, pf->addr, &pf->sp_die)) { + pr_warning("Failed to find probe point in any " + "functions.\n"); + return -ENOENT; + } + } else + memcpy(&pf->sp_die, sc_die, sizeof(Dwarf_Die)); + + /* Get the frame base attribute/ops from subprogram */ + dwarf_attr(&pf->sp_die, DW_AT_frame_base, &fb_attr); + ret = dwarf_getlocation_addr(&fb_attr, pf->addr, &pf->fb_ops, &nops, 1); + if (ret <= 0 || nops == 0) { + pf->fb_ops = NULL; +#if _ELFUTILS_PREREQ(0, 142) + } else if (nops == 1 && pf->fb_ops[0].atom == DW_OP_call_frame_cfa && + pf->cfi != NULL) { + Dwarf_Frame *frame; + if (dwarf_cfi_addrframe(pf->cfi, pf->addr, &frame) != 0 || + dwarf_frame_cfa(frame, &pf->fb_ops, &nops) != 0) { + pr_warning("Failed to get call frame on 0x%jx\n", + (uintmax_t)pf->addr); + return -ENOENT; + } +#endif + } + + /* Call finder's callback handler */ + ret = pf->callback(sc_die, pf); + + /* *pf->fb_ops will be cached in libdw. Don't free it. */ + pf->fb_ops = NULL; + + return ret; +} + +struct find_scope_param { + const char *function; + const char *file; + int line; + int diff; + Dwarf_Die *die_mem; + bool found; +}; + +static int find_best_scope_cb(Dwarf_Die *fn_die, void *data) +{ + struct find_scope_param *fsp = data; + const char *file; + int lno; + + /* Skip if declared file name does not match */ + if (fsp->file) { + file = dwarf_decl_file(fn_die); + if (!file || strcmp(fsp->file, file) != 0) + return 0; + } + /* If the function name is given, that's what user expects */ + if (fsp->function) { + if (die_compare_name(fn_die, fsp->function)) { + memcpy(fsp->die_mem, fn_die, sizeof(Dwarf_Die)); + fsp->found = true; + return 1; + } + } else { + /* With the line number, find the nearest declared DIE */ + dwarf_decl_line(fn_die, &lno); + if (lno < fsp->line && fsp->diff > fsp->line - lno) { + /* Keep a candidate and continue */ + fsp->diff = fsp->line - lno; + memcpy(fsp->die_mem, fn_die, sizeof(Dwarf_Die)); + fsp->found = true; + } + } + return 0; +} + +/* Find an appropriate scope fits to given conditions */ +static Dwarf_Die *find_best_scope(struct probe_finder *pf, Dwarf_Die *die_mem) +{ + struct find_scope_param fsp = { + .function = pf->pev->point.function, + .file = pf->fname, + .line = pf->lno, + .diff = INT_MAX, + .die_mem = die_mem, + .found = false, + }; + + cu_walk_functions_at(&pf->cu_die, pf->addr, find_best_scope_cb, &fsp); + + return fsp.found ? die_mem : NULL; +} + +static int probe_point_line_walker(const char *fname, int lineno, + Dwarf_Addr addr, void *data) +{ + struct probe_finder *pf = data; + Dwarf_Die *sc_die, die_mem; + int ret; + + if (lineno != pf->lno || strtailcmp(fname, pf->fname) != 0) + return 0; + + pf->addr = addr; + sc_die = find_best_scope(pf, &die_mem); + if (!sc_die) { + pr_warning("Failed to find scope of probe point.\n"); + return -ENOENT; + } + + ret = call_probe_finder(sc_die, pf); + + /* Continue if no error, because the line will be in inline function */ + return ret < 0 ? ret : 0; +} + +/* Find probe point from its line number */ +static int find_probe_point_by_line(struct probe_finder *pf) +{ + return die_walk_lines(&pf->cu_die, probe_point_line_walker, pf); +} + +/* Find lines which match lazy pattern */ +static int find_lazy_match_lines(struct list_head *head, + const char *fname, const char *pat) +{ + FILE *fp; + char *line = NULL; + size_t line_len; + ssize_t len; + int count = 0, linenum = 1; + + fp = fopen(fname, "r"); + if (!fp) { + pr_warning("Failed to open %s: %s\n", fname, strerror(errno)); + return -errno; + } + + while ((len = getline(&line, &line_len, fp)) > 0) { + + if (line[len - 1] == '\n') + line[len - 1] = '\0'; + + if (strlazymatch(line, pat)) { + line_list__add_line(head, linenum); + count++; + } + linenum++; + } + + if (ferror(fp)) + count = -errno; + free(line); + fclose(fp); + + if (count == 0) + pr_debug("No matched lines found in %s.\n", fname); + return count; +} + +static int probe_point_lazy_walker(const char *fname, int lineno, + Dwarf_Addr addr, void *data) +{ + struct probe_finder *pf = data; + Dwarf_Die *sc_die, die_mem; + int ret; + + if (!line_list__has_line(&pf->lcache, lineno) || + strtailcmp(fname, pf->fname) != 0) + return 0; + + pr_debug("Probe line found: line:%d addr:0x%llx\n", + lineno, (unsigned long long)addr); + pf->addr = addr; + pf->lno = lineno; + sc_die = find_best_scope(pf, &die_mem); + if (!sc_die) { + pr_warning("Failed to find scope of probe point.\n"); + return -ENOENT; + } + + ret = call_probe_finder(sc_die, pf); + + /* + * Continue if no error, because the lazy pattern will match + * to other lines + */ + return ret < 0 ? ret : 0; +} + +/* Find probe points from lazy pattern */ +static int find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf) +{ + int ret = 0; + + if (list_empty(&pf->lcache)) { + /* Matching lazy line pattern */ + ret = find_lazy_match_lines(&pf->lcache, pf->fname, + pf->pev->point.lazy_line); + if (ret <= 0) + return ret; + } + + return die_walk_lines(sp_die, probe_point_lazy_walker, pf); +} + +static int probe_point_inline_cb(Dwarf_Die *in_die, void *data) +{ + struct probe_finder *pf = data; + struct perf_probe_point *pp = &pf->pev->point; + Dwarf_Addr addr; + int ret; + + if (pp->lazy_line) + ret = find_probe_point_lazy(in_die, pf); + else { + /* Get probe address */ + if (dwarf_entrypc(in_die, &addr) != 0) { + pr_warning("Failed to get entry address of %s.\n", + dwarf_diename(in_die)); + return -ENOENT; + } + pf->addr = addr; + pf->addr += pp->offset; + pr_debug("found inline addr: 0x%jx\n", + (uintmax_t)pf->addr); + + ret = call_probe_finder(in_die, pf); + } + + return ret; +} + +/* Callback parameter with return value for libdw */ +struct dwarf_callback_param { + void *data; + int retval; +}; + +/* Search function from function name */ +static int probe_point_search_cb(Dwarf_Die *sp_die, void *data) +{ + struct dwarf_callback_param *param = data; + struct probe_finder *pf = param->data; + struct perf_probe_point *pp = &pf->pev->point; + Dwarf_Attribute attr; + + /* Check tag and diename */ + if (dwarf_tag(sp_die) != DW_TAG_subprogram || + !die_compare_name(sp_die, pp->function) || + dwarf_attr(sp_die, DW_AT_declaration, &attr)) + return DWARF_CB_OK; + + /* Check declared file */ + if (pp->file && strtailcmp(pp->file, dwarf_decl_file(sp_die))) + return DWARF_CB_OK; + + pf->fname = dwarf_decl_file(sp_die); + if (pp->line) { /* Function relative line */ + dwarf_decl_line(sp_die, &pf->lno); + pf->lno += pp->line; + param->retval = find_probe_point_by_line(pf); + } else if (!dwarf_func_inline(sp_die)) { + /* Real function */ + if (pp->lazy_line) + param->retval = find_probe_point_lazy(sp_die, pf); + else { + if (dwarf_entrypc(sp_die, &pf->addr) != 0) { + pr_warning("Failed to get entry address of " + "%s.\n", dwarf_diename(sp_die)); + param->retval = -ENOENT; + return DWARF_CB_ABORT; + } + pf->addr += pp->offset; + /* TODO: Check the address in this function */ + param->retval = call_probe_finder(sp_die, pf); + } + } else + /* Inlined function: search instances */ + param->retval = die_walk_instances(sp_die, + probe_point_inline_cb, (void *)pf); + + return DWARF_CB_ABORT; /* Exit; no same symbol in this CU. */ +} + +static int find_probe_point_by_func(struct probe_finder *pf) +{ + struct dwarf_callback_param _param = {.data = (void *)pf, + .retval = 0}; + dwarf_getfuncs(&pf->cu_die, probe_point_search_cb, &_param, 0); + return _param.retval; +} + +struct pubname_callback_param { + char *function; + char *file; + Dwarf_Die *cu_die; + Dwarf_Die *sp_die; + int found; +}; + +static int pubname_search_cb(Dwarf *dbg, Dwarf_Global *gl, void *data) +{ + struct pubname_callback_param *param = data; + + if (dwarf_offdie(dbg, gl->die_offset, param->sp_die)) { + if (dwarf_tag(param->sp_die) != DW_TAG_subprogram) + return DWARF_CB_OK; + + if (die_compare_name(param->sp_die, param->function)) { + if (!dwarf_offdie(dbg, gl->cu_offset, param->cu_die)) + return DWARF_CB_OK; + + if (param->file && + strtailcmp(param->file, dwarf_decl_file(param->sp_die))) + return DWARF_CB_OK; + + param->found = 1; + return DWARF_CB_ABORT; + } + } + + return DWARF_CB_OK; +} + +/* Find probe points from debuginfo */ +static int debuginfo__find_probes(struct debuginfo *self, + struct probe_finder *pf) +{ + struct perf_probe_point *pp = &pf->pev->point; + Dwarf_Off off, noff; + size_t cuhl; + Dwarf_Die *diep; + int ret = 0; + +#if _ELFUTILS_PREREQ(0, 142) + /* Get the call frame information from this dwarf */ + pf->cfi = dwarf_getcfi(self->dbg); +#endif + + off = 0; + line_list__init(&pf->lcache); + + /* Fastpath: lookup by function name from .debug_pubnames section */ + if (pp->function) { + struct pubname_callback_param pubname_param = { + .function = pp->function, + .file = pp->file, + .cu_die = &pf->cu_die, + .sp_die = &pf->sp_die, + .found = 0, + }; + struct dwarf_callback_param probe_param = { + .data = pf, + }; + + dwarf_getpubnames(self->dbg, pubname_search_cb, + &pubname_param, 0); + if (pubname_param.found) { + ret = probe_point_search_cb(&pf->sp_die, &probe_param); + if (ret) + goto found; + } + } + + /* Loop on CUs (Compilation Unit) */ + while (!dwarf_nextcu(self->dbg, off, &noff, &cuhl, NULL, NULL, NULL)) { + /* Get the DIE(Debugging Information Entry) of this CU */ + diep = dwarf_offdie(self->dbg, off + cuhl, &pf->cu_die); + if (!diep) + continue; + + /* Check if target file is included. */ + if (pp->file) + pf->fname = cu_find_realpath(&pf->cu_die, pp->file); + else + pf->fname = NULL; + + if (!pp->file || pf->fname) { + if (pp->function) + ret = find_probe_point_by_func(pf); + else if (pp->lazy_line) + ret = find_probe_point_lazy(NULL, pf); + else { + pf->lno = pp->line; + ret = find_probe_point_by_line(pf); + } + if (ret < 0) + break; + } + off = noff; + } + +found: + line_list__free(&pf->lcache); + + return ret; +} + +/* Add a found probe point into trace event list */ +static int add_probe_trace_event(Dwarf_Die *sc_die, struct probe_finder *pf) +{ + struct trace_event_finder *tf = + container_of(pf, struct trace_event_finder, pf); + struct probe_trace_event *tev; + int ret, i; + + /* Check number of tevs */ + if (tf->ntevs == tf->max_tevs) { + pr_warning("Too many( > %d) probe point found.\n", + tf->max_tevs); + return -ERANGE; + } + tev = &tf->tevs[tf->ntevs++]; + + /* Trace point should be converted from subprogram DIE */ + ret = convert_to_trace_point(&pf->sp_die, pf->addr, + pf->pev->point.retprobe, &tev->point); + if (ret < 0) + return ret; + + pr_debug("Probe point found: %s+%lu\n", tev->point.symbol, + tev->point.offset); + + /* Find each argument */ + tev->nargs = pf->pev->nargs; + tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs); + if (tev->args == NULL) + return -ENOMEM; + for (i = 0; i < pf->pev->nargs; i++) { + pf->pvar = &pf->pev->args[i]; + pf->tvar = &tev->args[i]; + /* Variable should be found from scope DIE */ + ret = find_variable(sc_die, pf); + if (ret != 0) + return ret; + } + + return 0; +} + +/* Find probe_trace_events specified by perf_probe_event from debuginfo */ +int debuginfo__find_trace_events(struct debuginfo *self, + struct perf_probe_event *pev, + struct probe_trace_event **tevs, int max_tevs) +{ + struct trace_event_finder tf = { + .pf = {.pev = pev, .callback = add_probe_trace_event}, + .max_tevs = max_tevs}; + int ret; + + /* Allocate result tevs array */ + *tevs = zalloc(sizeof(struct probe_trace_event) * max_tevs); + if (*tevs == NULL) + return -ENOMEM; + + tf.tevs = *tevs; + tf.ntevs = 0; + + ret = debuginfo__find_probes(self, &tf.pf); + if (ret < 0) { + free(*tevs); + *tevs = NULL; + return ret; + } + + return (ret < 0) ? ret : tf.ntevs; +} + +#define MAX_VAR_LEN 64 + +/* Collect available variables in this scope */ +static int collect_variables_cb(Dwarf_Die *die_mem, void *data) +{ + struct available_var_finder *af = data; + struct variable_list *vl; + char buf[MAX_VAR_LEN]; + int tag, ret; + + vl = &af->vls[af->nvls - 1]; + + tag = dwarf_tag(die_mem); + if (tag == DW_TAG_formal_parameter || + tag == DW_TAG_variable) { + ret = convert_variable_location(die_mem, af->pf.addr, + af->pf.fb_ops, NULL); + if (ret == 0) { + ret = die_get_varname(die_mem, buf, MAX_VAR_LEN); + pr_debug2("Add new var: %s\n", buf); + if (ret > 0) + strlist__add(vl->vars, buf); + } + } + + if (af->child && dwarf_haspc(die_mem, af->pf.addr)) + return DIE_FIND_CB_CONTINUE; + else + return DIE_FIND_CB_SIBLING; +} + +/* Add a found vars into available variables list */ +static int add_available_vars(Dwarf_Die *sc_die, struct probe_finder *pf) +{ + struct available_var_finder *af = + container_of(pf, struct available_var_finder, pf); + struct variable_list *vl; + Dwarf_Die die_mem; + int ret; + + /* Check number of tevs */ + if (af->nvls == af->max_vls) { + pr_warning("Too many( > %d) probe point found.\n", af->max_vls); + return -ERANGE; + } + vl = &af->vls[af->nvls++]; + + /* Trace point should be converted from subprogram DIE */ + ret = convert_to_trace_point(&pf->sp_die, pf->addr, + pf->pev->point.retprobe, &vl->point); + if (ret < 0) + return ret; + + pr_debug("Probe point found: %s+%lu\n", vl->point.symbol, + vl->point.offset); + + /* Find local variables */ + vl->vars = strlist__new(true, NULL); + if (vl->vars == NULL) + return -ENOMEM; + af->child = true; + die_find_child(sc_die, collect_variables_cb, (void *)af, &die_mem); + + /* Find external variables */ + if (!af->externs) + goto out; + /* Don't need to search child DIE for externs. */ + af->child = false; + die_find_child(&pf->cu_die, collect_variables_cb, (void *)af, &die_mem); + +out: + if (strlist__empty(vl->vars)) { + strlist__delete(vl->vars); + vl->vars = NULL; + } + + return ret; +} + +/* Find available variables at given probe point */ +int debuginfo__find_available_vars_at(struct debuginfo *self, + struct perf_probe_event *pev, + struct variable_list **vls, + int max_vls, bool externs) +{ + struct available_var_finder af = { + .pf = {.pev = pev, .callback = add_available_vars}, + .max_vls = max_vls, .externs = externs}; + int ret; + + /* Allocate result vls array */ + *vls = zalloc(sizeof(struct variable_list) * max_vls); + if (*vls == NULL) + return -ENOMEM; + + af.vls = *vls; + af.nvls = 0; + + ret = debuginfo__find_probes(self, &af.pf); + if (ret < 0) { + /* Free vlist for error */ + while (af.nvls--) { + if (af.vls[af.nvls].point.symbol) + free(af.vls[af.nvls].point.symbol); + if (af.vls[af.nvls].vars) + strlist__delete(af.vls[af.nvls].vars); + } + free(af.vls); + *vls = NULL; + return ret; + } + + return (ret < 0) ? ret : af.nvls; +} + +/* Reverse search */ +int debuginfo__find_probe_point(struct debuginfo *self, unsigned long addr, + struct perf_probe_point *ppt) +{ + Dwarf_Die cudie, spdie, indie; + Dwarf_Addr _addr, baseaddr; + const char *fname = NULL, *func = NULL, *tmp; + int baseline = 0, lineno = 0, ret = 0; + + /* Adjust address with bias */ + addr += self->bias; + + /* Find cu die */ + if (!dwarf_addrdie(self->dbg, (Dwarf_Addr)addr - self->bias, &cudie)) { + pr_warning("Failed to find debug information for address %lx\n", + addr); + ret = -EINVAL; + goto end; + } + + /* Find a corresponding line (filename and lineno) */ + cu_find_lineinfo(&cudie, addr, &fname, &lineno); + /* Don't care whether it failed or not */ + + /* Find a corresponding function (name, baseline and baseaddr) */ + if (die_find_realfunc(&cudie, (Dwarf_Addr)addr, &spdie)) { + /* Get function entry information */ + tmp = dwarf_diename(&spdie); + if (!tmp || + dwarf_entrypc(&spdie, &baseaddr) != 0 || + dwarf_decl_line(&spdie, &baseline) != 0) + goto post; + func = tmp; + + if (addr == (unsigned long)baseaddr) + /* Function entry - Relative line number is 0 */ + lineno = baseline; + else if (die_find_inlinefunc(&spdie, (Dwarf_Addr)addr, + &indie)) { + if (dwarf_entrypc(&indie, &_addr) == 0 && + _addr == addr) + /* + * addr is at an inline function entry. + * In this case, lineno should be the call-site + * line number. + */ + lineno = die_get_call_lineno(&indie); + else { + /* + * addr is in an inline function body. + * Since lineno points one of the lines + * of the inline function, baseline should + * be the entry line of the inline function. + */ + tmp = dwarf_diename(&indie); + if (tmp && + dwarf_decl_line(&spdie, &baseline) == 0) + func = tmp; + } + } + } + +post: + /* Make a relative line number or an offset */ + if (lineno) + ppt->line = lineno - baseline; + else if (func) + ppt->offset = addr - (unsigned long)baseaddr; + + /* Duplicate strings */ + if (func) { + ppt->function = strdup(func); + if (ppt->function == NULL) { + ret = -ENOMEM; + goto end; + } + } + if (fname) { + ppt->file = strdup(fname); + if (ppt->file == NULL) { + if (ppt->function) { + free(ppt->function); + ppt->function = NULL; + } + ret = -ENOMEM; + goto end; + } + } +end: + if (ret == 0 && (fname || func)) + ret = 1; /* Found a point */ + return ret; +} + +/* Add a line and store the src path */ +static int line_range_add_line(const char *src, unsigned int lineno, + struct line_range *lr) +{ + /* Copy source path */ + if (!lr->path) { + lr->path = strdup(src); + if (lr->path == NULL) + return -ENOMEM; + } + return line_list__add_line(&lr->line_list, lineno); +} + +static int line_range_walk_cb(const char *fname, int lineno, + Dwarf_Addr addr __maybe_unused, + void *data) +{ + struct line_finder *lf = data; + + if ((strtailcmp(fname, lf->fname) != 0) || + (lf->lno_s > lineno || lf->lno_e < lineno)) + return 0; + + if (line_range_add_line(fname, lineno, lf->lr) < 0) + return -EINVAL; + + return 0; +} + +/* Find line range from its line number */ +static int find_line_range_by_line(Dwarf_Die *sp_die, struct line_finder *lf) +{ + int ret; + + ret = die_walk_lines(sp_die ?: &lf->cu_die, line_range_walk_cb, lf); + + /* Update status */ + if (ret >= 0) + if (!list_empty(&lf->lr->line_list)) + ret = lf->found = 1; + else + ret = 0; /* Lines are not found */ + else { + free(lf->lr->path); + lf->lr->path = NULL; + } + return ret; +} + +static int line_range_inline_cb(Dwarf_Die *in_die, void *data) +{ + find_line_range_by_line(in_die, data); + + /* + * We have to check all instances of inlined function, because + * some execution paths can be optimized out depends on the + * function argument of instances + */ + return 0; +} + +/* Search function from function name */ +static int line_range_search_cb(Dwarf_Die *sp_die, void *data) +{ + struct dwarf_callback_param *param = data; + struct line_finder *lf = param->data; + struct line_range *lr = lf->lr; + + /* Check declared file */ + if (lr->file && strtailcmp(lr->file, dwarf_decl_file(sp_die))) + return DWARF_CB_OK; + + if (dwarf_tag(sp_die) == DW_TAG_subprogram && + die_compare_name(sp_die, lr->function)) { + lf->fname = dwarf_decl_file(sp_die); + dwarf_decl_line(sp_die, &lr->offset); + pr_debug("fname: %s, lineno:%d\n", lf->fname, lr->offset); + lf->lno_s = lr->offset + lr->start; + if (lf->lno_s < 0) /* Overflow */ + lf->lno_s = INT_MAX; + lf->lno_e = lr->offset + lr->end; + if (lf->lno_e < 0) /* Overflow */ + lf->lno_e = INT_MAX; + pr_debug("New line range: %d to %d\n", lf->lno_s, lf->lno_e); + lr->start = lf->lno_s; + lr->end = lf->lno_e; + if (dwarf_func_inline(sp_die)) + param->retval = die_walk_instances(sp_die, + line_range_inline_cb, lf); + else + param->retval = find_line_range_by_line(sp_die, lf); + return DWARF_CB_ABORT; + } + return DWARF_CB_OK; +} + +static int find_line_range_by_func(struct line_finder *lf) +{ + struct dwarf_callback_param param = {.data = (void *)lf, .retval = 0}; + dwarf_getfuncs(&lf->cu_die, line_range_search_cb, ¶m, 0); + return param.retval; +} + +int debuginfo__find_line_range(struct debuginfo *self, struct line_range *lr) +{ + struct line_finder lf = {.lr = lr, .found = 0}; + int ret = 0; + Dwarf_Off off = 0, noff; + size_t cuhl; + Dwarf_Die *diep; + const char *comp_dir; + + /* Fastpath: lookup by function name from .debug_pubnames section */ + if (lr->function) { + struct pubname_callback_param pubname_param = { + .function = lr->function, .file = lr->file, + .cu_die = &lf.cu_die, .sp_die = &lf.sp_die, .found = 0}; + struct dwarf_callback_param line_range_param = { + .data = (void *)&lf, .retval = 0}; + + dwarf_getpubnames(self->dbg, pubname_search_cb, + &pubname_param, 0); + if (pubname_param.found) { + line_range_search_cb(&lf.sp_die, &line_range_param); + if (lf.found) + goto found; + } + } + + /* Loop on CUs (Compilation Unit) */ + while (!lf.found && ret >= 0) { + if (dwarf_nextcu(self->dbg, off, &noff, &cuhl, + NULL, NULL, NULL) != 0) + break; + + /* Get the DIE(Debugging Information Entry) of this CU */ + diep = dwarf_offdie(self->dbg, off + cuhl, &lf.cu_die); + if (!diep) + continue; + + /* Check if target file is included. */ + if (lr->file) + lf.fname = cu_find_realpath(&lf.cu_die, lr->file); + else + lf.fname = 0; + + if (!lr->file || lf.fname) { + if (lr->function) + ret = find_line_range_by_func(&lf); + else { + lf.lno_s = lr->start; + lf.lno_e = lr->end; + ret = find_line_range_by_line(NULL, &lf); + } + } + off = noff; + } + +found: + /* Store comp_dir */ + if (lf.found) { + comp_dir = cu_get_comp_dir(&lf.cu_die); + if (comp_dir) { + lr->comp_dir = strdup(comp_dir); + if (!lr->comp_dir) + ret = -ENOMEM; + } + } + + pr_debug("path: %s\n", lr->path); + return (ret < 0) ? ret : lf.found; +} + diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h new file mode 100644 index 00000000..17e94d0c --- /dev/null +++ b/tools/perf/util/probe-finder.h @@ -0,0 +1,107 @@ +#ifndef _PROBE_FINDER_H +#define _PROBE_FINDER_H + +#include <stdbool.h> +#include "util.h" +#include "probe-event.h" + +#define MAX_PROBE_BUFFER 1024 +#define MAX_PROBES 128 + +static inline int is_c_varname(const char *name) +{ + /* TODO */ + return isalpha(name[0]) || name[0] == '_'; +} + +#ifdef DWARF_SUPPORT + +#include "dwarf-aux.h" + +/* TODO: export debuginfo data structure even if no dwarf support */ + +/* debug information structure */ +struct debuginfo { + Dwarf *dbg; + Dwfl *dwfl; + Dwarf_Addr bias; +}; + +extern struct debuginfo *debuginfo__new(const char *path); +extern struct debuginfo *debuginfo__new_online_kernel(unsigned long addr); +extern void debuginfo__delete(struct debuginfo *self); + +/* Find probe_trace_events specified by perf_probe_event from debuginfo */ +extern int debuginfo__find_trace_events(struct debuginfo *self, + struct perf_probe_event *pev, + struct probe_trace_event **tevs, + int max_tevs); + +/* Find a perf_probe_point from debuginfo */ +extern int debuginfo__find_probe_point(struct debuginfo *self, + unsigned long addr, + struct perf_probe_point *ppt); + +/* Find a line range */ +extern int debuginfo__find_line_range(struct debuginfo *self, + struct line_range *lr); + +/* Find available variables */ +extern int debuginfo__find_available_vars_at(struct debuginfo *self, + struct perf_probe_event *pev, + struct variable_list **vls, + int max_points, bool externs); + +struct probe_finder { + struct perf_probe_event *pev; /* Target probe event */ + + /* Callback when a probe point is found */ + int (*callback)(Dwarf_Die *sc_die, struct probe_finder *pf); + + /* For function searching */ + int lno; /* Line number */ + Dwarf_Addr addr; /* Address */ + const char *fname; /* Real file name */ + Dwarf_Die cu_die; /* Current CU */ + Dwarf_Die sp_die; + struct list_head lcache; /* Line cache for lazy match */ + + /* For variable searching */ +#if _ELFUTILS_PREREQ(0, 142) + Dwarf_CFI *cfi; /* Call Frame Information */ +#endif + Dwarf_Op *fb_ops; /* Frame base attribute */ + struct perf_probe_arg *pvar; /* Current target variable */ + struct probe_trace_arg *tvar; /* Current result variable */ +}; + +struct trace_event_finder { + struct probe_finder pf; + struct probe_trace_event *tevs; /* Found trace events */ + int ntevs; /* Number of trace events */ + int max_tevs; /* Max number of trace events */ +}; + +struct available_var_finder { + struct probe_finder pf; + struct variable_list *vls; /* Found variable lists */ + int nvls; /* Number of variable lists */ + int max_vls; /* Max no. of variable lists */ + bool externs; /* Find external vars too */ + bool child; /* Search child scopes */ +}; + +struct line_finder { + struct line_range *lr; /* Target line range */ + + const char *fname; /* File name */ + int lno_s; /* Start line number */ + int lno_e; /* End line number */ + Dwarf_Die cu_die; /* Current CU */ + Dwarf_Die sp_die; + int found; +}; + +#endif /* DWARF_SUPPORT */ + +#endif /*_PROBE_FINDER_H */ diff --git a/tools/perf/util/pstack.c b/tools/perf/util/pstack.c new file mode 100644 index 00000000..daa17aeb --- /dev/null +++ b/tools/perf/util/pstack.c @@ -0,0 +1,75 @@ +/* + * Simple pointer stack + * + * (c) 2010 Arnaldo Carvalho de Melo <acme@redhat.com> + */ + +#include "util.h" +#include "pstack.h" +#include <linux/kernel.h> +#include <stdlib.h> + +struct pstack { + unsigned short top; + unsigned short max_nr_entries; + void *entries[0]; +}; + +struct pstack *pstack__new(unsigned short max_nr_entries) +{ + struct pstack *pstack = zalloc((sizeof(*pstack) + + max_nr_entries * sizeof(void *))); + if (pstack != NULL) + pstack->max_nr_entries = max_nr_entries; + return pstack; +} + +void pstack__delete(struct pstack *pstack) +{ + free(pstack); +} + +bool pstack__empty(const struct pstack *pstack) +{ + return pstack->top == 0; +} + +void pstack__remove(struct pstack *pstack, void *key) +{ + unsigned short i = pstack->top, last_index = pstack->top - 1; + + while (i-- != 0) { + if (pstack->entries[i] == key) { + if (i < last_index) + memmove(pstack->entries + i, + pstack->entries + i + 1, + (last_index - i) * sizeof(void *)); + --pstack->top; + return; + } + } + pr_err("%s: %p not on the pstack!\n", __func__, key); +} + +void pstack__push(struct pstack *pstack, void *key) +{ + if (pstack->top == pstack->max_nr_entries) { + pr_err("%s: top=%d, overflow!\n", __func__, pstack->top); + return; + } + pstack->entries[pstack->top++] = key; +} + +void *pstack__pop(struct pstack *pstack) +{ + void *ret; + + if (pstack->top == 0) { + pr_err("%s: underflow!\n", __func__); + return NULL; + } + + ret = pstack->entries[--pstack->top]; + pstack->entries[pstack->top] = NULL; + return ret; +} diff --git a/tools/perf/util/pstack.h b/tools/perf/util/pstack.h new file mode 100644 index 00000000..4cedea59 --- /dev/null +++ b/tools/perf/util/pstack.h @@ -0,0 +1,14 @@ +#ifndef _PERF_PSTACK_ +#define _PERF_PSTACK_ + +#include <stdbool.h> + +struct pstack; +struct pstack *pstack__new(unsigned short max_nr_entries); +void pstack__delete(struct pstack *self); +bool pstack__empty(const struct pstack *self); +void pstack__remove(struct pstack *self, void *key); +void pstack__push(struct pstack *self, void *key); +void *pstack__pop(struct pstack *self); + +#endif /* _PERF_PSTACK_ */ diff --git a/tools/perf/util/python-ext-sources b/tools/perf/util/python-ext-sources new file mode 100644 index 00000000..64536a99 --- /dev/null +++ b/tools/perf/util/python-ext-sources @@ -0,0 +1,22 @@ +# +# List of files needed by perf python extension +# +# Each source file must be placed on its own line so that it can be +# processed by Makefile and util/setup.py accordingly. +# + +util/python.c +util/ctype.c +util/evlist.c +util/evsel.c +util/cpumap.c +util/hweight.c +util/thread_map.c +util/util.c +util/xyarray.c +util/cgroup.c +util/debugfs.c +util/rblist.c +util/strlist.c +util/sysfs.c +../../lib/rbtree.c diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c new file mode 100644 index 00000000..925e0c3e --- /dev/null +++ b/tools/perf/util/python.c @@ -0,0 +1,1056 @@ +#include <Python.h> +#include <structmember.h> +#include <inttypes.h> +#include <poll.h> +#include "evlist.h" +#include "evsel.h" +#include "event.h" +#include "cpumap.h" +#include "thread_map.h" + +/* Define PyVarObject_HEAD_INIT for python 2.5 */ +#ifndef PyVarObject_HEAD_INIT +# define PyVarObject_HEAD_INIT(type, size) PyObject_HEAD_INIT(type) size, +#endif + +struct throttle_event { + struct perf_event_header header; + u64 time; + u64 id; + u64 stream_id; +}; + +PyMODINIT_FUNC initperf(void); + +#define member_def(type, member, ptype, help) \ + { #member, ptype, \ + offsetof(struct pyrf_event, event) + offsetof(struct type, member), \ + 0, help } + +#define sample_member_def(name, member, ptype, help) \ + { #name, ptype, \ + offsetof(struct pyrf_event, sample) + offsetof(struct perf_sample, member), \ + 0, help } + +struct pyrf_event { + PyObject_HEAD + struct perf_sample sample; + union perf_event event; +}; + +#define sample_members \ + sample_member_def(sample_ip, ip, T_ULONGLONG, "event type"), \ + sample_member_def(sample_pid, pid, T_INT, "event pid"), \ + sample_member_def(sample_tid, tid, T_INT, "event tid"), \ + sample_member_def(sample_time, time, T_ULONGLONG, "event timestamp"), \ + sample_member_def(sample_addr, addr, T_ULONGLONG, "event addr"), \ + sample_member_def(sample_id, id, T_ULONGLONG, "event id"), \ + sample_member_def(sample_stream_id, stream_id, T_ULONGLONG, "event stream id"), \ + sample_member_def(sample_period, period, T_ULONGLONG, "event period"), \ + sample_member_def(sample_cpu, cpu, T_UINT, "event cpu"), + +static char pyrf_mmap_event__doc[] = PyDoc_STR("perf mmap event object."); + +static PyMemberDef pyrf_mmap_event__members[] = { + sample_members + member_def(perf_event_header, type, T_UINT, "event type"), + member_def(mmap_event, pid, T_UINT, "event pid"), + member_def(mmap_event, tid, T_UINT, "event tid"), + member_def(mmap_event, start, T_ULONGLONG, "start of the map"), + member_def(mmap_event, len, T_ULONGLONG, "map length"), + member_def(mmap_event, pgoff, T_ULONGLONG, "page offset"), + member_def(mmap_event, filename, T_STRING_INPLACE, "backing store"), + { .name = NULL, }, +}; + +static PyObject *pyrf_mmap_event__repr(struct pyrf_event *pevent) +{ + PyObject *ret; + char *s; + + if (asprintf(&s, "{ type: mmap, pid: %u, tid: %u, start: %#" PRIx64 ", " + "length: %#" PRIx64 ", offset: %#" PRIx64 ", " + "filename: %s }", + pevent->event.mmap.pid, pevent->event.mmap.tid, + pevent->event.mmap.start, pevent->event.mmap.len, + pevent->event.mmap.pgoff, pevent->event.mmap.filename) < 0) { + ret = PyErr_NoMemory(); + } else { + ret = PyString_FromString(s); + free(s); + } + return ret; +} + +static PyTypeObject pyrf_mmap_event__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.mmap_event", + .tp_basicsize = sizeof(struct pyrf_event), + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_mmap_event__doc, + .tp_members = pyrf_mmap_event__members, + .tp_repr = (reprfunc)pyrf_mmap_event__repr, +}; + +static char pyrf_task_event__doc[] = PyDoc_STR("perf task (fork/exit) event object."); + +static PyMemberDef pyrf_task_event__members[] = { + sample_members + member_def(perf_event_header, type, T_UINT, "event type"), + member_def(fork_event, pid, T_UINT, "event pid"), + member_def(fork_event, ppid, T_UINT, "event ppid"), + member_def(fork_event, tid, T_UINT, "event tid"), + member_def(fork_event, ptid, T_UINT, "event ptid"), + member_def(fork_event, time, T_ULONGLONG, "timestamp"), + { .name = NULL, }, +}; + +static PyObject *pyrf_task_event__repr(struct pyrf_event *pevent) +{ + return PyString_FromFormat("{ type: %s, pid: %u, ppid: %u, tid: %u, " + "ptid: %u, time: %" PRIu64 "}", + pevent->event.header.type == PERF_RECORD_FORK ? "fork" : "exit", + pevent->event.fork.pid, + pevent->event.fork.ppid, + pevent->event.fork.tid, + pevent->event.fork.ptid, + pevent->event.fork.time); +} + +static PyTypeObject pyrf_task_event__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.task_event", + .tp_basicsize = sizeof(struct pyrf_event), + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_task_event__doc, + .tp_members = pyrf_task_event__members, + .tp_repr = (reprfunc)pyrf_task_event__repr, +}; + +static char pyrf_comm_event__doc[] = PyDoc_STR("perf comm event object."); + +static PyMemberDef pyrf_comm_event__members[] = { + sample_members + member_def(perf_event_header, type, T_UINT, "event type"), + member_def(comm_event, pid, T_UINT, "event pid"), + member_def(comm_event, tid, T_UINT, "event tid"), + member_def(comm_event, comm, T_STRING_INPLACE, "process name"), + { .name = NULL, }, +}; + +static PyObject *pyrf_comm_event__repr(struct pyrf_event *pevent) +{ + return PyString_FromFormat("{ type: comm, pid: %u, tid: %u, comm: %s }", + pevent->event.comm.pid, + pevent->event.comm.tid, + pevent->event.comm.comm); +} + +static PyTypeObject pyrf_comm_event__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.comm_event", + .tp_basicsize = sizeof(struct pyrf_event), + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_comm_event__doc, + .tp_members = pyrf_comm_event__members, + .tp_repr = (reprfunc)pyrf_comm_event__repr, +}; + +static char pyrf_throttle_event__doc[] = PyDoc_STR("perf throttle event object."); + +static PyMemberDef pyrf_throttle_event__members[] = { + sample_members + member_def(perf_event_header, type, T_UINT, "event type"), + member_def(throttle_event, time, T_ULONGLONG, "timestamp"), + member_def(throttle_event, id, T_ULONGLONG, "event id"), + member_def(throttle_event, stream_id, T_ULONGLONG, "event stream id"), + { .name = NULL, }, +}; + +static PyObject *pyrf_throttle_event__repr(struct pyrf_event *pevent) +{ + struct throttle_event *te = (struct throttle_event *)(&pevent->event.header + 1); + + return PyString_FromFormat("{ type: %sthrottle, time: %" PRIu64 ", id: %" PRIu64 + ", stream_id: %" PRIu64 " }", + pevent->event.header.type == PERF_RECORD_THROTTLE ? "" : "un", + te->time, te->id, te->stream_id); +} + +static PyTypeObject pyrf_throttle_event__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.throttle_event", + .tp_basicsize = sizeof(struct pyrf_event), + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_throttle_event__doc, + .tp_members = pyrf_throttle_event__members, + .tp_repr = (reprfunc)pyrf_throttle_event__repr, +}; + +static char pyrf_lost_event__doc[] = PyDoc_STR("perf lost event object."); + +static PyMemberDef pyrf_lost_event__members[] = { + sample_members + member_def(lost_event, id, T_ULONGLONG, "event id"), + member_def(lost_event, lost, T_ULONGLONG, "number of lost events"), + { .name = NULL, }, +}; + +static PyObject *pyrf_lost_event__repr(struct pyrf_event *pevent) +{ + PyObject *ret; + char *s; + + if (asprintf(&s, "{ type: lost, id: %#" PRIx64 ", " + "lost: %#" PRIx64 " }", + pevent->event.lost.id, pevent->event.lost.lost) < 0) { + ret = PyErr_NoMemory(); + } else { + ret = PyString_FromString(s); + free(s); + } + return ret; +} + +static PyTypeObject pyrf_lost_event__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.lost_event", + .tp_basicsize = sizeof(struct pyrf_event), + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_lost_event__doc, + .tp_members = pyrf_lost_event__members, + .tp_repr = (reprfunc)pyrf_lost_event__repr, +}; + +static char pyrf_read_event__doc[] = PyDoc_STR("perf read event object."); + +static PyMemberDef pyrf_read_event__members[] = { + sample_members + member_def(read_event, pid, T_UINT, "event pid"), + member_def(read_event, tid, T_UINT, "event tid"), + { .name = NULL, }, +}; + +static PyObject *pyrf_read_event__repr(struct pyrf_event *pevent) +{ + return PyString_FromFormat("{ type: read, pid: %u, tid: %u }", + pevent->event.read.pid, + pevent->event.read.tid); + /* + * FIXME: return the array of read values, + * making this method useful ;-) + */ +} + +static PyTypeObject pyrf_read_event__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.read_event", + .tp_basicsize = sizeof(struct pyrf_event), + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_read_event__doc, + .tp_members = pyrf_read_event__members, + .tp_repr = (reprfunc)pyrf_read_event__repr, +}; + +static char pyrf_sample_event__doc[] = PyDoc_STR("perf sample event object."); + +static PyMemberDef pyrf_sample_event__members[] = { + sample_members + member_def(perf_event_header, type, T_UINT, "event type"), + { .name = NULL, }, +}; + +static PyObject *pyrf_sample_event__repr(struct pyrf_event *pevent) +{ + PyObject *ret; + char *s; + + if (asprintf(&s, "{ type: sample }") < 0) { + ret = PyErr_NoMemory(); + } else { + ret = PyString_FromString(s); + free(s); + } + return ret; +} + +static PyTypeObject pyrf_sample_event__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.sample_event", + .tp_basicsize = sizeof(struct pyrf_event), + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_sample_event__doc, + .tp_members = pyrf_sample_event__members, + .tp_repr = (reprfunc)pyrf_sample_event__repr, +}; + +static int pyrf_event__setup_types(void) +{ + int err; + pyrf_mmap_event__type.tp_new = + pyrf_task_event__type.tp_new = + pyrf_comm_event__type.tp_new = + pyrf_lost_event__type.tp_new = + pyrf_read_event__type.tp_new = + pyrf_sample_event__type.tp_new = + pyrf_throttle_event__type.tp_new = PyType_GenericNew; + err = PyType_Ready(&pyrf_mmap_event__type); + if (err < 0) + goto out; + err = PyType_Ready(&pyrf_lost_event__type); + if (err < 0) + goto out; + err = PyType_Ready(&pyrf_task_event__type); + if (err < 0) + goto out; + err = PyType_Ready(&pyrf_comm_event__type); + if (err < 0) + goto out; + err = PyType_Ready(&pyrf_throttle_event__type); + if (err < 0) + goto out; + err = PyType_Ready(&pyrf_read_event__type); + if (err < 0) + goto out; + err = PyType_Ready(&pyrf_sample_event__type); + if (err < 0) + goto out; +out: + return err; +} + +static PyTypeObject *pyrf_event__type[] = { + [PERF_RECORD_MMAP] = &pyrf_mmap_event__type, + [PERF_RECORD_LOST] = &pyrf_lost_event__type, + [PERF_RECORD_COMM] = &pyrf_comm_event__type, + [PERF_RECORD_EXIT] = &pyrf_task_event__type, + [PERF_RECORD_THROTTLE] = &pyrf_throttle_event__type, + [PERF_RECORD_UNTHROTTLE] = &pyrf_throttle_event__type, + [PERF_RECORD_FORK] = &pyrf_task_event__type, + [PERF_RECORD_READ] = &pyrf_read_event__type, + [PERF_RECORD_SAMPLE] = &pyrf_sample_event__type, +}; + +static PyObject *pyrf_event__new(union perf_event *event) +{ + struct pyrf_event *pevent; + PyTypeObject *ptype; + + if (event->header.type < PERF_RECORD_MMAP || + event->header.type > PERF_RECORD_SAMPLE) + return NULL; + + ptype = pyrf_event__type[event->header.type]; + pevent = PyObject_New(struct pyrf_event, ptype); + if (pevent != NULL) + memcpy(&pevent->event, event, event->header.size); + return (PyObject *)pevent; +} + +struct pyrf_cpu_map { + PyObject_HEAD + + struct cpu_map *cpus; +}; + +static int pyrf_cpu_map__init(struct pyrf_cpu_map *pcpus, + PyObject *args, PyObject *kwargs) +{ + static char *kwlist[] = { "cpustr", NULL }; + char *cpustr = NULL; + + if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|s", + kwlist, &cpustr)) + return -1; + + pcpus->cpus = cpu_map__new(cpustr); + if (pcpus->cpus == NULL) + return -1; + return 0; +} + +static void pyrf_cpu_map__delete(struct pyrf_cpu_map *pcpus) +{ + cpu_map__delete(pcpus->cpus); + pcpus->ob_type->tp_free((PyObject*)pcpus); +} + +static Py_ssize_t pyrf_cpu_map__length(PyObject *obj) +{ + struct pyrf_cpu_map *pcpus = (void *)obj; + + return pcpus->cpus->nr; +} + +static PyObject *pyrf_cpu_map__item(PyObject *obj, Py_ssize_t i) +{ + struct pyrf_cpu_map *pcpus = (void *)obj; + + if (i >= pcpus->cpus->nr) + return NULL; + + return Py_BuildValue("i", pcpus->cpus->map[i]); +} + +static PySequenceMethods pyrf_cpu_map__sequence_methods = { + .sq_length = pyrf_cpu_map__length, + .sq_item = pyrf_cpu_map__item, +}; + +static char pyrf_cpu_map__doc[] = PyDoc_STR("cpu map object."); + +static PyTypeObject pyrf_cpu_map__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.cpu_map", + .tp_basicsize = sizeof(struct pyrf_cpu_map), + .tp_dealloc = (destructor)pyrf_cpu_map__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_cpu_map__doc, + .tp_as_sequence = &pyrf_cpu_map__sequence_methods, + .tp_init = (initproc)pyrf_cpu_map__init, +}; + +static int pyrf_cpu_map__setup_types(void) +{ + pyrf_cpu_map__type.tp_new = PyType_GenericNew; + return PyType_Ready(&pyrf_cpu_map__type); +} + +struct pyrf_thread_map { + PyObject_HEAD + + struct thread_map *threads; +}; + +static int pyrf_thread_map__init(struct pyrf_thread_map *pthreads, + PyObject *args, PyObject *kwargs) +{ + static char *kwlist[] = { "pid", "tid", "uid", NULL }; + int pid = -1, tid = -1, uid = UINT_MAX; + + if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|iii", + kwlist, &pid, &tid, &uid)) + return -1; + + pthreads->threads = thread_map__new(pid, tid, uid); + if (pthreads->threads == NULL) + return -1; + return 0; +} + +static void pyrf_thread_map__delete(struct pyrf_thread_map *pthreads) +{ + thread_map__delete(pthreads->threads); + pthreads->ob_type->tp_free((PyObject*)pthreads); +} + +static Py_ssize_t pyrf_thread_map__length(PyObject *obj) +{ + struct pyrf_thread_map *pthreads = (void *)obj; + + return pthreads->threads->nr; +} + +static PyObject *pyrf_thread_map__item(PyObject *obj, Py_ssize_t i) +{ + struct pyrf_thread_map *pthreads = (void *)obj; + + if (i >= pthreads->threads->nr) + return NULL; + + return Py_BuildValue("i", pthreads->threads->map[i]); +} + +static PySequenceMethods pyrf_thread_map__sequence_methods = { + .sq_length = pyrf_thread_map__length, + .sq_item = pyrf_thread_map__item, +}; + +static char pyrf_thread_map__doc[] = PyDoc_STR("thread map object."); + +static PyTypeObject pyrf_thread_map__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.thread_map", + .tp_basicsize = sizeof(struct pyrf_thread_map), + .tp_dealloc = (destructor)pyrf_thread_map__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_thread_map__doc, + .tp_as_sequence = &pyrf_thread_map__sequence_methods, + .tp_init = (initproc)pyrf_thread_map__init, +}; + +static int pyrf_thread_map__setup_types(void) +{ + pyrf_thread_map__type.tp_new = PyType_GenericNew; + return PyType_Ready(&pyrf_thread_map__type); +} + +struct pyrf_evsel { + PyObject_HEAD + + struct perf_evsel evsel; +}; + +static int pyrf_evsel__init(struct pyrf_evsel *pevsel, + PyObject *args, PyObject *kwargs) +{ + struct perf_event_attr attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES, + .sample_type = PERF_SAMPLE_PERIOD | PERF_SAMPLE_TID, + }; + static char *kwlist[] = { + "type", + "config", + "sample_freq", + "sample_period", + "sample_type", + "read_format", + "disabled", + "inherit", + "pinned", + "exclusive", + "exclude_user", + "exclude_kernel", + "exclude_hv", + "exclude_idle", + "mmap", + "comm", + "freq", + "inherit_stat", + "enable_on_exec", + "task", + "watermark", + "precise_ip", + "mmap_data", + "sample_id_all", + "wakeup_events", + "bp_type", + "bp_addr", + "bp_len", + NULL + }; + u64 sample_period = 0; + u32 disabled = 0, + inherit = 0, + pinned = 0, + exclusive = 0, + exclude_user = 0, + exclude_kernel = 0, + exclude_hv = 0, + exclude_idle = 0, + mmap = 0, + comm = 0, + freq = 1, + inherit_stat = 0, + enable_on_exec = 0, + task = 0, + watermark = 0, + precise_ip = 0, + mmap_data = 0, + sample_id_all = 1; + int idx = 0; + + if (!PyArg_ParseTupleAndKeywords(args, kwargs, + "|iKiKKiiiiiiiiiiiiiiiiiiiiiKK", kwlist, + &attr.type, &attr.config, &attr.sample_freq, + &sample_period, &attr.sample_type, + &attr.read_format, &disabled, &inherit, + &pinned, &exclusive, &exclude_user, + &exclude_kernel, &exclude_hv, &exclude_idle, + &mmap, &comm, &freq, &inherit_stat, + &enable_on_exec, &task, &watermark, + &precise_ip, &mmap_data, &sample_id_all, + &attr.wakeup_events, &attr.bp_type, + &attr.bp_addr, &attr.bp_len, &idx)) + return -1; + + /* union... */ + if (sample_period != 0) { + if (attr.sample_freq != 0) + return -1; /* FIXME: throw right exception */ + attr.sample_period = sample_period; + } + + /* Bitfields */ + attr.disabled = disabled; + attr.inherit = inherit; + attr.pinned = pinned; + attr.exclusive = exclusive; + attr.exclude_user = exclude_user; + attr.exclude_kernel = exclude_kernel; + attr.exclude_hv = exclude_hv; + attr.exclude_idle = exclude_idle; + attr.mmap = mmap; + attr.comm = comm; + attr.freq = freq; + attr.inherit_stat = inherit_stat; + attr.enable_on_exec = enable_on_exec; + attr.task = task; + attr.watermark = watermark; + attr.precise_ip = precise_ip; + attr.mmap_data = mmap_data; + attr.sample_id_all = sample_id_all; + + perf_evsel__init(&pevsel->evsel, &attr, idx); + return 0; +} + +static void pyrf_evsel__delete(struct pyrf_evsel *pevsel) +{ + perf_evsel__exit(&pevsel->evsel); + pevsel->ob_type->tp_free((PyObject*)pevsel); +} + +static PyObject *pyrf_evsel__open(struct pyrf_evsel *pevsel, + PyObject *args, PyObject *kwargs) +{ + struct perf_evsel *evsel = &pevsel->evsel; + struct cpu_map *cpus = NULL; + struct thread_map *threads = NULL; + PyObject *pcpus = NULL, *pthreads = NULL; + int group = 0, inherit = 0; + static char *kwlist[] = { "cpus", "threads", "group", "inherit", NULL }; + + if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|OOii", kwlist, + &pcpus, &pthreads, &group, &inherit)) + return NULL; + + if (pthreads != NULL) + threads = ((struct pyrf_thread_map *)pthreads)->threads; + + if (pcpus != NULL) + cpus = ((struct pyrf_cpu_map *)pcpus)->cpus; + + evsel->attr.inherit = inherit; + /* + * This will group just the fds for this single evsel, to group + * multiple events, use evlist.open(). + */ + if (perf_evsel__open(evsel, cpus, threads) < 0) { + PyErr_SetFromErrno(PyExc_OSError); + return NULL; + } + + Py_INCREF(Py_None); + return Py_None; +} + +static PyMethodDef pyrf_evsel__methods[] = { + { + .ml_name = "open", + .ml_meth = (PyCFunction)pyrf_evsel__open, + .ml_flags = METH_VARARGS | METH_KEYWORDS, + .ml_doc = PyDoc_STR("open the event selector file descriptor table.") + }, + { .ml_name = NULL, } +}; + +static char pyrf_evsel__doc[] = PyDoc_STR("perf event selector list object."); + +static PyTypeObject pyrf_evsel__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.evsel", + .tp_basicsize = sizeof(struct pyrf_evsel), + .tp_dealloc = (destructor)pyrf_evsel__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_doc = pyrf_evsel__doc, + .tp_methods = pyrf_evsel__methods, + .tp_init = (initproc)pyrf_evsel__init, +}; + +static int pyrf_evsel__setup_types(void) +{ + pyrf_evsel__type.tp_new = PyType_GenericNew; + return PyType_Ready(&pyrf_evsel__type); +} + +struct pyrf_evlist { + PyObject_HEAD + + struct perf_evlist evlist; +}; + +static int pyrf_evlist__init(struct pyrf_evlist *pevlist, + PyObject *args, PyObject *kwargs __maybe_unused) +{ + PyObject *pcpus = NULL, *pthreads = NULL; + struct cpu_map *cpus; + struct thread_map *threads; + + if (!PyArg_ParseTuple(args, "OO", &pcpus, &pthreads)) + return -1; + + threads = ((struct pyrf_thread_map *)pthreads)->threads; + cpus = ((struct pyrf_cpu_map *)pcpus)->cpus; + perf_evlist__init(&pevlist->evlist, cpus, threads); + return 0; +} + +static void pyrf_evlist__delete(struct pyrf_evlist *pevlist) +{ + perf_evlist__exit(&pevlist->evlist); + pevlist->ob_type->tp_free((PyObject*)pevlist); +} + +static PyObject *pyrf_evlist__mmap(struct pyrf_evlist *pevlist, + PyObject *args, PyObject *kwargs) +{ + struct perf_evlist *evlist = &pevlist->evlist; + static char *kwlist[] = { "pages", "overwrite", NULL }; + int pages = 128, overwrite = false; + + if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|ii", kwlist, + &pages, &overwrite)) + return NULL; + + if (perf_evlist__mmap(evlist, pages, overwrite) < 0) { + PyErr_SetFromErrno(PyExc_OSError); + return NULL; + } + + Py_INCREF(Py_None); + return Py_None; +} + +static PyObject *pyrf_evlist__poll(struct pyrf_evlist *pevlist, + PyObject *args, PyObject *kwargs) +{ + struct perf_evlist *evlist = &pevlist->evlist; + static char *kwlist[] = { "timeout", NULL }; + int timeout = -1, n; + + if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|i", kwlist, &timeout)) + return NULL; + + n = poll(evlist->pollfd, evlist->nr_fds, timeout); + if (n < 0) { + PyErr_SetFromErrno(PyExc_OSError); + return NULL; + } + + return Py_BuildValue("i", n); +} + +static PyObject *pyrf_evlist__get_pollfd(struct pyrf_evlist *pevlist, + PyObject *args __maybe_unused, + PyObject *kwargs __maybe_unused) +{ + struct perf_evlist *evlist = &pevlist->evlist; + PyObject *list = PyList_New(0); + int i; + + for (i = 0; i < evlist->nr_fds; ++i) { + PyObject *file; + FILE *fp = fdopen(evlist->pollfd[i].fd, "r"); + + if (fp == NULL) + goto free_list; + + file = PyFile_FromFile(fp, "perf", "r", NULL); + if (file == NULL) + goto free_list; + + if (PyList_Append(list, file) != 0) { + Py_DECREF(file); + goto free_list; + } + + Py_DECREF(file); + } + + return list; +free_list: + return PyErr_NoMemory(); +} + + +static PyObject *pyrf_evlist__add(struct pyrf_evlist *pevlist, + PyObject *args, + PyObject *kwargs __maybe_unused) +{ + struct perf_evlist *evlist = &pevlist->evlist; + PyObject *pevsel; + struct perf_evsel *evsel; + + if (!PyArg_ParseTuple(args, "O", &pevsel)) + return NULL; + + Py_INCREF(pevsel); + evsel = &((struct pyrf_evsel *)pevsel)->evsel; + evsel->idx = evlist->nr_entries; + perf_evlist__add(evlist, evsel); + + return Py_BuildValue("i", evlist->nr_entries); +} + +static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist, + PyObject *args, PyObject *kwargs) +{ + struct perf_evlist *evlist = &pevlist->evlist; + union perf_event *event; + int sample_id_all = 1, cpu; + static char *kwlist[] = { "cpu", "sample_id_all", NULL }; + int err; + + if (!PyArg_ParseTupleAndKeywords(args, kwargs, "i|i", kwlist, + &cpu, &sample_id_all)) + return NULL; + + event = perf_evlist__mmap_read(evlist, cpu); + if (event != NULL) { + PyObject *pyevent = pyrf_event__new(event); + struct pyrf_event *pevent = (struct pyrf_event *)pyevent; + + if (pyevent == NULL) + return PyErr_NoMemory(); + + err = perf_evlist__parse_sample(evlist, event, &pevent->sample); + if (err) + return PyErr_Format(PyExc_OSError, + "perf: can't parse sample, err=%d", err); + return pyevent; + } + + Py_INCREF(Py_None); + return Py_None; +} + +static PyObject *pyrf_evlist__open(struct pyrf_evlist *pevlist, + PyObject *args, PyObject *kwargs) +{ + struct perf_evlist *evlist = &pevlist->evlist; + int group = 0; + static char *kwlist[] = { "group", NULL }; + + if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|OOii", kwlist, &group)) + return NULL; + + if (group) + perf_evlist__set_leader(evlist); + + if (perf_evlist__open(evlist) < 0) { + PyErr_SetFromErrno(PyExc_OSError); + return NULL; + } + + Py_INCREF(Py_None); + return Py_None; +} + +static PyMethodDef pyrf_evlist__methods[] = { + { + .ml_name = "mmap", + .ml_meth = (PyCFunction)pyrf_evlist__mmap, + .ml_flags = METH_VARARGS | METH_KEYWORDS, + .ml_doc = PyDoc_STR("mmap the file descriptor table.") + }, + { + .ml_name = "open", + .ml_meth = (PyCFunction)pyrf_evlist__open, + .ml_flags = METH_VARARGS | METH_KEYWORDS, + .ml_doc = PyDoc_STR("open the file descriptors.") + }, + { + .ml_name = "poll", + .ml_meth = (PyCFunction)pyrf_evlist__poll, + .ml_flags = METH_VARARGS | METH_KEYWORDS, + .ml_doc = PyDoc_STR("poll the file descriptor table.") + }, + { + .ml_name = "get_pollfd", + .ml_meth = (PyCFunction)pyrf_evlist__get_pollfd, + .ml_flags = METH_VARARGS | METH_KEYWORDS, + .ml_doc = PyDoc_STR("get the poll file descriptor table.") + }, + { + .ml_name = "add", + .ml_meth = (PyCFunction)pyrf_evlist__add, + .ml_flags = METH_VARARGS | METH_KEYWORDS, + .ml_doc = PyDoc_STR("adds an event selector to the list.") + }, + { + .ml_name = "read_on_cpu", + .ml_meth = (PyCFunction)pyrf_evlist__read_on_cpu, + .ml_flags = METH_VARARGS | METH_KEYWORDS, + .ml_doc = PyDoc_STR("reads an event.") + }, + { .ml_name = NULL, } +}; + +static Py_ssize_t pyrf_evlist__length(PyObject *obj) +{ + struct pyrf_evlist *pevlist = (void *)obj; + + return pevlist->evlist.nr_entries; +} + +static PyObject *pyrf_evlist__item(PyObject *obj, Py_ssize_t i) +{ + struct pyrf_evlist *pevlist = (void *)obj; + struct perf_evsel *pos; + + if (i >= pevlist->evlist.nr_entries) + return NULL; + + list_for_each_entry(pos, &pevlist->evlist.entries, node) + if (i-- == 0) + break; + + return Py_BuildValue("O", container_of(pos, struct pyrf_evsel, evsel)); +} + +static PySequenceMethods pyrf_evlist__sequence_methods = { + .sq_length = pyrf_evlist__length, + .sq_item = pyrf_evlist__item, +}; + +static char pyrf_evlist__doc[] = PyDoc_STR("perf event selector list object."); + +static PyTypeObject pyrf_evlist__type = { + PyVarObject_HEAD_INIT(NULL, 0) + .tp_name = "perf.evlist", + .tp_basicsize = sizeof(struct pyrf_evlist), + .tp_dealloc = (destructor)pyrf_evlist__delete, + .tp_flags = Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, + .tp_as_sequence = &pyrf_evlist__sequence_methods, + .tp_doc = pyrf_evlist__doc, + .tp_methods = pyrf_evlist__methods, + .tp_init = (initproc)pyrf_evlist__init, +}; + +static int pyrf_evlist__setup_types(void) +{ + pyrf_evlist__type.tp_new = PyType_GenericNew; + return PyType_Ready(&pyrf_evlist__type); +} + +static struct { + const char *name; + int value; +} perf__constants[] = { + { "TYPE_HARDWARE", PERF_TYPE_HARDWARE }, + { "TYPE_SOFTWARE", PERF_TYPE_SOFTWARE }, + { "TYPE_TRACEPOINT", PERF_TYPE_TRACEPOINT }, + { "TYPE_HW_CACHE", PERF_TYPE_HW_CACHE }, + { "TYPE_RAW", PERF_TYPE_RAW }, + { "TYPE_BREAKPOINT", PERF_TYPE_BREAKPOINT }, + + { "COUNT_HW_CPU_CYCLES", PERF_COUNT_HW_CPU_CYCLES }, + { "COUNT_HW_INSTRUCTIONS", PERF_COUNT_HW_INSTRUCTIONS }, + { "COUNT_HW_CACHE_REFERENCES", PERF_COUNT_HW_CACHE_REFERENCES }, + { "COUNT_HW_CACHE_MISSES", PERF_COUNT_HW_CACHE_MISSES }, + { "COUNT_HW_BRANCH_INSTRUCTIONS", PERF_COUNT_HW_BRANCH_INSTRUCTIONS }, + { "COUNT_HW_BRANCH_MISSES", PERF_COUNT_HW_BRANCH_MISSES }, + { "COUNT_HW_BUS_CYCLES", PERF_COUNT_HW_BUS_CYCLES }, + { "COUNT_HW_CACHE_L1D", PERF_COUNT_HW_CACHE_L1D }, + { "COUNT_HW_CACHE_L1I", PERF_COUNT_HW_CACHE_L1I }, + { "COUNT_HW_CACHE_LL", PERF_COUNT_HW_CACHE_LL }, + { "COUNT_HW_CACHE_DTLB", PERF_COUNT_HW_CACHE_DTLB }, + { "COUNT_HW_CACHE_ITLB", PERF_COUNT_HW_CACHE_ITLB }, + { "COUNT_HW_CACHE_BPU", PERF_COUNT_HW_CACHE_BPU }, + { "COUNT_HW_CACHE_OP_READ", PERF_COUNT_HW_CACHE_OP_READ }, + { "COUNT_HW_CACHE_OP_WRITE", PERF_COUNT_HW_CACHE_OP_WRITE }, + { "COUNT_HW_CACHE_OP_PREFETCH", PERF_COUNT_HW_CACHE_OP_PREFETCH }, + { "COUNT_HW_CACHE_RESULT_ACCESS", PERF_COUNT_HW_CACHE_RESULT_ACCESS }, + { "COUNT_HW_CACHE_RESULT_MISS", PERF_COUNT_HW_CACHE_RESULT_MISS }, + + { "COUNT_HW_STALLED_CYCLES_FRONTEND", PERF_COUNT_HW_STALLED_CYCLES_FRONTEND }, + { "COUNT_HW_STALLED_CYCLES_BACKEND", PERF_COUNT_HW_STALLED_CYCLES_BACKEND }, + + { "COUNT_SW_CPU_CLOCK", PERF_COUNT_SW_CPU_CLOCK }, + { "COUNT_SW_TASK_CLOCK", PERF_COUNT_SW_TASK_CLOCK }, + { "COUNT_SW_PAGE_FAULTS", PERF_COUNT_SW_PAGE_FAULTS }, + { "COUNT_SW_CONTEXT_SWITCHES", PERF_COUNT_SW_CONTEXT_SWITCHES }, + { "COUNT_SW_CPU_MIGRATIONS", PERF_COUNT_SW_CPU_MIGRATIONS }, + { "COUNT_SW_PAGE_FAULTS_MIN", PERF_COUNT_SW_PAGE_FAULTS_MIN }, + { "COUNT_SW_PAGE_FAULTS_MAJ", PERF_COUNT_SW_PAGE_FAULTS_MAJ }, + { "COUNT_SW_ALIGNMENT_FAULTS", PERF_COUNT_SW_ALIGNMENT_FAULTS }, + { "COUNT_SW_EMULATION_FAULTS", PERF_COUNT_SW_EMULATION_FAULTS }, + + { "SAMPLE_IP", PERF_SAMPLE_IP }, + { "SAMPLE_TID", PERF_SAMPLE_TID }, + { "SAMPLE_TIME", PERF_SAMPLE_TIME }, + { "SAMPLE_ADDR", PERF_SAMPLE_ADDR }, + { "SAMPLE_READ", PERF_SAMPLE_READ }, + { "SAMPLE_CALLCHAIN", PERF_SAMPLE_CALLCHAIN }, + { "SAMPLE_ID", PERF_SAMPLE_ID }, + { "SAMPLE_CPU", PERF_SAMPLE_CPU }, + { "SAMPLE_PERIOD", PERF_SAMPLE_PERIOD }, + { "SAMPLE_STREAM_ID", PERF_SAMPLE_STREAM_ID }, + { "SAMPLE_RAW", PERF_SAMPLE_RAW }, + + { "FORMAT_TOTAL_TIME_ENABLED", PERF_FORMAT_TOTAL_TIME_ENABLED }, + { "FORMAT_TOTAL_TIME_RUNNING", PERF_FORMAT_TOTAL_TIME_RUNNING }, + { "FORMAT_ID", PERF_FORMAT_ID }, + { "FORMAT_GROUP", PERF_FORMAT_GROUP }, + + { "RECORD_MMAP", PERF_RECORD_MMAP }, + { "RECORD_LOST", PERF_RECORD_LOST }, + { "RECORD_COMM", PERF_RECORD_COMM }, + { "RECORD_EXIT", PERF_RECORD_EXIT }, + { "RECORD_THROTTLE", PERF_RECORD_THROTTLE }, + { "RECORD_UNTHROTTLE", PERF_RECORD_UNTHROTTLE }, + { "RECORD_FORK", PERF_RECORD_FORK }, + { "RECORD_READ", PERF_RECORD_READ }, + { "RECORD_SAMPLE", PERF_RECORD_SAMPLE }, + { .name = NULL, }, +}; + +static PyMethodDef perf__methods[] = { + { .ml_name = NULL, } +}; + +PyMODINIT_FUNC initperf(void) +{ + PyObject *obj; + int i; + PyObject *dict, *module = Py_InitModule("perf", perf__methods); + + if (module == NULL || + pyrf_event__setup_types() < 0 || + pyrf_evlist__setup_types() < 0 || + pyrf_evsel__setup_types() < 0 || + pyrf_thread_map__setup_types() < 0 || + pyrf_cpu_map__setup_types() < 0) + return; + + page_size = sysconf(_SC_PAGE_SIZE); + + Py_INCREF(&pyrf_evlist__type); + PyModule_AddObject(module, "evlist", (PyObject*)&pyrf_evlist__type); + + Py_INCREF(&pyrf_evsel__type); + PyModule_AddObject(module, "evsel", (PyObject*)&pyrf_evsel__type); + + Py_INCREF(&pyrf_thread_map__type); + PyModule_AddObject(module, "thread_map", (PyObject*)&pyrf_thread_map__type); + + Py_INCREF(&pyrf_cpu_map__type); + PyModule_AddObject(module, "cpu_map", (PyObject*)&pyrf_cpu_map__type); + + dict = PyModule_GetDict(module); + if (dict == NULL) + goto error; + + for (i = 0; perf__constants[i].name != NULL; i++) { + obj = PyInt_FromLong(perf__constants[i].value); + if (obj == NULL) + goto error; + PyDict_SetItemString(dict, perf__constants[i].name, obj); + Py_DECREF(obj); + } + +error: + if (PyErr_Occurred()) + PyErr_SetString(PyExc_ImportError, "perf: Init failed!"); +} + +/* + * Dummy, to avoid dragging all the test_attr infrastructure in the python + * binding. + */ +void test_attr__open(struct perf_event_attr *attr, pid_t pid, int cpu, + int fd, int group_fd, unsigned long flags) +{ +} diff --git a/tools/perf/util/quote.c b/tools/perf/util/quote.c new file mode 100644 index 00000000..01f03242 --- /dev/null +++ b/tools/perf/util/quote.c @@ -0,0 +1,54 @@ +#include "cache.h" +#include "quote.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 + * + * 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 == '!'); +} + +static 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_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"); + } +} diff --git a/tools/perf/util/quote.h b/tools/perf/util/quote.h new file mode 100644 index 00000000..172889ea --- /dev/null +++ b/tools/perf/util/quote.h @@ -0,0 +1,29 @@ +#ifndef __PERF_QUOTE_H +#define __PERF_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. + */ + +extern void sq_quote_argv(struct strbuf *, const char **argv, size_t maxlen); + +#endif /* __PERF_QUOTE_H */ diff --git a/tools/perf/util/rblist.c b/tools/perf/util/rblist.c new file mode 100644 index 00000000..a16cdd26 --- /dev/null +++ b/tools/perf/util/rblist.c @@ -0,0 +1,107 @@ +/* + * Based on strlist.c by: + * (c) 2009 Arnaldo Carvalho de Melo <acme@redhat.com> + * + * Licensed under the GPLv2. + */ + +#include <errno.h> +#include <stdio.h> +#include <stdlib.h> + +#include "rblist.h" + +int rblist__add_node(struct rblist *rblist, const void *new_entry) +{ + struct rb_node **p = &rblist->entries.rb_node; + struct rb_node *parent = NULL, *new_node; + + while (*p != NULL) { + int rc; + + parent = *p; + + rc = rblist->node_cmp(parent, new_entry); + if (rc > 0) + p = &(*p)->rb_left; + else if (rc < 0) + p = &(*p)->rb_right; + else + return -EEXIST; + } + + new_node = rblist->node_new(rblist, new_entry); + if (new_node == NULL) + return -ENOMEM; + + rb_link_node(new_node, parent, p); + rb_insert_color(new_node, &rblist->entries); + ++rblist->nr_entries; + + return 0; +} + +void rblist__remove_node(struct rblist *rblist, struct rb_node *rb_node) +{ + rb_erase(rb_node, &rblist->entries); + --rblist->nr_entries; + rblist->node_delete(rblist, rb_node); +} + +struct rb_node *rblist__find(struct rblist *rblist, const void *entry) +{ + struct rb_node **p = &rblist->entries.rb_node; + struct rb_node *parent = NULL; + + while (*p != NULL) { + int rc; + + parent = *p; + + rc = rblist->node_cmp(parent, entry); + if (rc > 0) + p = &(*p)->rb_left; + else if (rc < 0) + p = &(*p)->rb_right; + else + return parent; + } + + return NULL; +} + +void rblist__init(struct rblist *rblist) +{ + if (rblist != NULL) { + rblist->entries = RB_ROOT; + rblist->nr_entries = 0; + } + + return; +} + +void rblist__delete(struct rblist *rblist) +{ + if (rblist != NULL) { + struct rb_node *pos, *next = rb_first(&rblist->entries); + + while (next) { + pos = next; + next = rb_next(pos); + rblist__remove_node(rblist, pos); + } + free(rblist); + } +} + +struct rb_node *rblist__entry(const struct rblist *rblist, unsigned int idx) +{ + struct rb_node *node; + + for (node = rb_first(&rblist->entries); node; node = rb_next(node)) { + if (!idx--) + return node; + } + + return NULL; +} diff --git a/tools/perf/util/rblist.h b/tools/perf/util/rblist.h new file mode 100644 index 00000000..6d0cae5a --- /dev/null +++ b/tools/perf/util/rblist.h @@ -0,0 +1,47 @@ +#ifndef __PERF_RBLIST_H +#define __PERF_RBLIST_H + +#include <linux/rbtree.h> +#include <stdbool.h> + +/* + * create node structs of the form: + * struct my_node { + * struct rb_node rb_node; + * ... my data ... + * }; + * + * create list structs of the form: + * struct mylist { + * struct rblist rblist; + * ... my data ... + * }; + */ + +struct rblist { + struct rb_root entries; + unsigned int nr_entries; + + int (*node_cmp)(struct rb_node *rbn, const void *entry); + struct rb_node *(*node_new)(struct rblist *rlist, const void *new_entry); + void (*node_delete)(struct rblist *rblist, struct rb_node *rb_node); +}; + +void rblist__init(struct rblist *rblist); +void rblist__delete(struct rblist *rblist); +int rblist__add_node(struct rblist *rblist, const void *new_entry); +void rblist__remove_node(struct rblist *rblist, struct rb_node *rb_node); +struct rb_node *rblist__find(struct rblist *rblist, const void *entry); +struct rb_node *rblist__entry(const struct rblist *rblist, unsigned int idx); + +static inline bool rblist__empty(const struct rblist *rblist) +{ + return rblist->nr_entries == 0; +} + +static inline unsigned int rblist__nr_entries(const struct rblist *rblist) +{ + return rblist->nr_entries; +} + +#endif /* __PERF_RBLIST_H */ diff --git a/tools/perf/util/run-command.c b/tools/perf/util/run-command.c new file mode 100644 index 00000000..da8e9b28 --- /dev/null +++ b/tools/perf/util/run-command.c @@ -0,0 +1,214 @@ +#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]; + } + + 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); + } + + 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); +} diff --git a/tools/perf/util/run-command.h b/tools/perf/util/run-command.h new file mode 100644 index 00000000..1ef264d5 --- /dev/null +++ b/tools/perf/util/run-command.h @@ -0,0 +1,58 @@ +#ifndef __PERF_RUN_COMMAND_H +#define __PERF_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 *); + +#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); + +#endif /* __PERF_RUN_COMMAND_H */ diff --git a/tools/perf/util/scripting-engines/trace-event-perl.c b/tools/perf/util/scripting-engines/trace-event-perl.c new file mode 100644 index 00000000..eacec859 --- /dev/null +++ b/tools/perf/util/scripting-engines/trace-event-perl.c @@ -0,0 +1,643 @@ +/* + * trace-event-perl. Feed perf script events to an embedded Perl interpreter. + * + * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <ctype.h> +#include <errno.h> + +#include "../util.h" +#include <EXTERN.h> +#include <perl.h> + +#include "../../perf.h" +#include "../thread.h" +#include "../event.h" +#include "../trace-event.h" +#include "../evsel.h" + +void boot_Perf__Trace__Context(pTHX_ CV *cv); +void boot_DynaLoader(pTHX_ CV *cv); +typedef PerlInterpreter * INTERP; + +void xs_init(pTHX); + +void xs_init(pTHX) +{ + const char *file = __FILE__; + dXSUB_SYS; + + newXS("Perf::Trace::Context::bootstrap", boot_Perf__Trace__Context, + file); + newXS("DynaLoader::boot_DynaLoader", boot_DynaLoader, file); +} + +INTERP my_perl; + +#define FTRACE_MAX_EVENT \ + ((1 << (sizeof(unsigned short) * 8)) - 1) + +struct event_format *events[FTRACE_MAX_EVENT]; + +extern struct scripting_context *scripting_context; + +static char *cur_field_name; +static int zero_flag_atom; + +static void define_symbolic_value(const char *ev_name, + const char *field_name, + const char *field_value, + const char *field_str) +{ + unsigned long long value; + dSP; + + value = eval_flag(field_value); + + ENTER; + SAVETMPS; + PUSHMARK(SP); + + XPUSHs(sv_2mortal(newSVpv(ev_name, 0))); + XPUSHs(sv_2mortal(newSVpv(field_name, 0))); + XPUSHs(sv_2mortal(newSVuv(value))); + XPUSHs(sv_2mortal(newSVpv(field_str, 0))); + + PUTBACK; + if (get_cv("main::define_symbolic_value", 0)) + call_pv("main::define_symbolic_value", G_SCALAR); + SPAGAIN; + PUTBACK; + FREETMPS; + LEAVE; +} + +static void define_symbolic_values(struct print_flag_sym *field, + const char *ev_name, + const char *field_name) +{ + define_symbolic_value(ev_name, field_name, field->value, field->str); + if (field->next) + define_symbolic_values(field->next, ev_name, field_name); +} + +static void define_symbolic_field(const char *ev_name, + const char *field_name) +{ + dSP; + + ENTER; + SAVETMPS; + PUSHMARK(SP); + + XPUSHs(sv_2mortal(newSVpv(ev_name, 0))); + XPUSHs(sv_2mortal(newSVpv(field_name, 0))); + + PUTBACK; + if (get_cv("main::define_symbolic_field", 0)) + call_pv("main::define_symbolic_field", G_SCALAR); + SPAGAIN; + PUTBACK; + FREETMPS; + LEAVE; +} + +static void define_flag_value(const char *ev_name, + const char *field_name, + const char *field_value, + const char *field_str) +{ + unsigned long long value; + dSP; + + value = eval_flag(field_value); + + ENTER; + SAVETMPS; + PUSHMARK(SP); + + XPUSHs(sv_2mortal(newSVpv(ev_name, 0))); + XPUSHs(sv_2mortal(newSVpv(field_name, 0))); + XPUSHs(sv_2mortal(newSVuv(value))); + XPUSHs(sv_2mortal(newSVpv(field_str, 0))); + + PUTBACK; + if (get_cv("main::define_flag_value", 0)) + call_pv("main::define_flag_value", G_SCALAR); + SPAGAIN; + PUTBACK; + FREETMPS; + LEAVE; +} + +static void define_flag_values(struct print_flag_sym *field, + const char *ev_name, + const char *field_name) +{ + define_flag_value(ev_name, field_name, field->value, field->str); + if (field->next) + define_flag_values(field->next, ev_name, field_name); +} + +static void define_flag_field(const char *ev_name, + const char *field_name, + const char *delim) +{ + dSP; + + ENTER; + SAVETMPS; + PUSHMARK(SP); + + XPUSHs(sv_2mortal(newSVpv(ev_name, 0))); + XPUSHs(sv_2mortal(newSVpv(field_name, 0))); + XPUSHs(sv_2mortal(newSVpv(delim, 0))); + + PUTBACK; + if (get_cv("main::define_flag_field", 0)) + call_pv("main::define_flag_field", G_SCALAR); + SPAGAIN; + PUTBACK; + FREETMPS; + LEAVE; +} + +static void define_event_symbols(struct event_format *event, + const char *ev_name, + struct print_arg *args) +{ + switch (args->type) { + case PRINT_NULL: + break; + case PRINT_ATOM: + define_flag_value(ev_name, cur_field_name, "0", + args->atom.atom); + zero_flag_atom = 0; + break; + case PRINT_FIELD: + if (cur_field_name) + free(cur_field_name); + cur_field_name = strdup(args->field.name); + break; + case PRINT_FLAGS: + define_event_symbols(event, ev_name, args->flags.field); + define_flag_field(ev_name, cur_field_name, args->flags.delim); + define_flag_values(args->flags.flags, ev_name, cur_field_name); + break; + case PRINT_SYMBOL: + define_event_symbols(event, ev_name, args->symbol.field); + define_symbolic_field(ev_name, cur_field_name); + define_symbolic_values(args->symbol.symbols, ev_name, + cur_field_name); + break; + case PRINT_HEX: + define_event_symbols(event, ev_name, args->hex.field); + define_event_symbols(event, ev_name, args->hex.size); + break; + case PRINT_BSTRING: + case PRINT_DYNAMIC_ARRAY: + case PRINT_STRING: + break; + case PRINT_TYPE: + define_event_symbols(event, ev_name, args->typecast.item); + break; + case PRINT_OP: + if (strcmp(args->op.op, ":") == 0) + zero_flag_atom = 1; + define_event_symbols(event, ev_name, args->op.left); + define_event_symbols(event, ev_name, args->op.right); + break; + case PRINT_FUNC: + default: + pr_err("Unsupported print arg type\n"); + /* we should warn... */ + return; + } + + if (args->next) + define_event_symbols(event, ev_name, args->next); +} + +static inline struct event_format *find_cache_event(struct perf_evsel *evsel) +{ + static char ev_name[256]; + struct event_format *event; + int type = evsel->attr.config; + + if (events[type]) + return events[type]; + + events[type] = event = evsel->tp_format; + if (!event) + return NULL; + + sprintf(ev_name, "%s::%s", event->system, event->name); + + define_event_symbols(event, ev_name, event->print_fmt.args); + + return event; +} + +static void perl_process_tracepoint(union perf_event *perf_event __maybe_unused, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine __maybe_unused, + struct addr_location *al) +{ + struct format_field *field; + static char handler[256]; + unsigned long long val; + unsigned long s, ns; + struct event_format *event; + int pid; + int cpu = sample->cpu; + void *data = sample->raw_data; + unsigned long long nsecs = sample->time; + struct thread *thread = al->thread; + char *comm = thread->comm; + + dSP; + + if (evsel->attr.type != PERF_TYPE_TRACEPOINT) + return; + + event = find_cache_event(evsel); + if (!event) + die("ug! no event found for type %" PRIu64, evsel->attr.config); + + pid = raw_field_value(event, "common_pid", data); + + sprintf(handler, "%s::%s", event->system, event->name); + + s = nsecs / NSECS_PER_SEC; + ns = nsecs - s * NSECS_PER_SEC; + + scripting_context->event_data = data; + scripting_context->pevent = evsel->tp_format->pevent; + + ENTER; + SAVETMPS; + PUSHMARK(SP); + + XPUSHs(sv_2mortal(newSVpv(handler, 0))); + XPUSHs(sv_2mortal(newSViv(PTR2IV(scripting_context)))); + XPUSHs(sv_2mortal(newSVuv(cpu))); + XPUSHs(sv_2mortal(newSVuv(s))); + XPUSHs(sv_2mortal(newSVuv(ns))); + XPUSHs(sv_2mortal(newSViv(pid))); + XPUSHs(sv_2mortal(newSVpv(comm, 0))); + + /* common fields other than pid can be accessed via xsub fns */ + + for (field = event->format.fields; field; field = field->next) { + if (field->flags & FIELD_IS_STRING) { + int offset; + if (field->flags & FIELD_IS_DYNAMIC) { + offset = *(int *)(data + field->offset); + offset &= 0xffff; + } else + offset = field->offset; + XPUSHs(sv_2mortal(newSVpv((char *)data + offset, 0))); + } else { /* FIELD_IS_NUMERIC */ + val = read_size(event, data + field->offset, + field->size); + if (field->flags & FIELD_IS_SIGNED) { + XPUSHs(sv_2mortal(newSViv(val))); + } else { + XPUSHs(sv_2mortal(newSVuv(val))); + } + } + } + + PUTBACK; + + if (get_cv(handler, 0)) + call_pv(handler, G_SCALAR); + else if (get_cv("main::trace_unhandled", 0)) { + XPUSHs(sv_2mortal(newSVpv(handler, 0))); + XPUSHs(sv_2mortal(newSViv(PTR2IV(scripting_context)))); + XPUSHs(sv_2mortal(newSVuv(cpu))); + XPUSHs(sv_2mortal(newSVuv(nsecs))); + XPUSHs(sv_2mortal(newSViv(pid))); + XPUSHs(sv_2mortal(newSVpv(comm, 0))); + call_pv("main::trace_unhandled", G_SCALAR); + } + SPAGAIN; + PUTBACK; + FREETMPS; + LEAVE; +} + +static void perl_process_event_generic(union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine __maybe_unused, + struct addr_location *al __maybe_unused) +{ + dSP; + + if (!get_cv("process_event", 0)) + return; + + ENTER; + SAVETMPS; + PUSHMARK(SP); + XPUSHs(sv_2mortal(newSVpvn((const char *)event, event->header.size))); + XPUSHs(sv_2mortal(newSVpvn((const char *)&evsel->attr, sizeof(evsel->attr)))); + XPUSHs(sv_2mortal(newSVpvn((const char *)sample, sizeof(*sample)))); + XPUSHs(sv_2mortal(newSVpvn((const char *)sample->raw_data, sample->raw_size))); + PUTBACK; + call_pv("process_event", G_SCALAR); + SPAGAIN; + PUTBACK; + FREETMPS; + LEAVE; +} + +static void perl_process_event(union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine, + struct addr_location *al) +{ + perl_process_tracepoint(event, sample, evsel, machine, al); + perl_process_event_generic(event, sample, evsel, machine, al); +} + +static void run_start_sub(void) +{ + dSP; /* access to Perl stack */ + PUSHMARK(SP); + + if (get_cv("main::trace_begin", 0)) + call_pv("main::trace_begin", G_DISCARD | G_NOARGS); +} + +/* + * Start trace script + */ +static int perl_start_script(const char *script, int argc, const char **argv) +{ + const char **command_line; + int i, err = 0; + + command_line = malloc((argc + 2) * sizeof(const char *)); + command_line[0] = ""; + command_line[1] = script; + for (i = 2; i < argc + 2; i++) + command_line[i] = argv[i - 2]; + + my_perl = perl_alloc(); + perl_construct(my_perl); + + if (perl_parse(my_perl, xs_init, argc + 2, (char **)command_line, + (char **)NULL)) { + err = -1; + goto error; + } + + if (perl_run(my_perl)) { + err = -1; + goto error; + } + + if (SvTRUE(ERRSV)) { + err = -1; + goto error; + } + + run_start_sub(); + + free(command_line); + return 0; +error: + perl_free(my_perl); + free(command_line); + + return err; +} + +/* + * Stop trace script + */ +static int perl_stop_script(void) +{ + dSP; /* access to Perl stack */ + PUSHMARK(SP); + + if (get_cv("main::trace_end", 0)) + call_pv("main::trace_end", G_DISCARD | G_NOARGS); + + perl_destruct(my_perl); + perl_free(my_perl); + + return 0; +} + +static int perl_generate_script(struct pevent *pevent, const char *outfile) +{ + struct event_format *event = NULL; + struct format_field *f; + char fname[PATH_MAX]; + int not_first, count; + FILE *ofp; + + sprintf(fname, "%s.pl", outfile); + ofp = fopen(fname, "w"); + if (ofp == NULL) { + fprintf(stderr, "couldn't open %s\n", fname); + return -1; + } + + fprintf(ofp, "# perf script event handlers, " + "generated by perf script -g perl\n"); + + fprintf(ofp, "# Licensed under the terms of the GNU GPL" + " License version 2\n\n"); + + fprintf(ofp, "# The common_* event handler fields are the most useful " + "fields common to\n"); + + fprintf(ofp, "# all events. They don't necessarily correspond to " + "the 'common_*' fields\n"); + + fprintf(ofp, "# in the format files. Those fields not available as " + "handler params can\n"); + + fprintf(ofp, "# be retrieved using Perl functions of the form " + "common_*($context).\n"); + + fprintf(ofp, "# See Context.pm for the list of available " + "functions.\n\n"); + + fprintf(ofp, "use lib \"$ENV{'PERF_EXEC_PATH'}/scripts/perl/" + "Perf-Trace-Util/lib\";\n"); + + fprintf(ofp, "use lib \"./Perf-Trace-Util/lib\";\n"); + fprintf(ofp, "use Perf::Trace::Core;\n"); + fprintf(ofp, "use Perf::Trace::Context;\n"); + fprintf(ofp, "use Perf::Trace::Util;\n\n"); + + fprintf(ofp, "sub trace_begin\n{\n\t# optional\n}\n\n"); + fprintf(ofp, "sub trace_end\n{\n\t# optional\n}\n\n"); + + while ((event = trace_find_next_event(pevent, event))) { + fprintf(ofp, "sub %s::%s\n{\n", event->system, event->name); + fprintf(ofp, "\tmy ("); + + fprintf(ofp, "$event_name, "); + fprintf(ofp, "$context, "); + fprintf(ofp, "$common_cpu, "); + fprintf(ofp, "$common_secs, "); + fprintf(ofp, "$common_nsecs,\n"); + fprintf(ofp, "\t $common_pid, "); + fprintf(ofp, "$common_comm,\n\t "); + + not_first = 0; + count = 0; + + for (f = event->format.fields; f; f = f->next) { + if (not_first++) + fprintf(ofp, ", "); + if (++count % 5 == 0) + fprintf(ofp, "\n\t "); + + fprintf(ofp, "$%s", f->name); + } + fprintf(ofp, ") = @_;\n\n"); + + fprintf(ofp, "\tprint_header($event_name, $common_cpu, " + "$common_secs, $common_nsecs,\n\t " + "$common_pid, $common_comm);\n\n"); + + fprintf(ofp, "\tprintf(\""); + + not_first = 0; + count = 0; + + for (f = event->format.fields; f; f = f->next) { + if (not_first++) + fprintf(ofp, ", "); + if (count && count % 4 == 0) { + fprintf(ofp, "\".\n\t \""); + } + count++; + + fprintf(ofp, "%s=", f->name); + if (f->flags & FIELD_IS_STRING || + f->flags & FIELD_IS_FLAG || + f->flags & FIELD_IS_SYMBOLIC) + fprintf(ofp, "%%s"); + else if (f->flags & FIELD_IS_SIGNED) + fprintf(ofp, "%%d"); + else + fprintf(ofp, "%%u"); + } + + fprintf(ofp, "\\n\",\n\t "); + + not_first = 0; + count = 0; + + for (f = event->format.fields; f; f = f->next) { + if (not_first++) + fprintf(ofp, ", "); + + if (++count % 5 == 0) + fprintf(ofp, "\n\t "); + + if (f->flags & FIELD_IS_FLAG) { + if ((count - 1) % 5 != 0) { + fprintf(ofp, "\n\t "); + count = 4; + } + fprintf(ofp, "flag_str(\""); + fprintf(ofp, "%s::%s\", ", event->system, + event->name); + fprintf(ofp, "\"%s\", $%s)", f->name, + f->name); + } else if (f->flags & FIELD_IS_SYMBOLIC) { + if ((count - 1) % 5 != 0) { + fprintf(ofp, "\n\t "); + count = 4; + } + fprintf(ofp, "symbol_str(\""); + fprintf(ofp, "%s::%s\", ", event->system, + event->name); + fprintf(ofp, "\"%s\", $%s)", f->name, + f->name); + } else + fprintf(ofp, "$%s", f->name); + } + + fprintf(ofp, ");\n"); + fprintf(ofp, "}\n\n"); + } + + fprintf(ofp, "sub trace_unhandled\n{\n\tmy ($event_name, $context, " + "$common_cpu, $common_secs, $common_nsecs,\n\t " + "$common_pid, $common_comm) = @_;\n\n"); + + fprintf(ofp, "\tprint_header($event_name, $common_cpu, " + "$common_secs, $common_nsecs,\n\t $common_pid, " + "$common_comm);\n}\n\n"); + + fprintf(ofp, "sub print_header\n{\n" + "\tmy ($event_name, $cpu, $secs, $nsecs, $pid, $comm) = @_;\n\n" + "\tprintf(\"%%-20s %%5u %%05u.%%09u %%8u %%-20s \",\n\t " + "$event_name, $cpu, $secs, $nsecs, $pid, $comm);\n}\n"); + + fprintf(ofp, + "\n# Packed byte string args of process_event():\n" + "#\n" + "# $event:\tunion perf_event\tutil/event.h\n" + "# $attr:\tstruct perf_event_attr\tlinux/perf_event.h\n" + "# $sample:\tstruct perf_sample\tutil/event.h\n" + "# $raw_data:\tperf_sample->raw_data\tutil/event.h\n" + "\n" + "sub process_event\n" + "{\n" + "\tmy ($event, $attr, $sample, $raw_data) = @_;\n" + "\n" + "\tmy @event\t= unpack(\"LSS\", $event);\n" + "\tmy @attr\t= unpack(\"LLQQQQQLLQQ\", $attr);\n" + "\tmy @sample\t= unpack(\"QLLQQQQQLL\", $sample);\n" + "\tmy @raw_data\t= unpack(\"C*\", $raw_data);\n" + "\n" + "\tuse Data::Dumper;\n" + "\tprint Dumper \\@event, \\@attr, \\@sample, \\@raw_data;\n" + "}\n"); + + fclose(ofp); + + fprintf(stderr, "generated Perl script: %s\n", fname); + + return 0; +} + +struct scripting_ops perl_scripting_ops = { + .name = "Perl", + .start_script = perl_start_script, + .stop_script = perl_stop_script, + .process_event = perl_process_event, + .generate_script = perl_generate_script, +}; diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c new file mode 100644 index 00000000..e87aa5d9 --- /dev/null +++ b/tools/perf/util/scripting-engines/trace-event-python.c @@ -0,0 +1,694 @@ +/* + * trace-event-python. Feed trace events to an embedded Python interpreter. + * + * Copyright (C) 2010 Tom Zanussi <tzanussi@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <Python.h> + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <errno.h> + +#include "../../perf.h" +#include "../evsel.h" +#include "../util.h" +#include "../event.h" +#include "../thread.h" +#include "../trace-event.h" + +PyMODINIT_FUNC initperf_trace_context(void); + +#define FTRACE_MAX_EVENT \ + ((1 << (sizeof(unsigned short) * 8)) - 1) + +struct event_format *events[FTRACE_MAX_EVENT]; + +#define MAX_FIELDS 64 +#define N_COMMON_FIELDS 7 + +extern struct scripting_context *scripting_context; + +static char *cur_field_name; +static int zero_flag_atom; + +static PyObject *main_module, *main_dict; + +static void handler_call_die(const char *handler_name) +{ + PyErr_Print(); + Py_FatalError("problem in Python trace event handler"); +} + +static void define_value(enum print_arg_type field_type, + const char *ev_name, + const char *field_name, + const char *field_value, + const char *field_str) +{ + const char *handler_name = "define_flag_value"; + PyObject *handler, *t, *retval; + unsigned long long value; + unsigned n = 0; + + if (field_type == PRINT_SYMBOL) + handler_name = "define_symbolic_value"; + + t = PyTuple_New(4); + if (!t) + Py_FatalError("couldn't create Python tuple"); + + value = eval_flag(field_value); + + PyTuple_SetItem(t, n++, PyString_FromString(ev_name)); + PyTuple_SetItem(t, n++, PyString_FromString(field_name)); + PyTuple_SetItem(t, n++, PyInt_FromLong(value)); + PyTuple_SetItem(t, n++, PyString_FromString(field_str)); + + handler = PyDict_GetItemString(main_dict, handler_name); + if (handler && PyCallable_Check(handler)) { + retval = PyObject_CallObject(handler, t); + if (retval == NULL) + handler_call_die(handler_name); + } + + Py_DECREF(t); +} + +static void define_values(enum print_arg_type field_type, + struct print_flag_sym *field, + const char *ev_name, + const char *field_name) +{ + define_value(field_type, ev_name, field_name, field->value, + field->str); + + if (field->next) + define_values(field_type, field->next, ev_name, field_name); +} + +static void define_field(enum print_arg_type field_type, + const char *ev_name, + const char *field_name, + const char *delim) +{ + const char *handler_name = "define_flag_field"; + PyObject *handler, *t, *retval; + unsigned n = 0; + + if (field_type == PRINT_SYMBOL) + handler_name = "define_symbolic_field"; + + if (field_type == PRINT_FLAGS) + t = PyTuple_New(3); + else + t = PyTuple_New(2); + if (!t) + Py_FatalError("couldn't create Python tuple"); + + PyTuple_SetItem(t, n++, PyString_FromString(ev_name)); + PyTuple_SetItem(t, n++, PyString_FromString(field_name)); + if (field_type == PRINT_FLAGS) + PyTuple_SetItem(t, n++, PyString_FromString(delim)); + + handler = PyDict_GetItemString(main_dict, handler_name); + if (handler && PyCallable_Check(handler)) { + retval = PyObject_CallObject(handler, t); + if (retval == NULL) + handler_call_die(handler_name); + } + + Py_DECREF(t); +} + +static void define_event_symbols(struct event_format *event, + const char *ev_name, + struct print_arg *args) +{ + switch (args->type) { + case PRINT_NULL: + break; + case PRINT_ATOM: + define_value(PRINT_FLAGS, ev_name, cur_field_name, "0", + args->atom.atom); + zero_flag_atom = 0; + break; + case PRINT_FIELD: + if (cur_field_name) + free(cur_field_name); + cur_field_name = strdup(args->field.name); + break; + case PRINT_FLAGS: + define_event_symbols(event, ev_name, args->flags.field); + define_field(PRINT_FLAGS, ev_name, cur_field_name, + args->flags.delim); + define_values(PRINT_FLAGS, args->flags.flags, ev_name, + cur_field_name); + break; + case PRINT_SYMBOL: + define_event_symbols(event, ev_name, args->symbol.field); + define_field(PRINT_SYMBOL, ev_name, cur_field_name, NULL); + define_values(PRINT_SYMBOL, args->symbol.symbols, ev_name, + cur_field_name); + break; + case PRINT_HEX: + define_event_symbols(event, ev_name, args->hex.field); + define_event_symbols(event, ev_name, args->hex.size); + break; + case PRINT_STRING: + break; + case PRINT_TYPE: + define_event_symbols(event, ev_name, args->typecast.item); + break; + case PRINT_OP: + if (strcmp(args->op.op, ":") == 0) + zero_flag_atom = 1; + define_event_symbols(event, ev_name, args->op.left); + define_event_symbols(event, ev_name, args->op.right); + break; + default: + /* gcc warns for these? */ + case PRINT_BSTRING: + case PRINT_DYNAMIC_ARRAY: + case PRINT_FUNC: + /* we should warn... */ + return; + } + + if (args->next) + define_event_symbols(event, ev_name, args->next); +} + +static inline struct event_format *find_cache_event(struct perf_evsel *evsel) +{ + static char ev_name[256]; + struct event_format *event; + int type = evsel->attr.config; + + /* + * XXX: Do we really need to cache this since now we have evsel->tp_format + * cached already? Need to re-read this "cache" routine that as well calls + * define_event_symbols() :-\ + */ + if (events[type]) + return events[type]; + + events[type] = event = evsel->tp_format; + if (!event) + return NULL; + + sprintf(ev_name, "%s__%s", event->system, event->name); + + define_event_symbols(event, ev_name, event->print_fmt.args); + + return event; +} + +static void python_process_tracepoint(union perf_event *perf_event + __maybe_unused, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine __maybe_unused, + struct addr_location *al) +{ + PyObject *handler, *retval, *context, *t, *obj, *dict = NULL; + static char handler_name[256]; + struct format_field *field; + unsigned long long val; + unsigned long s, ns; + struct event_format *event; + unsigned n = 0; + int pid; + int cpu = sample->cpu; + void *data = sample->raw_data; + unsigned long long nsecs = sample->time; + struct thread *thread = al->thread; + char *comm = thread->comm; + + t = PyTuple_New(MAX_FIELDS); + if (!t) + Py_FatalError("couldn't create Python tuple"); + + event = find_cache_event(evsel); + if (!event) + die("ug! no event found for type %d", (int)evsel->attr.config); + + pid = raw_field_value(event, "common_pid", data); + + sprintf(handler_name, "%s__%s", event->system, event->name); + + handler = PyDict_GetItemString(main_dict, handler_name); + if (handler && !PyCallable_Check(handler)) + handler = NULL; + if (!handler) { + dict = PyDict_New(); + if (!dict) + Py_FatalError("couldn't create Python dict"); + } + s = nsecs / NSECS_PER_SEC; + ns = nsecs - s * NSECS_PER_SEC; + + scripting_context->event_data = data; + scripting_context->pevent = evsel->tp_format->pevent; + + context = PyCObject_FromVoidPtr(scripting_context, NULL); + + PyTuple_SetItem(t, n++, PyString_FromString(handler_name)); + PyTuple_SetItem(t, n++, context); + + if (handler) { + PyTuple_SetItem(t, n++, PyInt_FromLong(cpu)); + PyTuple_SetItem(t, n++, PyInt_FromLong(s)); + PyTuple_SetItem(t, n++, PyInt_FromLong(ns)); + PyTuple_SetItem(t, n++, PyInt_FromLong(pid)); + PyTuple_SetItem(t, n++, PyString_FromString(comm)); + } else { + PyDict_SetItemString(dict, "common_cpu", PyInt_FromLong(cpu)); + PyDict_SetItemString(dict, "common_s", PyInt_FromLong(s)); + PyDict_SetItemString(dict, "common_ns", PyInt_FromLong(ns)); + PyDict_SetItemString(dict, "common_pid", PyInt_FromLong(pid)); + PyDict_SetItemString(dict, "common_comm", PyString_FromString(comm)); + } + for (field = event->format.fields; field; field = field->next) { + if (field->flags & FIELD_IS_STRING) { + int offset; + if (field->flags & FIELD_IS_DYNAMIC) { + offset = *(int *)(data + field->offset); + offset &= 0xffff; + } else + offset = field->offset; + obj = PyString_FromString((char *)data + offset); + } else { /* FIELD_IS_NUMERIC */ + val = read_size(event, data + field->offset, + field->size); + if (field->flags & FIELD_IS_SIGNED) { + if ((long long)val >= LONG_MIN && + (long long)val <= LONG_MAX) + obj = PyInt_FromLong(val); + else + obj = PyLong_FromLongLong(val); + } else { + if (val <= LONG_MAX) + obj = PyInt_FromLong(val); + else + obj = PyLong_FromUnsignedLongLong(val); + } + } + if (handler) + PyTuple_SetItem(t, n++, obj); + else + PyDict_SetItemString(dict, field->name, obj); + + } + if (!handler) + PyTuple_SetItem(t, n++, dict); + + if (_PyTuple_Resize(&t, n) == -1) + Py_FatalError("error resizing Python tuple"); + + if (handler) { + retval = PyObject_CallObject(handler, t); + if (retval == NULL) + handler_call_die(handler_name); + } else { + handler = PyDict_GetItemString(main_dict, "trace_unhandled"); + if (handler && PyCallable_Check(handler)) { + + retval = PyObject_CallObject(handler, t); + if (retval == NULL) + handler_call_die("trace_unhandled"); + } + Py_DECREF(dict); + } + + Py_DECREF(t); +} + +static void python_process_general_event(union perf_event *perf_event + __maybe_unused, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine __maybe_unused, + struct addr_location *al) +{ + PyObject *handler, *retval, *t, *dict; + static char handler_name[64]; + unsigned n = 0; + struct thread *thread = al->thread; + + /* + * Use the MAX_FIELDS to make the function expandable, though + * currently there is only one item for the tuple. + */ + t = PyTuple_New(MAX_FIELDS); + if (!t) + Py_FatalError("couldn't create Python tuple"); + + dict = PyDict_New(); + if (!dict) + Py_FatalError("couldn't create Python dictionary"); + + snprintf(handler_name, sizeof(handler_name), "%s", "process_event"); + + handler = PyDict_GetItemString(main_dict, handler_name); + if (!handler || !PyCallable_Check(handler)) + goto exit; + + PyDict_SetItemString(dict, "ev_name", PyString_FromString(perf_evsel__name(evsel))); + PyDict_SetItemString(dict, "attr", PyString_FromStringAndSize( + (const char *)&evsel->attr, sizeof(evsel->attr))); + PyDict_SetItemString(dict, "sample", PyString_FromStringAndSize( + (const char *)sample, sizeof(*sample))); + PyDict_SetItemString(dict, "raw_buf", PyString_FromStringAndSize( + (const char *)sample->raw_data, sample->raw_size)); + PyDict_SetItemString(dict, "comm", + PyString_FromString(thread->comm)); + if (al->map) { + PyDict_SetItemString(dict, "dso", + PyString_FromString(al->map->dso->name)); + } + if (al->sym) { + PyDict_SetItemString(dict, "symbol", + PyString_FromString(al->sym->name)); + } + + PyTuple_SetItem(t, n++, dict); + if (_PyTuple_Resize(&t, n) == -1) + Py_FatalError("error resizing Python tuple"); + + retval = PyObject_CallObject(handler, t); + if (retval == NULL) + handler_call_die(handler_name); +exit: + Py_DECREF(dict); + Py_DECREF(t); +} + +static void python_process_event(union perf_event *perf_event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine, + struct addr_location *al) +{ + switch (evsel->attr.type) { + case PERF_TYPE_TRACEPOINT: + python_process_tracepoint(perf_event, sample, evsel, + machine, al); + break; + /* Reserve for future process_hw/sw/raw APIs */ + default: + python_process_general_event(perf_event, sample, evsel, + machine, al); + } +} + +static int run_start_sub(void) +{ + PyObject *handler, *retval; + int err = 0; + + main_module = PyImport_AddModule("__main__"); + if (main_module == NULL) + return -1; + Py_INCREF(main_module); + + main_dict = PyModule_GetDict(main_module); + if (main_dict == NULL) { + err = -1; + goto error; + } + Py_INCREF(main_dict); + + handler = PyDict_GetItemString(main_dict, "trace_begin"); + if (handler == NULL || !PyCallable_Check(handler)) + goto out; + + retval = PyObject_CallObject(handler, NULL); + if (retval == NULL) + handler_call_die("trace_begin"); + + Py_DECREF(retval); + return err; +error: + Py_XDECREF(main_dict); + Py_XDECREF(main_module); +out: + return err; +} + +/* + * Start trace script + */ +static int python_start_script(const char *script, int argc, const char **argv) +{ + const char **command_line; + char buf[PATH_MAX]; + int i, err = 0; + FILE *fp; + + command_line = malloc((argc + 1) * sizeof(const char *)); + command_line[0] = script; + for (i = 1; i < argc + 1; i++) + command_line[i] = argv[i - 1]; + + Py_Initialize(); + + initperf_trace_context(); + + PySys_SetArgv(argc + 1, (char **)command_line); + + fp = fopen(script, "r"); + if (!fp) { + sprintf(buf, "Can't open python script \"%s\"", script); + perror(buf); + err = -1; + goto error; + } + + err = PyRun_SimpleFile(fp, script); + if (err) { + fprintf(stderr, "Error running python script %s\n", script); + goto error; + } + + err = run_start_sub(); + if (err) { + fprintf(stderr, "Error starting python script %s\n", script); + goto error; + } + + free(command_line); + + return err; +error: + Py_Finalize(); + free(command_line); + + return err; +} + +/* + * Stop trace script + */ +static int python_stop_script(void) +{ + PyObject *handler, *retval; + int err = 0; + + handler = PyDict_GetItemString(main_dict, "trace_end"); + if (handler == NULL || !PyCallable_Check(handler)) + goto out; + + retval = PyObject_CallObject(handler, NULL); + if (retval == NULL) + handler_call_die("trace_end"); + else + Py_DECREF(retval); +out: + Py_XDECREF(main_dict); + Py_XDECREF(main_module); + Py_Finalize(); + + return err; +} + +static int python_generate_script(struct pevent *pevent, const char *outfile) +{ + struct event_format *event = NULL; + struct format_field *f; + char fname[PATH_MAX]; + int not_first, count; + FILE *ofp; + + sprintf(fname, "%s.py", outfile); + ofp = fopen(fname, "w"); + if (ofp == NULL) { + fprintf(stderr, "couldn't open %s\n", fname); + return -1; + } + fprintf(ofp, "# perf script event handlers, " + "generated by perf script -g python\n"); + + fprintf(ofp, "# Licensed under the terms of the GNU GPL" + " License version 2\n\n"); + + fprintf(ofp, "# The common_* event handler fields are the most useful " + "fields common to\n"); + + fprintf(ofp, "# all events. They don't necessarily correspond to " + "the 'common_*' fields\n"); + + fprintf(ofp, "# in the format files. Those fields not available as " + "handler params can\n"); + + fprintf(ofp, "# be retrieved using Python functions of the form " + "common_*(context).\n"); + + fprintf(ofp, "# See the perf-trace-python Documentation for the list " + "of available functions.\n\n"); + + fprintf(ofp, "import os\n"); + fprintf(ofp, "import sys\n\n"); + + fprintf(ofp, "sys.path.append(os.environ['PERF_EXEC_PATH'] + \\\n"); + fprintf(ofp, "\t'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')\n"); + fprintf(ofp, "\nfrom perf_trace_context import *\n"); + fprintf(ofp, "from Core import *\n\n\n"); + + fprintf(ofp, "def trace_begin():\n"); + fprintf(ofp, "\tprint \"in trace_begin\"\n\n"); + + fprintf(ofp, "def trace_end():\n"); + fprintf(ofp, "\tprint \"in trace_end\"\n\n"); + + while ((event = trace_find_next_event(pevent, event))) { + fprintf(ofp, "def %s__%s(", event->system, event->name); + fprintf(ofp, "event_name, "); + fprintf(ofp, "context, "); + fprintf(ofp, "common_cpu,\n"); + fprintf(ofp, "\tcommon_secs, "); + fprintf(ofp, "common_nsecs, "); + fprintf(ofp, "common_pid, "); + fprintf(ofp, "common_comm,\n\t"); + + not_first = 0; + count = 0; + + for (f = event->format.fields; f; f = f->next) { + if (not_first++) + fprintf(ofp, ", "); + if (++count % 5 == 0) + fprintf(ofp, "\n\t"); + + fprintf(ofp, "%s", f->name); + } + fprintf(ofp, "):\n"); + + fprintf(ofp, "\t\tprint_header(event_name, common_cpu, " + "common_secs, common_nsecs,\n\t\t\t" + "common_pid, common_comm)\n\n"); + + fprintf(ofp, "\t\tprint \""); + + not_first = 0; + count = 0; + + for (f = event->format.fields; f; f = f->next) { + if (not_first++) + fprintf(ofp, ", "); + if (count && count % 3 == 0) { + fprintf(ofp, "\" \\\n\t\t\""); + } + count++; + + fprintf(ofp, "%s=", f->name); + if (f->flags & FIELD_IS_STRING || + f->flags & FIELD_IS_FLAG || + f->flags & FIELD_IS_SYMBOLIC) + fprintf(ofp, "%%s"); + else if (f->flags & FIELD_IS_SIGNED) + fprintf(ofp, "%%d"); + else + fprintf(ofp, "%%u"); + } + + fprintf(ofp, "\\n\" %% \\\n\t\t("); + + not_first = 0; + count = 0; + + for (f = event->format.fields; f; f = f->next) { + if (not_first++) + fprintf(ofp, ", "); + + if (++count % 5 == 0) + fprintf(ofp, "\n\t\t"); + + if (f->flags & FIELD_IS_FLAG) { + if ((count - 1) % 5 != 0) { + fprintf(ofp, "\n\t\t"); + count = 4; + } + fprintf(ofp, "flag_str(\""); + fprintf(ofp, "%s__%s\", ", event->system, + event->name); + fprintf(ofp, "\"%s\", %s)", f->name, + f->name); + } else if (f->flags & FIELD_IS_SYMBOLIC) { + if ((count - 1) % 5 != 0) { + fprintf(ofp, "\n\t\t"); + count = 4; + } + fprintf(ofp, "symbol_str(\""); + fprintf(ofp, "%s__%s\", ", event->system, + event->name); + fprintf(ofp, "\"%s\", %s)", f->name, + f->name); + } else + fprintf(ofp, "%s", f->name); + } + + fprintf(ofp, "),\n\n"); + } + + fprintf(ofp, "def trace_unhandled(event_name, context, " + "event_fields_dict):\n"); + + fprintf(ofp, "\t\tprint ' '.join(['%%s=%%s'%%(k,str(v))" + "for k,v in sorted(event_fields_dict.items())])\n\n"); + + fprintf(ofp, "def print_header(" + "event_name, cpu, secs, nsecs, pid, comm):\n" + "\tprint \"%%-20s %%5u %%05u.%%09u %%8u %%-20s \" %% \\\n\t" + "(event_name, cpu, secs, nsecs, pid, comm),\n"); + + fclose(ofp); + + fprintf(stderr, "generated Python script: %s\n", fname); + + return 0; +} + +struct scripting_ops python_scripting_ops = { + .name = "Python", + .start_script = python_start_script, + .stop_script = python_stop_script, + .process_event = python_process_event, + .generate_script = python_generate_script, +}; diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c new file mode 100644 index 00000000..bd85280b --- /dev/null +++ b/tools/perf/util/session.c @@ -0,0 +1,1569 @@ +#define _FILE_OFFSET_BITS 64 + +#include <linux/kernel.h> + +#include <byteswap.h> +#include <unistd.h> +#include <sys/types.h> +#include <sys/mman.h> + +#include "evlist.h" +#include "evsel.h" +#include "session.h" +#include "tool.h" +#include "sort.h" +#include "util.h" +#include "cpumap.h" +#include "event-parse.h" +#include "perf_regs.h" +#include "vdso.h" + +static int perf_session__open(struct perf_session *self, bool force) +{ + struct stat input_stat; + + if (!strcmp(self->filename, "-")) { + self->fd_pipe = true; + self->fd = STDIN_FILENO; + + if (perf_session__read_header(self, self->fd) < 0) + pr_err("incompatible file format (rerun with -v to learn more)"); + + return 0; + } + + self->fd = open(self->filename, O_RDONLY); + if (self->fd < 0) { + int err = errno; + + pr_err("failed to open %s: %s", self->filename, strerror(err)); + if (err == ENOENT && !strcmp(self->filename, "perf.data")) + pr_err(" (try 'perf record' first)"); + pr_err("\n"); + return -errno; + } + + if (fstat(self->fd, &input_stat) < 0) + goto out_close; + + if (!force && input_stat.st_uid && (input_stat.st_uid != geteuid())) { + pr_err("file %s not owned by current user or root\n", + self->filename); + goto out_close; + } + + if (!input_stat.st_size) { + pr_info("zero-sized file (%s), nothing to do!\n", + self->filename); + goto out_close; + } + + if (perf_session__read_header(self, self->fd) < 0) { + pr_err("incompatible file format (rerun with -v to learn more)"); + goto out_close; + } + + if (!perf_evlist__valid_sample_type(self->evlist)) { + pr_err("non matching sample_type"); + goto out_close; + } + + if (!perf_evlist__valid_sample_id_all(self->evlist)) { + pr_err("non matching sample_id_all"); + goto out_close; + } + + self->size = input_stat.st_size; + return 0; + +out_close: + close(self->fd); + self->fd = -1; + return -1; +} + +void perf_session__set_id_hdr_size(struct perf_session *session) +{ + u16 id_hdr_size = perf_evlist__id_hdr_size(session->evlist); + + machines__set_id_hdr_size(&session->machines, id_hdr_size); +} + +int perf_session__create_kernel_maps(struct perf_session *self) +{ + int ret = machine__create_kernel_maps(&self->machines.host); + + if (ret >= 0) + ret = machines__create_guest_kernel_maps(&self->machines); + return ret; +} + +static void perf_session__destroy_kernel_maps(struct perf_session *self) +{ + machines__destroy_kernel_maps(&self->machines); +} + +struct perf_session *perf_session__new(const char *filename, int mode, + bool force, bool repipe, + struct perf_tool *tool) +{ + struct perf_session *self; + struct stat st; + size_t len; + + if (!filename || !strlen(filename)) { + if (!fstat(STDIN_FILENO, &st) && S_ISFIFO(st.st_mode)) + filename = "-"; + else + filename = "perf.data"; + } + + len = strlen(filename); + self = zalloc(sizeof(*self) + len); + + if (self == NULL) + goto out; + + memcpy(self->filename, filename, len); + self->repipe = repipe; + INIT_LIST_HEAD(&self->ordered_samples.samples); + INIT_LIST_HEAD(&self->ordered_samples.sample_cache); + INIT_LIST_HEAD(&self->ordered_samples.to_free); + machines__init(&self->machines); + + if (mode == O_RDONLY) { + if (perf_session__open(self, force) < 0) + goto out_delete; + perf_session__set_id_hdr_size(self); + } else if (mode == O_WRONLY) { + /* + * In O_RDONLY mode this will be performed when reading the + * kernel MMAP event, in perf_event__process_mmap(). + */ + if (perf_session__create_kernel_maps(self) < 0) + goto out_delete; + } + + if (tool && tool->ordering_requires_timestamps && + tool->ordered_samples && !perf_evlist__sample_id_all(self->evlist)) { + dump_printf("WARNING: No sample_id_all support, falling back to unordered processing\n"); + tool->ordered_samples = false; + } + +out: + return self; +out_delete: + perf_session__delete(self); + return NULL; +} + +static void perf_session__delete_dead_threads(struct perf_session *session) +{ + machine__delete_dead_threads(&session->machines.host); +} + +static void perf_session__delete_threads(struct perf_session *session) +{ + machine__delete_threads(&session->machines.host); +} + +static void perf_session_env__delete(struct perf_session_env *env) +{ + free(env->hostname); + free(env->os_release); + free(env->version); + free(env->arch); + free(env->cpu_desc); + free(env->cpuid); + + free(env->cmdline); + free(env->sibling_cores); + free(env->sibling_threads); + free(env->numa_nodes); + free(env->pmu_mappings); +} + +void perf_session__delete(struct perf_session *self) +{ + perf_session__destroy_kernel_maps(self); + perf_session__delete_dead_threads(self); + perf_session__delete_threads(self); + perf_session_env__delete(&self->header.env); + machines__exit(&self->machines); + close(self->fd); + free(self); + vdso__exit(); +} + +static int process_event_synth_tracing_data_stub(union perf_event *event + __maybe_unused, + struct perf_session *session + __maybe_unused) +{ + dump_printf(": unhandled!\n"); + return 0; +} + +static int process_event_synth_attr_stub(union perf_event *event __maybe_unused, + struct perf_evlist **pevlist + __maybe_unused) +{ + dump_printf(": unhandled!\n"); + return 0; +} + +static int process_event_sample_stub(struct perf_tool *tool __maybe_unused, + union perf_event *event __maybe_unused, + struct perf_sample *sample __maybe_unused, + struct perf_evsel *evsel __maybe_unused, + struct machine *machine __maybe_unused) +{ + dump_printf(": unhandled!\n"); + return 0; +} + +static int process_event_stub(struct perf_tool *tool __maybe_unused, + union perf_event *event __maybe_unused, + struct perf_sample *sample __maybe_unused, + struct machine *machine __maybe_unused) +{ + dump_printf(": unhandled!\n"); + return 0; +} + +static int process_finished_round_stub(struct perf_tool *tool __maybe_unused, + union perf_event *event __maybe_unused, + struct perf_session *perf_session + __maybe_unused) +{ + dump_printf(": unhandled!\n"); + return 0; +} + +static int process_event_type_stub(struct perf_tool *tool __maybe_unused, + union perf_event *event __maybe_unused) +{ + dump_printf(": unhandled!\n"); + return 0; +} + +static int process_finished_round(struct perf_tool *tool, + union perf_event *event, + struct perf_session *session); + +static void perf_tool__fill_defaults(struct perf_tool *tool) +{ + if (tool->sample == NULL) + tool->sample = process_event_sample_stub; + if (tool->mmap == NULL) + tool->mmap = process_event_stub; + if (tool->comm == NULL) + tool->comm = process_event_stub; + if (tool->fork == NULL) + tool->fork = process_event_stub; + if (tool->exit == NULL) + tool->exit = process_event_stub; + if (tool->lost == NULL) + tool->lost = perf_event__process_lost; + if (tool->read == NULL) + tool->read = process_event_sample_stub; + if (tool->throttle == NULL) + tool->throttle = process_event_stub; + if (tool->unthrottle == NULL) + tool->unthrottle = process_event_stub; + if (tool->attr == NULL) + tool->attr = process_event_synth_attr_stub; + if (tool->event_type == NULL) + tool->event_type = process_event_type_stub; + if (tool->tracing_data == NULL) + tool->tracing_data = process_event_synth_tracing_data_stub; + if (tool->build_id == NULL) + tool->build_id = process_finished_round_stub; + if (tool->finished_round == NULL) { + if (tool->ordered_samples) + tool->finished_round = process_finished_round; + else + tool->finished_round = process_finished_round_stub; + } +} + +void mem_bswap_32(void *src, int byte_size) +{ + u32 *m = src; + while (byte_size > 0) { + *m = bswap_32(*m); + byte_size -= sizeof(u32); + ++m; + } +} + +void mem_bswap_64(void *src, int byte_size) +{ + u64 *m = src; + + while (byte_size > 0) { + *m = bswap_64(*m); + byte_size -= sizeof(u64); + ++m; + } +} + +static void swap_sample_id_all(union perf_event *event, void *data) +{ + void *end = (void *) event + event->header.size; + int size = end - data; + + BUG_ON(size % sizeof(u64)); + mem_bswap_64(data, size); +} + +static void perf_event__all64_swap(union perf_event *event, + bool sample_id_all __maybe_unused) +{ + struct perf_event_header *hdr = &event->header; + mem_bswap_64(hdr + 1, event->header.size - sizeof(*hdr)); +} + +static void perf_event__comm_swap(union perf_event *event, bool sample_id_all) +{ + event->comm.pid = bswap_32(event->comm.pid); + event->comm.tid = bswap_32(event->comm.tid); + + if (sample_id_all) { + void *data = &event->comm.comm; + + data += PERF_ALIGN(strlen(data) + 1, sizeof(u64)); + swap_sample_id_all(event, data); + } +} + +static void perf_event__mmap_swap(union perf_event *event, + bool sample_id_all) +{ + event->mmap.pid = bswap_32(event->mmap.pid); + event->mmap.tid = bswap_32(event->mmap.tid); + event->mmap.start = bswap_64(event->mmap.start); + event->mmap.len = bswap_64(event->mmap.len); + event->mmap.pgoff = bswap_64(event->mmap.pgoff); + + if (sample_id_all) { + void *data = &event->mmap.filename; + + data += PERF_ALIGN(strlen(data) + 1, sizeof(u64)); + swap_sample_id_all(event, data); + } +} + +static void perf_event__task_swap(union perf_event *event, bool sample_id_all) +{ + event->fork.pid = bswap_32(event->fork.pid); + event->fork.tid = bswap_32(event->fork.tid); + event->fork.ppid = bswap_32(event->fork.ppid); + event->fork.ptid = bswap_32(event->fork.ptid); + event->fork.time = bswap_64(event->fork.time); + + if (sample_id_all) + swap_sample_id_all(event, &event->fork + 1); +} + +static void perf_event__read_swap(union perf_event *event, bool sample_id_all) +{ + event->read.pid = bswap_32(event->read.pid); + event->read.tid = bswap_32(event->read.tid); + event->read.value = bswap_64(event->read.value); + event->read.time_enabled = bswap_64(event->read.time_enabled); + event->read.time_running = bswap_64(event->read.time_running); + event->read.id = bswap_64(event->read.id); + + if (sample_id_all) + swap_sample_id_all(event, &event->read + 1); +} + +static u8 revbyte(u8 b) +{ + int rev = (b >> 4) | ((b & 0xf) << 4); + rev = ((rev & 0xcc) >> 2) | ((rev & 0x33) << 2); + rev = ((rev & 0xaa) >> 1) | ((rev & 0x55) << 1); + return (u8) rev; +} + +/* + * XXX this is hack in attempt to carry flags bitfield + * throught endian village. ABI says: + * + * Bit-fields are allocated from right to left (least to most significant) + * on little-endian implementations and from left to right (most to least + * significant) on big-endian implementations. + * + * The above seems to be byte specific, so we need to reverse each + * byte of the bitfield. 'Internet' also says this might be implementation + * specific and we probably need proper fix and carry perf_event_attr + * bitfield flags in separate data file FEAT_ section. Thought this seems + * to work for now. + */ +static void swap_bitfield(u8 *p, unsigned len) +{ + unsigned i; + + for (i = 0; i < len; i++) { + *p = revbyte(*p); + p++; + } +} + +/* exported for swapping attributes in file header */ +void perf_event__attr_swap(struct perf_event_attr *attr) +{ + attr->type = bswap_32(attr->type); + attr->size = bswap_32(attr->size); + attr->config = bswap_64(attr->config); + attr->sample_period = bswap_64(attr->sample_period); + attr->sample_type = bswap_64(attr->sample_type); + attr->read_format = bswap_64(attr->read_format); + attr->wakeup_events = bswap_32(attr->wakeup_events); + attr->bp_type = bswap_32(attr->bp_type); + attr->bp_addr = bswap_64(attr->bp_addr); + attr->bp_len = bswap_64(attr->bp_len); + + swap_bitfield((u8 *) (&attr->read_format + 1), sizeof(u64)); +} + +static void perf_event__hdr_attr_swap(union perf_event *event, + bool sample_id_all __maybe_unused) +{ + size_t size; + + perf_event__attr_swap(&event->attr.attr); + + size = event->header.size; + size -= (void *)&event->attr.id - (void *)event; + mem_bswap_64(event->attr.id, size); +} + +static void perf_event__event_type_swap(union perf_event *event, + bool sample_id_all __maybe_unused) +{ + event->event_type.event_type.event_id = + bswap_64(event->event_type.event_type.event_id); +} + +static void perf_event__tracing_data_swap(union perf_event *event, + bool sample_id_all __maybe_unused) +{ + event->tracing_data.size = bswap_32(event->tracing_data.size); +} + +typedef void (*perf_event__swap_op)(union perf_event *event, + bool sample_id_all); + +static perf_event__swap_op perf_event__swap_ops[] = { + [PERF_RECORD_MMAP] = perf_event__mmap_swap, + [PERF_RECORD_COMM] = perf_event__comm_swap, + [PERF_RECORD_FORK] = perf_event__task_swap, + [PERF_RECORD_EXIT] = perf_event__task_swap, + [PERF_RECORD_LOST] = perf_event__all64_swap, + [PERF_RECORD_READ] = perf_event__read_swap, + [PERF_RECORD_SAMPLE] = perf_event__all64_swap, + [PERF_RECORD_HEADER_ATTR] = perf_event__hdr_attr_swap, + [PERF_RECORD_HEADER_EVENT_TYPE] = perf_event__event_type_swap, + [PERF_RECORD_HEADER_TRACING_DATA] = perf_event__tracing_data_swap, + [PERF_RECORD_HEADER_BUILD_ID] = NULL, + [PERF_RECORD_HEADER_MAX] = NULL, +}; + +struct sample_queue { + u64 timestamp; + u64 file_offset; + union perf_event *event; + struct list_head list; +}; + +static void perf_session_free_sample_buffers(struct perf_session *session) +{ + struct ordered_samples *os = &session->ordered_samples; + + while (!list_empty(&os->to_free)) { + struct sample_queue *sq; + + sq = list_entry(os->to_free.next, struct sample_queue, list); + list_del(&sq->list); + free(sq); + } +} + +static int perf_session_deliver_event(struct perf_session *session, + union perf_event *event, + struct perf_sample *sample, + struct perf_tool *tool, + u64 file_offset); + +static int flush_sample_queue(struct perf_session *s, + struct perf_tool *tool) +{ + struct ordered_samples *os = &s->ordered_samples; + struct list_head *head = &os->samples; + struct sample_queue *tmp, *iter; + struct perf_sample sample; + u64 limit = os->next_flush; + u64 last_ts = os->last_sample ? os->last_sample->timestamp : 0ULL; + unsigned idx = 0, progress_next = os->nr_samples / 16; + int ret; + + if (!tool->ordered_samples || !limit) + return 0; + + list_for_each_entry_safe(iter, tmp, head, list) { + if (iter->timestamp > limit) + break; + + ret = perf_evlist__parse_sample(s->evlist, iter->event, &sample); + if (ret) + pr_err("Can't parse sample, err = %d\n", ret); + else { + ret = perf_session_deliver_event(s, iter->event, &sample, tool, + iter->file_offset); + if (ret) + return ret; + } + + os->last_flush = iter->timestamp; + list_del(&iter->list); + list_add(&iter->list, &os->sample_cache); + if (++idx >= progress_next) { + progress_next += os->nr_samples / 16; + ui_progress__update(idx, os->nr_samples, + "Processing time ordered events..."); + } + } + + if (list_empty(head)) { + os->last_sample = NULL; + } else if (last_ts <= limit) { + os->last_sample = + list_entry(head->prev, struct sample_queue, list); + } + + os->nr_samples = 0; + + return 0; +} + +/* + * When perf record finishes a pass on every buffers, it records this pseudo + * event. + * We record the max timestamp t found in the pass n. + * Assuming these timestamps are monotonic across cpus, we know that if + * a buffer still has events with timestamps below t, they will be all + * available and then read in the pass n + 1. + * Hence when we start to read the pass n + 2, we can safely flush every + * events with timestamps below t. + * + * ============ PASS n ================= + * CPU 0 | CPU 1 + * | + * cnt1 timestamps | cnt2 timestamps + * 1 | 2 + * 2 | 3 + * - | 4 <--- max recorded + * + * ============ PASS n + 1 ============== + * CPU 0 | CPU 1 + * | + * cnt1 timestamps | cnt2 timestamps + * 3 | 5 + * 4 | 6 + * 5 | 7 <---- max recorded + * + * Flush every events below timestamp 4 + * + * ============ PASS n + 2 ============== + * CPU 0 | CPU 1 + * | + * cnt1 timestamps | cnt2 timestamps + * 6 | 8 + * 7 | 9 + * - | 10 + * + * Flush every events below timestamp 7 + * etc... + */ +static int process_finished_round(struct perf_tool *tool, + union perf_event *event __maybe_unused, + struct perf_session *session) +{ + int ret = flush_sample_queue(session, tool); + if (!ret) + session->ordered_samples.next_flush = session->ordered_samples.max_timestamp; + + return ret; +} + +/* The queue is ordered by time */ +static void __queue_event(struct sample_queue *new, struct perf_session *s) +{ + struct ordered_samples *os = &s->ordered_samples; + struct sample_queue *sample = os->last_sample; + u64 timestamp = new->timestamp; + struct list_head *p; + + ++os->nr_samples; + os->last_sample = new; + + if (!sample) { + list_add(&new->list, &os->samples); + os->max_timestamp = timestamp; + return; + } + + /* + * last_sample might point to some random place in the list as it's + * the last queued event. We expect that the new event is close to + * this. + */ + if (sample->timestamp <= timestamp) { + while (sample->timestamp <= timestamp) { + p = sample->list.next; + if (p == &os->samples) { + list_add_tail(&new->list, &os->samples); + os->max_timestamp = timestamp; + return; + } + sample = list_entry(p, struct sample_queue, list); + } + list_add_tail(&new->list, &sample->list); + } else { + while (sample->timestamp > timestamp) { + p = sample->list.prev; + if (p == &os->samples) { + list_add(&new->list, &os->samples); + return; + } + sample = list_entry(p, struct sample_queue, list); + } + list_add(&new->list, &sample->list); + } +} + +#define MAX_SAMPLE_BUFFER (64 * 1024 / sizeof(struct sample_queue)) + +static int perf_session_queue_event(struct perf_session *s, union perf_event *event, + struct perf_sample *sample, u64 file_offset) +{ + struct ordered_samples *os = &s->ordered_samples; + struct list_head *sc = &os->sample_cache; + u64 timestamp = sample->time; + struct sample_queue *new; + + if (!timestamp || timestamp == ~0ULL) + return -ETIME; + + if (timestamp < s->ordered_samples.last_flush) { + printf("Warning: Timestamp below last timeslice flush\n"); + return -EINVAL; + } + + if (!list_empty(sc)) { + new = list_entry(sc->next, struct sample_queue, list); + list_del(&new->list); + } else if (os->sample_buffer) { + new = os->sample_buffer + os->sample_buffer_idx; + if (++os->sample_buffer_idx == MAX_SAMPLE_BUFFER) + os->sample_buffer = NULL; + } else { + os->sample_buffer = malloc(MAX_SAMPLE_BUFFER * sizeof(*new)); + if (!os->sample_buffer) + return -ENOMEM; + list_add(&os->sample_buffer->list, &os->to_free); + os->sample_buffer_idx = 2; + new = os->sample_buffer + 1; + } + + new->timestamp = timestamp; + new->file_offset = file_offset; + new->event = event; + + __queue_event(new, s); + + return 0; +} + +static void callchain__printf(struct perf_sample *sample) +{ + unsigned int i; + + printf("... chain: nr:%" PRIu64 "\n", sample->callchain->nr); + + for (i = 0; i < sample->callchain->nr; i++) + printf("..... %2d: %016" PRIx64 "\n", + i, sample->callchain->ips[i]); +} + +static void branch_stack__printf(struct perf_sample *sample) +{ + uint64_t i; + + printf("... branch stack: nr:%" PRIu64 "\n", sample->branch_stack->nr); + + for (i = 0; i < sample->branch_stack->nr; i++) + printf("..... %2"PRIu64": %016" PRIx64 " -> %016" PRIx64 "\n", + i, sample->branch_stack->entries[i].from, + sample->branch_stack->entries[i].to); +} + +static void regs_dump__printf(u64 mask, u64 *regs) +{ + unsigned rid, i = 0; + + for_each_set_bit(rid, (unsigned long *) &mask, sizeof(mask) * 8) { + u64 val = regs[i++]; + + printf(".... %-5s 0x%" PRIx64 "\n", + perf_reg_name(rid), val); + } +} + +static void regs_user__printf(struct perf_sample *sample, u64 mask) +{ + struct regs_dump *user_regs = &sample->user_regs; + + if (user_regs->regs) { + printf("... user regs: mask 0x%" PRIx64 "\n", mask); + regs_dump__printf(mask, user_regs->regs); + } +} + +static void stack_user__printf(struct stack_dump *dump) +{ + printf("... ustack: size %" PRIu64 ", offset 0x%x\n", + dump->size, dump->offset); +} + +static void perf_session__print_tstamp(struct perf_session *session, + union perf_event *event, + struct perf_sample *sample) +{ + u64 sample_type = perf_evlist__sample_type(session->evlist); + + if (event->header.type != PERF_RECORD_SAMPLE && + !perf_evlist__sample_id_all(session->evlist)) { + fputs("-1 -1 ", stdout); + return; + } + + if ((sample_type & PERF_SAMPLE_CPU)) + printf("%u ", sample->cpu); + + if (sample_type & PERF_SAMPLE_TIME) + printf("%" PRIu64 " ", sample->time); +} + +static void dump_event(struct perf_session *session, union perf_event *event, + u64 file_offset, struct perf_sample *sample) +{ + if (!dump_trace) + return; + + printf("\n%#" PRIx64 " [%#x]: event: %d\n", + file_offset, event->header.size, event->header.type); + + trace_event(event); + + if (sample) + perf_session__print_tstamp(session, event, sample); + + printf("%#" PRIx64 " [%#x]: PERF_RECORD_%s", file_offset, + event->header.size, perf_event__name(event->header.type)); +} + +static void dump_sample(struct perf_evsel *evsel, union perf_event *event, + struct perf_sample *sample) +{ + u64 sample_type; + + if (!dump_trace) + return; + + printf("(IP, %d): %d/%d: %#" PRIx64 " period: %" PRIu64 " addr: %#" PRIx64 "\n", + event->header.misc, sample->pid, sample->tid, sample->ip, + sample->period, sample->addr); + + sample_type = evsel->attr.sample_type; + + if (sample_type & PERF_SAMPLE_CALLCHAIN) + callchain__printf(sample); + + if (sample_type & PERF_SAMPLE_BRANCH_STACK) + branch_stack__printf(sample); + + if (sample_type & PERF_SAMPLE_REGS_USER) + regs_user__printf(sample, evsel->attr.sample_regs_user); + + if (sample_type & PERF_SAMPLE_STACK_USER) + stack_user__printf(&sample->user_stack); +} + +static struct machine * + perf_session__find_machine_for_cpumode(struct perf_session *session, + union perf_event *event) +{ + const u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; + + if (perf_guest && + ((cpumode == PERF_RECORD_MISC_GUEST_KERNEL) || + (cpumode == PERF_RECORD_MISC_GUEST_USER))) { + u32 pid; + + if (event->header.type == PERF_RECORD_MMAP) + pid = event->mmap.pid; + else + pid = event->ip.pid; + + return perf_session__findnew_machine(session, pid); + } + + return &session->machines.host; +} + +static int perf_session_deliver_event(struct perf_session *session, + union perf_event *event, + struct perf_sample *sample, + struct perf_tool *tool, + u64 file_offset) +{ + struct perf_evsel *evsel; + struct machine *machine; + + dump_event(session, event, file_offset, sample); + + evsel = perf_evlist__id2evsel(session->evlist, sample->id); + if (evsel != NULL && event->header.type != PERF_RECORD_SAMPLE) { + /* + * XXX We're leaving PERF_RECORD_SAMPLE unnacounted here + * because the tools right now may apply filters, discarding + * some of the samples. For consistency, in the future we + * should have something like nr_filtered_samples and remove + * the sample->period from total_sample_period, etc, KISS for + * now tho. + * + * Also testing against NULL allows us to handle files without + * attr.sample_id_all and/or without PERF_SAMPLE_ID. In the + * future probably it'll be a good idea to restrict event + * processing via perf_session to files with both set. + */ + hists__inc_nr_events(&evsel->hists, event->header.type); + } + + machine = perf_session__find_machine_for_cpumode(session, event); + + switch (event->header.type) { + case PERF_RECORD_SAMPLE: + dump_sample(evsel, event, sample); + if (evsel == NULL) { + ++session->stats.nr_unknown_id; + return 0; + } + if (machine == NULL) { + ++session->stats.nr_unprocessable_samples; + return 0; + } + return tool->sample(tool, event, sample, evsel, machine); + case PERF_RECORD_MMAP: + return tool->mmap(tool, event, sample, machine); + case PERF_RECORD_COMM: + return tool->comm(tool, event, sample, machine); + case PERF_RECORD_FORK: + return tool->fork(tool, event, sample, machine); + case PERF_RECORD_EXIT: + return tool->exit(tool, event, sample, machine); + case PERF_RECORD_LOST: + if (tool->lost == perf_event__process_lost) + session->stats.total_lost += event->lost.lost; + return tool->lost(tool, event, sample, machine); + case PERF_RECORD_READ: + return tool->read(tool, event, sample, evsel, machine); + case PERF_RECORD_THROTTLE: + return tool->throttle(tool, event, sample, machine); + case PERF_RECORD_UNTHROTTLE: + return tool->unthrottle(tool, event, sample, machine); + default: + ++session->stats.nr_unknown_events; + return -1; + } +} + +static int perf_session__preprocess_sample(struct perf_session *session, + union perf_event *event, struct perf_sample *sample) +{ + if (event->header.type != PERF_RECORD_SAMPLE || + !(perf_evlist__sample_type(session->evlist) & PERF_SAMPLE_CALLCHAIN)) + return 0; + + if (!ip_callchain__valid(sample->callchain, event)) { + pr_debug("call-chain problem with event, skipping it.\n"); + ++session->stats.nr_invalid_chains; + session->stats.total_invalid_chains += sample->period; + return -EINVAL; + } + return 0; +} + +static int perf_session__process_user_event(struct perf_session *session, union perf_event *event, + struct perf_tool *tool, u64 file_offset) +{ + int err; + + dump_event(session, event, file_offset, NULL); + + /* These events are processed right away */ + switch (event->header.type) { + case PERF_RECORD_HEADER_ATTR: + err = tool->attr(event, &session->evlist); + if (err == 0) + perf_session__set_id_hdr_size(session); + return err; + case PERF_RECORD_HEADER_EVENT_TYPE: + return tool->event_type(tool, event); + case PERF_RECORD_HEADER_TRACING_DATA: + /* setup for reading amidst mmap */ + lseek(session->fd, file_offset, SEEK_SET); + return tool->tracing_data(event, session); + case PERF_RECORD_HEADER_BUILD_ID: + return tool->build_id(tool, event, session); + case PERF_RECORD_FINISHED_ROUND: + return tool->finished_round(tool, event, session); + default: + return -EINVAL; + } +} + +static void event_swap(union perf_event *event, bool sample_id_all) +{ + perf_event__swap_op swap; + + swap = perf_event__swap_ops[event->header.type]; + if (swap) + swap(event, sample_id_all); +} + +static int perf_session__process_event(struct perf_session *session, + union perf_event *event, + struct perf_tool *tool, + u64 file_offset) +{ + struct perf_sample sample; + int ret; + + if (session->header.needs_swap) + event_swap(event, perf_evlist__sample_id_all(session->evlist)); + + if (event->header.type >= PERF_RECORD_HEADER_MAX) + return -EINVAL; + + events_stats__inc(&session->stats, event->header.type); + + if (event->header.type >= PERF_RECORD_USER_TYPE_START) + return perf_session__process_user_event(session, event, tool, file_offset); + + /* + * For all kernel events we get the sample data + */ + ret = perf_evlist__parse_sample(session->evlist, event, &sample); + if (ret) + return ret; + + /* Preprocess sample records - precheck callchains */ + if (perf_session__preprocess_sample(session, event, &sample)) + return 0; + + if (tool->ordered_samples) { + ret = perf_session_queue_event(session, event, &sample, + file_offset); + if (ret != -ETIME) + return ret; + } + + return perf_session_deliver_event(session, event, &sample, tool, + file_offset); +} + +void perf_event_header__bswap(struct perf_event_header *self) +{ + self->type = bswap_32(self->type); + self->misc = bswap_16(self->misc); + self->size = bswap_16(self->size); +} + +struct thread *perf_session__findnew(struct perf_session *session, pid_t pid) +{ + return machine__findnew_thread(&session->machines.host, pid); +} + +static struct thread *perf_session__register_idle_thread(struct perf_session *self) +{ + struct thread *thread = perf_session__findnew(self, 0); + + if (thread == NULL || thread__set_comm(thread, "swapper")) { + pr_err("problem inserting idle task.\n"); + thread = NULL; + } + + return thread; +} + +static void perf_session__warn_about_errors(const struct perf_session *session, + const struct perf_tool *tool) +{ + if (tool->lost == perf_event__process_lost && + session->stats.nr_events[PERF_RECORD_LOST] != 0) { + ui__warning("Processed %d events and lost %d chunks!\n\n" + "Check IO/CPU overload!\n\n", + session->stats.nr_events[0], + session->stats.nr_events[PERF_RECORD_LOST]); + } + + if (session->stats.nr_unknown_events != 0) { + ui__warning("Found %u unknown events!\n\n" + "Is this an older tool processing a perf.data " + "file generated by a more recent tool?\n\n" + "If that is not the case, consider " + "reporting to linux-kernel@vger.kernel.org.\n\n", + session->stats.nr_unknown_events); + } + + if (session->stats.nr_unknown_id != 0) { + ui__warning("%u samples with id not present in the header\n", + session->stats.nr_unknown_id); + } + + if (session->stats.nr_invalid_chains != 0) { + ui__warning("Found invalid callchains!\n\n" + "%u out of %u events were discarded for this reason.\n\n" + "Consider reporting to linux-kernel@vger.kernel.org.\n\n", + session->stats.nr_invalid_chains, + session->stats.nr_events[PERF_RECORD_SAMPLE]); + } + + if (session->stats.nr_unprocessable_samples != 0) { + ui__warning("%u unprocessable samples recorded.\n" + "Do you have a KVM guest running and not using 'perf kvm'?\n", + session->stats.nr_unprocessable_samples); + } +} + +#define session_done() (*(volatile int *)(&session_done)) +volatile int session_done; + +static int __perf_session__process_pipe_events(struct perf_session *self, + struct perf_tool *tool) +{ + union perf_event *event; + uint32_t size, cur_size = 0; + void *buf = NULL; + int skip = 0; + u64 head; + int err; + void *p; + + perf_tool__fill_defaults(tool); + + head = 0; + cur_size = sizeof(union perf_event); + + buf = malloc(cur_size); + if (!buf) + return -errno; +more: + event = buf; + err = readn(self->fd, event, sizeof(struct perf_event_header)); + if (err <= 0) { + if (err == 0) + goto done; + + pr_err("failed to read event header\n"); + goto out_err; + } + + if (self->header.needs_swap) + perf_event_header__bswap(&event->header); + + size = event->header.size; + if (size == 0) + size = 8; + + if (size > cur_size) { + void *new = realloc(buf, size); + if (!new) { + pr_err("failed to allocate memory to read event\n"); + goto out_err; + } + buf = new; + cur_size = size; + event = buf; + } + p = event; + p += sizeof(struct perf_event_header); + + if (size - sizeof(struct perf_event_header)) { + err = readn(self->fd, p, size - sizeof(struct perf_event_header)); + if (err <= 0) { + if (err == 0) { + pr_err("unexpected end of event stream\n"); + goto done; + } + + pr_err("failed to read event data\n"); + goto out_err; + } + } + + if ((skip = perf_session__process_event(self, event, tool, head)) < 0) { + pr_err("%#" PRIx64 " [%#x]: failed to process type: %d\n", + head, event->header.size, event->header.type); + err = -EINVAL; + goto out_err; + } + + head += size; + + if (skip > 0) + head += skip; + + if (!session_done()) + goto more; +done: + err = 0; +out_err: + free(buf); + perf_session__warn_about_errors(self, tool); + perf_session_free_sample_buffers(self); + return err; +} + +static union perf_event * +fetch_mmaped_event(struct perf_session *session, + u64 head, size_t mmap_size, char *buf) +{ + union perf_event *event; + + /* + * Ensure we have enough space remaining to read + * the size of the event in the headers. + */ + if (head + sizeof(event->header) > mmap_size) + return NULL; + + event = (union perf_event *)(buf + head); + + if (session->header.needs_swap) + perf_event_header__bswap(&event->header); + + if (head + event->header.size > mmap_size) + return NULL; + + return event; +} + +/* + * On 64bit we can mmap the data file in one go. No need for tiny mmap + * slices. On 32bit we use 32MB. + */ +#if BITS_PER_LONG == 64 +#define MMAP_SIZE ULLONG_MAX +#define NUM_MMAPS 1 +#else +#define MMAP_SIZE (32 * 1024 * 1024ULL) +#define NUM_MMAPS 128 +#endif + +int __perf_session__process_events(struct perf_session *session, + u64 data_offset, u64 data_size, + u64 file_size, struct perf_tool *tool) +{ + u64 head, page_offset, file_offset, file_pos, progress_next; + int err, mmap_prot, mmap_flags, map_idx = 0; + size_t mmap_size; + char *buf, *mmaps[NUM_MMAPS]; + union perf_event *event; + uint32_t size; + + perf_tool__fill_defaults(tool); + + page_offset = page_size * (data_offset / page_size); + file_offset = page_offset; + head = data_offset - page_offset; + + if (data_offset + data_size < file_size) + file_size = data_offset + data_size; + + progress_next = file_size / 16; + + mmap_size = MMAP_SIZE; + if (mmap_size > file_size) + mmap_size = file_size; + + memset(mmaps, 0, sizeof(mmaps)); + + mmap_prot = PROT_READ; + mmap_flags = MAP_SHARED; + + if (session->header.needs_swap) { + mmap_prot |= PROT_WRITE; + mmap_flags = MAP_PRIVATE; + } +remap: + buf = mmap(NULL, mmap_size, mmap_prot, mmap_flags, session->fd, + file_offset); + if (buf == MAP_FAILED) { + pr_err("failed to mmap file\n"); + err = -errno; + goto out_err; + } + mmaps[map_idx] = buf; + map_idx = (map_idx + 1) & (ARRAY_SIZE(mmaps) - 1); + file_pos = file_offset + head; + +more: + event = fetch_mmaped_event(session, head, mmap_size, buf); + if (!event) { + if (mmaps[map_idx]) { + munmap(mmaps[map_idx], mmap_size); + mmaps[map_idx] = NULL; + } + + page_offset = page_size * (head / page_size); + file_offset += page_offset; + head -= page_offset; + goto remap; + } + + size = event->header.size; + + if (size == 0 || + perf_session__process_event(session, event, tool, file_pos) < 0) { + pr_err("%#" PRIx64 " [%#x]: failed to process type: %d\n", + file_offset + head, event->header.size, + event->header.type); + err = -EINVAL; + goto out_err; + } + + head += size; + file_pos += size; + + if (file_pos >= progress_next) { + progress_next += file_size / 16; + ui_progress__update(file_pos, file_size, + "Processing events..."); + } + + if (file_pos < file_size) + goto more; + + err = 0; + /* do the final flush for ordered samples */ + session->ordered_samples.next_flush = ULLONG_MAX; + err = flush_sample_queue(session, tool); +out_err: + ui_progress__finish(); + perf_session__warn_about_errors(session, tool); + perf_session_free_sample_buffers(session); + return err; +} + +int perf_session__process_events(struct perf_session *self, + struct perf_tool *tool) +{ + int err; + + if (perf_session__register_idle_thread(self) == NULL) + return -ENOMEM; + + if (!self->fd_pipe) + err = __perf_session__process_events(self, + self->header.data_offset, + self->header.data_size, + self->size, tool); + else + err = __perf_session__process_pipe_events(self, tool); + + return err; +} + +bool perf_session__has_traces(struct perf_session *session, const char *msg) +{ + if (!(perf_evlist__sample_type(session->evlist) & PERF_SAMPLE_RAW)) { + pr_err("No trace sample to read. Did you call 'perf %s'?\n", msg); + return false; + } + + return true; +} + +int maps__set_kallsyms_ref_reloc_sym(struct map **maps, + const char *symbol_name, u64 addr) +{ + char *bracket; + enum map_type i; + struct ref_reloc_sym *ref; + + ref = zalloc(sizeof(struct ref_reloc_sym)); + if (ref == NULL) + return -ENOMEM; + + ref->name = strdup(symbol_name); + if (ref->name == NULL) { + free(ref); + return -ENOMEM; + } + + bracket = strchr(ref->name, ']'); + if (bracket) + *bracket = '\0'; + + ref->addr = addr; + + for (i = 0; i < MAP__NR_TYPES; ++i) { + struct kmap *kmap = map__kmap(maps[i]); + kmap->ref_reloc_sym = ref; + } + + return 0; +} + +size_t perf_session__fprintf_dsos(struct perf_session *self, FILE *fp) +{ + return machines__fprintf_dsos(&self->machines, fp); +} + +size_t perf_session__fprintf_dsos_buildid(struct perf_session *self, FILE *fp, + bool (skip)(struct dso *dso, int parm), int parm) +{ + return machines__fprintf_dsos_buildid(&self->machines, fp, skip, parm); +} + +size_t perf_session__fprintf_nr_events(struct perf_session *session, FILE *fp) +{ + struct perf_evsel *pos; + size_t ret = fprintf(fp, "Aggregated stats:\n"); + + ret += events_stats__fprintf(&session->stats, fp); + + list_for_each_entry(pos, &session->evlist->entries, node) { + ret += fprintf(fp, "%s stats:\n", perf_evsel__name(pos)); + ret += events_stats__fprintf(&pos->hists.stats, fp); + } + + return ret; +} + +size_t perf_session__fprintf(struct perf_session *session, FILE *fp) +{ + /* + * FIXME: Here we have to actually print all the machines in this + * session, not just the host... + */ + return machine__fprintf(&session->machines.host, fp); +} + +void perf_session__remove_thread(struct perf_session *session, + struct thread *th) +{ + /* + * FIXME: This one makes no sense, we need to remove the thread from + * the machine it belongs to, perf_session can have many machines, so + * doing it always on ->machines.host is wrong. Fix when auditing all + * the 'perf kvm' code. + */ + machine__remove_thread(&session->machines.host, th); +} + +struct perf_evsel *perf_session__find_first_evtype(struct perf_session *session, + unsigned int type) +{ + struct perf_evsel *pos; + + list_for_each_entry(pos, &session->evlist->entries, node) { + if (pos->attr.type == type) + return pos; + } + return NULL; +} + +void perf_evsel__print_ip(struct perf_evsel *evsel, union perf_event *event, + struct perf_sample *sample, struct machine *machine, + int print_sym, int print_dso, int print_symoffset) +{ + struct addr_location al; + struct callchain_cursor_node *node; + + if (perf_event__preprocess_sample(event, machine, &al, sample, + NULL) < 0) { + error("problem processing %d event, skipping it.\n", + event->header.type); + return; + } + + if (symbol_conf.use_callchain && sample->callchain) { + + + if (machine__resolve_callchain(machine, evsel, al.thread, + sample, NULL) != 0) { + if (verbose) + error("Failed to resolve callchain. Skipping\n"); + return; + } + callchain_cursor_commit(&callchain_cursor); + + while (1) { + node = callchain_cursor_current(&callchain_cursor); + if (!node) + break; + + printf("\t%16" PRIx64, node->ip); + if (print_sym) { + printf(" "); + symbol__fprintf_symname(node->sym, stdout); + } + if (print_dso) { + printf(" ("); + map__fprintf_dsoname(node->map, stdout); + printf(")"); + } + printf("\n"); + + callchain_cursor_advance(&callchain_cursor); + } + + } else { + printf("%16" PRIx64, sample->ip); + if (print_sym) { + printf(" "); + if (print_symoffset) + symbol__fprintf_symname_offs(al.sym, &al, + stdout); + else + symbol__fprintf_symname(al.sym, stdout); + } + + if (print_dso) { + printf(" ("); + map__fprintf_dsoname(al.map, stdout); + printf(")"); + } + } +} + +int perf_session__cpu_bitmap(struct perf_session *session, + const char *cpu_list, unsigned long *cpu_bitmap) +{ + int i; + struct cpu_map *map; + + for (i = 0; i < PERF_TYPE_MAX; ++i) { + struct perf_evsel *evsel; + + evsel = perf_session__find_first_evtype(session, i); + if (!evsel) + continue; + + if (!(evsel->attr.sample_type & PERF_SAMPLE_CPU)) { + pr_err("File does not contain CPU events. " + "Remove -c option to proceed.\n"); + return -1; + } + } + + map = cpu_map__new(cpu_list); + if (map == NULL) { + pr_err("Invalid cpu_list\n"); + return -1; + } + + for (i = 0; i < map->nr; i++) { + int cpu = map->map[i]; + + if (cpu >= MAX_NR_CPUS) { + pr_err("Requested CPU %d too large. " + "Consider raising MAX_NR_CPUS\n", cpu); + return -1; + } + + set_bit(cpu, cpu_bitmap); + } + + return 0; +} + +void perf_session__fprintf_info(struct perf_session *session, FILE *fp, + bool full) +{ + struct stat st; + int ret; + + if (session == NULL || fp == NULL) + return; + + ret = fstat(session->fd, &st); + if (ret == -1) + return; + + fprintf(fp, "# ========\n"); + fprintf(fp, "# captured on: %s", ctime(&st.st_ctime)); + perf_header__fprintf_info(session, fp, full); + fprintf(fp, "# ========\n#\n"); +} + + +int __perf_session__set_tracepoints_handlers(struct perf_session *session, + const struct perf_evsel_str_handler *assocs, + size_t nr_assocs) +{ + struct perf_evlist *evlist = session->evlist; + struct event_format *format; + struct perf_evsel *evsel; + char *tracepoint, *name; + size_t i; + int err; + + for (i = 0; i < nr_assocs; i++) { + err = -ENOMEM; + tracepoint = strdup(assocs[i].name); + if (tracepoint == NULL) + goto out; + + err = -ENOENT; + name = strchr(tracepoint, ':'); + if (name == NULL) + goto out_free; + + *name++ = '\0'; + format = pevent_find_event_by_name(session->pevent, + tracepoint, name); + if (format == NULL) { + /* + * Adding a handler for an event not in the session, + * just ignore it. + */ + goto next; + } + + evsel = perf_evlist__find_tracepoint_by_id(evlist, format->id); + if (evsel == NULL) + goto next; + + err = -EEXIST; + if (evsel->handler.func != NULL) + goto out_free; + evsel->handler.func = assocs[i].handler; +next: + free(tracepoint); + } + + err = 0; +out: + return err; + +out_free: + free(tracepoint); + goto out; +} diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h new file mode 100644 index 00000000..b5c0847e --- /dev/null +++ b/tools/perf/util/session.h @@ -0,0 +1,119 @@ +#ifndef __PERF_SESSION_H +#define __PERF_SESSION_H + +#include "hist.h" +#include "event.h" +#include "header.h" +#include "machine.h" +#include "symbol.h" +#include "thread.h" +#include <linux/rbtree.h> +#include <linux/perf_event.h> + +struct sample_queue; +struct ip_callchain; +struct thread; + +struct ordered_samples { + u64 last_flush; + u64 next_flush; + u64 max_timestamp; + struct list_head samples; + struct list_head sample_cache; + struct list_head to_free; + struct sample_queue *sample_buffer; + struct sample_queue *last_sample; + int sample_buffer_idx; + unsigned int nr_samples; +}; + +struct perf_session { + struct perf_header header; + unsigned long size; + struct machines machines; + struct perf_evlist *evlist; + struct pevent *pevent; + struct events_stats stats; + int fd; + bool fd_pipe; + bool repipe; + int cwdlen; + char *cwd; + struct ordered_samples ordered_samples; + char filename[1]; +}; + +struct perf_tool; + +struct perf_session *perf_session__new(const char *filename, int mode, + bool force, bool repipe, + struct perf_tool *tool); +void perf_session__delete(struct perf_session *session); + +void perf_event_header__bswap(struct perf_event_header *self); + +int __perf_session__process_events(struct perf_session *self, + u64 data_offset, u64 data_size, u64 size, + struct perf_tool *tool); +int perf_session__process_events(struct perf_session *self, + struct perf_tool *tool); + +int perf_session__resolve_callchain(struct perf_session *self, struct perf_evsel *evsel, + struct thread *thread, + struct ip_callchain *chain, + struct symbol **parent); + +bool perf_session__has_traces(struct perf_session *self, const char *msg); + +void mem_bswap_64(void *src, int byte_size); +void mem_bswap_32(void *src, int byte_size); +void perf_event__attr_swap(struct perf_event_attr *attr); + +int perf_session__create_kernel_maps(struct perf_session *self); + +void perf_session__set_id_hdr_size(struct perf_session *session); +void perf_session__remove_thread(struct perf_session *self, struct thread *th); + +static inline +struct machine *perf_session__find_machine(struct perf_session *self, pid_t pid) +{ + return machines__find(&self->machines, pid); +} + +static inline +struct machine *perf_session__findnew_machine(struct perf_session *self, pid_t pid) +{ + return machines__findnew(&self->machines, pid); +} + +struct thread *perf_session__findnew(struct perf_session *self, pid_t pid); +size_t perf_session__fprintf(struct perf_session *self, FILE *fp); + +size_t perf_session__fprintf_dsos(struct perf_session *self, FILE *fp); + +size_t perf_session__fprintf_dsos_buildid(struct perf_session *session, FILE *fp, + bool (fn)(struct dso *dso, int parm), int parm); + +size_t perf_session__fprintf_nr_events(struct perf_session *session, FILE *fp); + +struct perf_evsel *perf_session__find_first_evtype(struct perf_session *session, + unsigned int type); + +void perf_evsel__print_ip(struct perf_evsel *evsel, union perf_event *event, + struct perf_sample *sample, struct machine *machine, + int print_sym, int print_dso, int print_symoffset); + +int perf_session__cpu_bitmap(struct perf_session *session, + const char *cpu_list, unsigned long *cpu_bitmap); + +void perf_session__fprintf_info(struct perf_session *s, FILE *fp, bool full); + +struct perf_evsel_str_handler; + +int __perf_session__set_tracepoints_handlers(struct perf_session *session, + const struct perf_evsel_str_handler *assocs, + size_t nr_assocs); + +#define perf_session__set_tracepoints_handlers(session, array) \ + __perf_session__set_tracepoints_handlers(session, array, ARRAY_SIZE(array)) +#endif /* __PERF_SESSION_H */ diff --git a/tools/perf/util/setup.py b/tools/perf/util/setup.py new file mode 100644 index 00000000..73d51026 --- /dev/null +++ b/tools/perf/util/setup.py @@ -0,0 +1,46 @@ +#!/usr/bin/python2 + +from distutils.core import setup, Extension +from os import getenv + +from distutils.command.build_ext import build_ext as _build_ext +from distutils.command.install_lib import install_lib as _install_lib + +class build_ext(_build_ext): + def finalize_options(self): + _build_ext.finalize_options(self) + self.build_lib = build_lib + self.build_temp = build_tmp + +class install_lib(_install_lib): + def finalize_options(self): + _install_lib.finalize_options(self) + self.build_dir = build_lib + + +cflags = ['-fno-strict-aliasing', '-Wno-write-strings'] +cflags += getenv('CFLAGS', '').split() + +build_lib = getenv('PYTHON_EXTBUILD_LIB') +build_tmp = getenv('PYTHON_EXTBUILD_TMP') +libtraceevent = getenv('LIBTRACEEVENT') + +ext_sources = [f.strip() for f in file('util/python-ext-sources') + if len(f.strip()) > 0 and f[0] != '#'] + +perf = Extension('perf', + sources = ext_sources, + include_dirs = ['util/include'], + extra_compile_args = cflags, + extra_objects = [libtraceevent], + ) + +setup(name='perf', + version='0.1', + description='Interface with the Linux profiling infrastructure', + author='Arnaldo Carvalho de Melo', + author_email='acme@redhat.com', + license='GPLv2', + url='http://perf.wiki.kernel.org', + ext_modules=[perf], + cmdclass={'build_ext': build_ext, 'install_lib': install_lib}) diff --git a/tools/perf/util/sigchain.c b/tools/perf/util/sigchain.c new file mode 100644 index 00000000..ba785e9b --- /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); +} + +static 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 00000000..959d64eb --- /dev/null +++ b/tools/perf/util/sigchain.h @@ -0,0 +1,10 @@ +#ifndef __PERF_SIGCHAIN_H +#define __PERF_SIGCHAIN_H + +typedef void (*sigchain_fun)(int); + +int sigchain_pop(int sig); + +void sigchain_push_common(sigchain_fun f); + +#endif /* __PERF_SIGCHAIN_H */ diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c new file mode 100644 index 00000000..d41926cb --- /dev/null +++ b/tools/perf/util/sort.c @@ -0,0 +1,603 @@ +#include "sort.h" +#include "hist.h" + +regex_t parent_regex; +const char default_parent_pattern[] = "^sys_|^do_page_fault"; +const char *parent_pattern = default_parent_pattern; +const char default_sort_order[] = "comm,dso,symbol"; +const char *sort_order = default_sort_order; +int sort__need_collapse = 0; +int sort__has_parent = 0; +int sort__has_sym = 0; +int sort__branch_mode = -1; /* -1 = means not set */ + +enum sort_type sort__first_dimension; + +LIST_HEAD(hist_entry__sort_list); + +static int repsep_snprintf(char *bf, size_t size, const char *fmt, ...) +{ + int n; + va_list ap; + + va_start(ap, fmt); + n = vsnprintf(bf, size, fmt, ap); + if (symbol_conf.field_sep && n > 0) { + char *sep = bf; + + while (1) { + sep = strchr(sep, *symbol_conf.field_sep); + if (sep == NULL) + break; + *sep = '.'; + } + } + va_end(ap); + + if (n >= (int)size) + return size - 1; + return n; +} + +static int64_t cmp_null(void *l, void *r) +{ + if (!l && !r) + return 0; + else if (!l) + return -1; + else + return 1; +} + +/* --sort pid */ + +static int64_t +sort__thread_cmp(struct hist_entry *left, struct hist_entry *right) +{ + return right->thread->pid - left->thread->pid; +} + +static int hist_entry__thread_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) +{ + return repsep_snprintf(bf, size, "%*s:%5d", width - 6, + self->thread->comm ?: "", self->thread->pid); +} + +struct sort_entry sort_thread = { + .se_header = "Command: Pid", + .se_cmp = sort__thread_cmp, + .se_snprintf = hist_entry__thread_snprintf, + .se_width_idx = HISTC_THREAD, +}; + +/* --sort comm */ + +static int64_t +sort__comm_cmp(struct hist_entry *left, struct hist_entry *right) +{ + return right->thread->pid - left->thread->pid; +} + +static int64_t +sort__comm_collapse(struct hist_entry *left, struct hist_entry *right) +{ + char *comm_l = left->thread->comm; + char *comm_r = right->thread->comm; + + if (!comm_l || !comm_r) + return cmp_null(comm_l, comm_r); + + return strcmp(comm_l, comm_r); +} + +static int hist_entry__comm_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) +{ + return repsep_snprintf(bf, size, "%*s", width, self->thread->comm); +} + +struct sort_entry sort_comm = { + .se_header = "Command", + .se_cmp = sort__comm_cmp, + .se_collapse = sort__comm_collapse, + .se_snprintf = hist_entry__comm_snprintf, + .se_width_idx = HISTC_COMM, +}; + +/* --sort dso */ + +static int64_t _sort__dso_cmp(struct map *map_l, struct map *map_r) +{ + struct dso *dso_l = map_l ? map_l->dso : NULL; + struct dso *dso_r = map_r ? map_r->dso : NULL; + const char *dso_name_l, *dso_name_r; + + if (!dso_l || !dso_r) + return cmp_null(dso_l, dso_r); + + if (verbose) { + dso_name_l = dso_l->long_name; + dso_name_r = dso_r->long_name; + } else { + dso_name_l = dso_l->short_name; + dso_name_r = dso_r->short_name; + } + + return strcmp(dso_name_l, dso_name_r); +} + +static int64_t +sort__dso_cmp(struct hist_entry *left, struct hist_entry *right) +{ + return _sort__dso_cmp(left->ms.map, right->ms.map); +} + +static int _hist_entry__dso_snprintf(struct map *map, char *bf, + size_t size, unsigned int width) +{ + if (map && map->dso) { + const char *dso_name = !verbose ? map->dso->short_name : + map->dso->long_name; + return repsep_snprintf(bf, size, "%-*s", width, dso_name); + } + + return repsep_snprintf(bf, size, "%-*s", width, "[unknown]"); +} + +static int hist_entry__dso_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) +{ + return _hist_entry__dso_snprintf(self->ms.map, bf, size, width); +} + +struct sort_entry sort_dso = { + .se_header = "Shared Object", + .se_cmp = sort__dso_cmp, + .se_snprintf = hist_entry__dso_snprintf, + .se_width_idx = HISTC_DSO, +}; + +/* --sort symbol */ + +static int64_t _sort__sym_cmp(struct symbol *sym_l, struct symbol *sym_r) +{ + u64 ip_l, ip_r; + + if (!sym_l || !sym_r) + return cmp_null(sym_l, sym_r); + + if (sym_l == sym_r) + return 0; + + ip_l = sym_l->start; + ip_r = sym_r->start; + + return (int64_t)(ip_r - ip_l); +} + +static int64_t +sort__sym_cmp(struct hist_entry *left, struct hist_entry *right) +{ + if (!left->ms.sym && !right->ms.sym) + return right->level - left->level; + + return _sort__sym_cmp(left->ms.sym, right->ms.sym); +} + +static int _hist_entry__sym_snprintf(struct map *map, struct symbol *sym, + u64 ip, char level, char *bf, size_t size, + unsigned int width) +{ + size_t ret = 0; + + if (verbose) { + char o = map ? dso__symtab_origin(map->dso) : '!'; + ret += repsep_snprintf(bf, size, "%-#*llx %c ", + BITS_PER_LONG / 4, ip, o); + } + + ret += repsep_snprintf(bf + ret, size - ret, "[%c] ", level); + if (sym) + ret += repsep_snprintf(bf + ret, size - ret, "%-*s", + width - ret, + sym->name); + else { + size_t len = BITS_PER_LONG / 4; + ret += repsep_snprintf(bf + ret, size - ret, "%-#.*llx", + len, ip); + ret += repsep_snprintf(bf + ret, size - ret, "%-*s", + width - ret, ""); + } + + return ret; +} + +static int hist_entry__sym_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) +{ + return _hist_entry__sym_snprintf(self->ms.map, self->ms.sym, self->ip, + self->level, bf, size, width); +} + +struct sort_entry sort_sym = { + .se_header = "Symbol", + .se_cmp = sort__sym_cmp, + .se_snprintf = hist_entry__sym_snprintf, + .se_width_idx = HISTC_SYMBOL, +}; + +/* --sort srcline */ + +static int64_t +sort__srcline_cmp(struct hist_entry *left, struct hist_entry *right) +{ + return (int64_t)(right->ip - left->ip); +} + +static int hist_entry__srcline_snprintf(struct hist_entry *self, char *bf, + size_t size, + unsigned int width __maybe_unused) +{ + FILE *fp = NULL; + char cmd[PATH_MAX + 2], *path = self->srcline, *nl; + size_t line_len; + + if (path != NULL) + goto out_path; + + if (!self->ms.map) + goto out_ip; + + if (!strncmp(self->ms.map->dso->long_name, "/tmp/perf-", 10)) + goto out_ip; + + snprintf(cmd, sizeof(cmd), "addr2line -e %s %016" PRIx64, + self->ms.map->dso->long_name, self->ip); + fp = popen(cmd, "r"); + if (!fp) + goto out_ip; + + if (getline(&path, &line_len, fp) < 0 || !line_len) + goto out_ip; + self->srcline = strdup(path); + if (self->srcline == NULL) + goto out_ip; + + nl = strchr(self->srcline, '\n'); + if (nl != NULL) + *nl = '\0'; + path = self->srcline; +out_path: + if (fp) + pclose(fp); + return repsep_snprintf(bf, size, "%s", path); +out_ip: + if (fp) + pclose(fp); + return repsep_snprintf(bf, size, "%-#*llx", BITS_PER_LONG / 4, self->ip); +} + +struct sort_entry sort_srcline = { + .se_header = "Source:Line", + .se_cmp = sort__srcline_cmp, + .se_snprintf = hist_entry__srcline_snprintf, + .se_width_idx = HISTC_SRCLINE, +}; + +/* --sort parent */ + +static int64_t +sort__parent_cmp(struct hist_entry *left, struct hist_entry *right) +{ + struct symbol *sym_l = left->parent; + struct symbol *sym_r = right->parent; + + if (!sym_l || !sym_r) + return cmp_null(sym_l, sym_r); + + return strcmp(sym_l->name, sym_r->name); +} + +static int hist_entry__parent_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) +{ + return repsep_snprintf(bf, size, "%-*s", width, + self->parent ? self->parent->name : "[other]"); +} + +struct sort_entry sort_parent = { + .se_header = "Parent symbol", + .se_cmp = sort__parent_cmp, + .se_snprintf = hist_entry__parent_snprintf, + .se_width_idx = HISTC_PARENT, +}; + +/* --sort cpu */ + +static int64_t +sort__cpu_cmp(struct hist_entry *left, struct hist_entry *right) +{ + return right->cpu - left->cpu; +} + +static int hist_entry__cpu_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) +{ + return repsep_snprintf(bf, size, "%*d", width, self->cpu); +} + +struct sort_entry sort_cpu = { + .se_header = "CPU", + .se_cmp = sort__cpu_cmp, + .se_snprintf = hist_entry__cpu_snprintf, + .se_width_idx = HISTC_CPU, +}; + +/* sort keys for branch stacks */ + +static int64_t +sort__dso_from_cmp(struct hist_entry *left, struct hist_entry *right) +{ + return _sort__dso_cmp(left->branch_info->from.map, + right->branch_info->from.map); +} + +static int hist_entry__dso_from_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) +{ + return _hist_entry__dso_snprintf(self->branch_info->from.map, + bf, size, width); +} + +static int64_t +sort__dso_to_cmp(struct hist_entry *left, struct hist_entry *right) +{ + return _sort__dso_cmp(left->branch_info->to.map, + right->branch_info->to.map); +} + +static int hist_entry__dso_to_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) +{ + return _hist_entry__dso_snprintf(self->branch_info->to.map, + bf, size, width); +} + +static int64_t +sort__sym_from_cmp(struct hist_entry *left, struct hist_entry *right) +{ + struct addr_map_symbol *from_l = &left->branch_info->from; + struct addr_map_symbol *from_r = &right->branch_info->from; + + if (!from_l->sym && !from_r->sym) + return right->level - left->level; + + return _sort__sym_cmp(from_l->sym, from_r->sym); +} + +static int64_t +sort__sym_to_cmp(struct hist_entry *left, struct hist_entry *right) +{ + struct addr_map_symbol *to_l = &left->branch_info->to; + struct addr_map_symbol *to_r = &right->branch_info->to; + + if (!to_l->sym && !to_r->sym) + return right->level - left->level; + + return _sort__sym_cmp(to_l->sym, to_r->sym); +} + +static int hist_entry__sym_from_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) +{ + struct addr_map_symbol *from = &self->branch_info->from; + return _hist_entry__sym_snprintf(from->map, from->sym, from->addr, + self->level, bf, size, width); + +} + +static int hist_entry__sym_to_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width) +{ + struct addr_map_symbol *to = &self->branch_info->to; + return _hist_entry__sym_snprintf(to->map, to->sym, to->addr, + self->level, bf, size, width); + +} + +struct sort_entry sort_dso_from = { + .se_header = "Source Shared Object", + .se_cmp = sort__dso_from_cmp, + .se_snprintf = hist_entry__dso_from_snprintf, + .se_width_idx = HISTC_DSO_FROM, +}; + +struct sort_entry sort_dso_to = { + .se_header = "Target Shared Object", + .se_cmp = sort__dso_to_cmp, + .se_snprintf = hist_entry__dso_to_snprintf, + .se_width_idx = HISTC_DSO_TO, +}; + +struct sort_entry sort_sym_from = { + .se_header = "Source Symbol", + .se_cmp = sort__sym_from_cmp, + .se_snprintf = hist_entry__sym_from_snprintf, + .se_width_idx = HISTC_SYMBOL_FROM, +}; + +struct sort_entry sort_sym_to = { + .se_header = "Target Symbol", + .se_cmp = sort__sym_to_cmp, + .se_snprintf = hist_entry__sym_to_snprintf, + .se_width_idx = HISTC_SYMBOL_TO, +}; + +static int64_t +sort__mispredict_cmp(struct hist_entry *left, struct hist_entry *right) +{ + const unsigned char mp = left->branch_info->flags.mispred != + right->branch_info->flags.mispred; + const unsigned char p = left->branch_info->flags.predicted != + right->branch_info->flags.predicted; + + return mp || p; +} + +static int hist_entry__mispredict_snprintf(struct hist_entry *self, char *bf, + size_t size, unsigned int width){ + static const char *out = "N/A"; + + if (self->branch_info->flags.predicted) + out = "N"; + else if (self->branch_info->flags.mispred) + out = "Y"; + + return repsep_snprintf(bf, size, "%-*s", width, out); +} + +struct sort_entry sort_mispredict = { + .se_header = "Branch Mispredicted", + .se_cmp = sort__mispredict_cmp, + .se_snprintf = hist_entry__mispredict_snprintf, + .se_width_idx = HISTC_MISPREDICT, +}; + +struct sort_dimension { + const char *name; + struct sort_entry *entry; + int taken; +}; + +#define DIM(d, n, func) [d] = { .name = n, .entry = &(func) } + +static struct sort_dimension common_sort_dimensions[] = { + DIM(SORT_PID, "pid", sort_thread), + DIM(SORT_COMM, "comm", sort_comm), + DIM(SORT_DSO, "dso", sort_dso), + DIM(SORT_SYM, "symbol", sort_sym), + DIM(SORT_PARENT, "parent", sort_parent), + DIM(SORT_CPU, "cpu", sort_cpu), + DIM(SORT_SRCLINE, "srcline", sort_srcline), +}; + +#undef DIM + +#define DIM(d, n, func) [d - __SORT_BRANCH_STACK] = { .name = n, .entry = &(func) } + +static struct sort_dimension bstack_sort_dimensions[] = { + DIM(SORT_DSO_FROM, "dso_from", sort_dso_from), + DIM(SORT_DSO_TO, "dso_to", sort_dso_to), + DIM(SORT_SYM_FROM, "symbol_from", sort_sym_from), + DIM(SORT_SYM_TO, "symbol_to", sort_sym_to), + DIM(SORT_MISPREDICT, "mispredict", sort_mispredict), +}; + +#undef DIM + +int sort_dimension__add(const char *tok) +{ + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(common_sort_dimensions); i++) { + struct sort_dimension *sd = &common_sort_dimensions[i]; + + if (strncasecmp(tok, sd->name, strlen(tok))) + continue; + + if (sd->entry == &sort_parent) { + int ret = regcomp(&parent_regex, parent_pattern, REG_EXTENDED); + if (ret) { + char err[BUFSIZ]; + + regerror(ret, &parent_regex, err, sizeof(err)); + pr_err("Invalid regex: %s\n%s", parent_pattern, err); + return -EINVAL; + } + sort__has_parent = 1; + } else if (sd->entry == &sort_sym) { + sort__has_sym = 1; + } + + if (sd->taken) + return 0; + + if (sd->entry->se_collapse) + sort__need_collapse = 1; + + if (list_empty(&hist_entry__sort_list)) + sort__first_dimension = i; + + list_add_tail(&sd->entry->list, &hist_entry__sort_list); + sd->taken = 1; + + return 0; + } + + for (i = 0; i < ARRAY_SIZE(bstack_sort_dimensions); i++) { + struct sort_dimension *sd = &bstack_sort_dimensions[i]; + + if (strncasecmp(tok, sd->name, strlen(tok))) + continue; + + if (sort__branch_mode != 1) + return -EINVAL; + + if (sd->entry == &sort_sym_from || sd->entry == &sort_sym_to) + sort__has_sym = 1; + + if (sd->taken) + return 0; + + if (sd->entry->se_collapse) + sort__need_collapse = 1; + + if (list_empty(&hist_entry__sort_list)) + sort__first_dimension = i + __SORT_BRANCH_STACK; + + list_add_tail(&sd->entry->list, &hist_entry__sort_list); + sd->taken = 1; + + return 0; + } + + return -ESRCH; +} + +int setup_sorting(void) +{ + char *tmp, *tok, *str = strdup(sort_order); + int ret = 0; + + if (str == NULL) { + error("Not enough memory to setup sort keys"); + return -ENOMEM; + } + + for (tok = strtok_r(str, ", ", &tmp); + tok; tok = strtok_r(NULL, ", ", &tmp)) { + ret = sort_dimension__add(tok); + if (ret == -EINVAL) { + error("Invalid --sort key: `%s'", tok); + break; + } else if (ret == -ESRCH) { + error("Unknown --sort key: `%s'", tok); + break; + } + } + + free(str); + return ret; +} + +void sort_entry__setup_elide(struct sort_entry *self, struct strlist *list, + const char *list_name, FILE *fp) +{ + if (list && strlist__nr_entries(list) == 1) { + if (fp != NULL) + fprintf(fp, "# %s: %s\n", list_name, + strlist__entry(list, 0)->s); + self->elide = true; + } +} diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h new file mode 100644 index 00000000..b13e56f6 --- /dev/null +++ b/tools/perf/util/sort.h @@ -0,0 +1,168 @@ +#ifndef __PERF_SORT_H +#define __PERF_SORT_H +#include "../builtin.h" + +#include "util.h" + +#include "color.h" +#include <linux/list.h> +#include "cache.h" +#include <linux/rbtree.h> +#include "symbol.h" +#include "string.h" +#include "callchain.h" +#include "strlist.h" +#include "values.h" + +#include "../perf.h" +#include "debug.h" +#include "header.h" + +#include "parse-options.h" +#include "parse-events.h" + +#include "thread.h" +#include "sort.h" + +extern regex_t parent_regex; +extern const char *sort_order; +extern const char default_parent_pattern[]; +extern const char *parent_pattern; +extern const char default_sort_order[]; +extern int sort__need_collapse; +extern int sort__has_parent; +extern int sort__has_sym; +extern int sort__branch_mode; +extern struct sort_entry sort_comm; +extern struct sort_entry sort_dso; +extern struct sort_entry sort_sym; +extern struct sort_entry sort_parent; +extern struct sort_entry sort_dso_from; +extern struct sort_entry sort_dso_to; +extern struct sort_entry sort_sym_from; +extern struct sort_entry sort_sym_to; +extern enum sort_type sort__first_dimension; + +struct he_stat { + u64 period; + u64 period_sys; + u64 period_us; + u64 period_guest_sys; + u64 period_guest_us; + u32 nr_events; +}; + +struct hist_entry_diff { + bool computed; + + /* PERF_HPP__DELTA */ + double period_ratio_delta; + + /* PERF_HPP__RATIO */ + double period_ratio; + + /* HISTC_WEIGHTED_DIFF */ + s64 wdiff; +}; + +/** + * struct hist_entry - histogram entry + * + * @row_offset - offset from the first callchain expanded to appear on screen + * @nr_rows - rows expanded in callchain, recalculated on folding/unfolding + */ +struct hist_entry { + struct rb_node rb_node_in; + struct rb_node rb_node; + union { + struct list_head node; + struct list_head head; + } pairs; + struct he_stat stat; + struct map_symbol ms; + struct thread *thread; + u64 ip; + s32 cpu; + + struct hist_entry_diff diff; + + /* XXX These two should move to some tree widget lib */ + u16 row_offset; + u16 nr_rows; + + bool init_have_children; + char level; + bool used; + u8 filtered; + char *srcline; + struct symbol *parent; + unsigned long position; + struct rb_root sorted_chain; + struct branch_info *branch_info; + struct hists *hists; + struct callchain_root callchain[0]; +}; + +static inline bool hist_entry__has_pairs(struct hist_entry *he) +{ + return !list_empty(&he->pairs.node); +} + +static inline struct hist_entry *hist_entry__next_pair(struct hist_entry *he) +{ + if (hist_entry__has_pairs(he)) + return list_entry(he->pairs.node.next, struct hist_entry, pairs.node); + return NULL; +} + +static inline void hist_entry__add_pair(struct hist_entry *he, + struct hist_entry *pair) +{ + list_add_tail(&he->pairs.head, &pair->pairs.node); +} + +enum sort_type { + /* common sort keys */ + SORT_PID, + SORT_COMM, + SORT_DSO, + SORT_SYM, + SORT_PARENT, + SORT_CPU, + SORT_SRCLINE, + + /* branch stack specific sort keys */ + __SORT_BRANCH_STACK, + SORT_DSO_FROM = __SORT_BRANCH_STACK, + SORT_DSO_TO, + SORT_SYM_FROM, + SORT_SYM_TO, + SORT_MISPREDICT, +}; + +/* + * configurable sorting bits + */ + +struct sort_entry { + struct list_head list; + + const char *se_header; + + int64_t (*se_cmp)(struct hist_entry *, struct hist_entry *); + int64_t (*se_collapse)(struct hist_entry *, struct hist_entry *); + int (*se_snprintf)(struct hist_entry *self, char *bf, size_t size, + unsigned int width); + u8 se_width_idx; + bool elide; +}; + +extern struct sort_entry sort_thread; +extern struct list_head hist_entry__sort_list; + +int setup_sorting(void); +extern int sort_dimension__add(const char *); +void sort_entry__setup_elide(struct sort_entry *self, struct strlist *list, + const char *list_name, FILE *fp); + +#endif /* __PERF_SORT_H */ diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c new file mode 100644 index 00000000..23742126 --- /dev/null +++ b/tools/perf/util/stat.c @@ -0,0 +1,57 @@ +#include <math.h> + +#include "stat.h" + +void update_stats(struct stats *stats, u64 val) +{ + double delta; + + stats->n++; + delta = val - stats->mean; + stats->mean += delta / stats->n; + stats->M2 += delta*(val - stats->mean); +} + +double avg_stats(struct stats *stats) +{ + return stats->mean; +} + +/* + * http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance + * + * (\Sum n_i^2) - ((\Sum n_i)^2)/n + * s^2 = ------------------------------- + * n - 1 + * + * http://en.wikipedia.org/wiki/Stddev + * + * The std dev of the mean is related to the std dev by: + * + * s + * s_mean = ------- + * sqrt(n) + * + */ +double stddev_stats(struct stats *stats) +{ + double variance, variance_mean; + + if (!stats->n) + return 0.0; + + variance = stats->M2 / (stats->n - 1); + variance_mean = variance / stats->n; + + return sqrt(variance_mean); +} + +double rel_stddev_stats(double stddev, double avg) +{ + double pct = 0.0; + + if (avg) + pct = 100.0 * stddev/avg; + + return pct; +} diff --git a/tools/perf/util/stat.h b/tools/perf/util/stat.h new file mode 100644 index 00000000..588367c3 --- /dev/null +++ b/tools/perf/util/stat.h @@ -0,0 +1,16 @@ +#ifndef __PERF_STATS_H +#define __PERF_STATS_H + +#include "types.h" + +struct stats +{ + double n, mean, M2; +}; + +void update_stats(struct stats *stats, u64 val); +double avg_stats(struct stats *stats); +double stddev_stats(struct stats *stats); +double rel_stddev_stats(double stddev, double avg); + +#endif diff --git a/tools/perf/util/strbuf.c b/tools/perf/util/strbuf.c new file mode 100644 index 00000000..cfa90688 --- /dev/null +++ b/tools/perf/util/strbuf.c @@ -0,0 +1,134 @@ +#include "cache.h" +#include <linux/kernel.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, ssize_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_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); +} + +static 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_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_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 vsnprintf is broken"); + } + } + strbuf_setlen(sb, sb->len + len); +} + +ssize_t strbuf_read(struct strbuf *sb, int fd, ssize_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; +} diff --git a/tools/perf/util/strbuf.h b/tools/perf/util/strbuf.h new file mode 100644 index 00000000..436ac319 --- /dev/null +++ b/tools/perf/util/strbuf.h @@ -0,0 +1,92 @@ +#ifndef __PERF_STRBUF_H +#define __PERF_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 *buf, ssize_t hint); +extern void strbuf_release(struct strbuf *); +extern char *strbuf_detach(struct strbuf *, size_t *); + +/*----- strbuf size related -----*/ +static inline ssize_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'; +} + +/*----- 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_remove(struct strbuf *, size_t pos, size_t len); + +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)); +} + +__attribute__((format(printf,2,3))) +extern void strbuf_addf(struct strbuf *sb, const char *fmt, ...); + +/* XXX: if read fails, any partial read is undone */ +extern ssize_t strbuf_read(struct strbuf *, int fd, ssize_t hint); + +#endif /* __PERF_STRBUF_H */ diff --git a/tools/perf/util/strfilter.c b/tools/perf/util/strfilter.c new file mode 100644 index 00000000..834c8ebf --- /dev/null +++ b/tools/perf/util/strfilter.c @@ -0,0 +1,199 @@ +#include "util.h" +#include "string.h" +#include "strfilter.h" + +/* Operators */ +static const char *OP_and = "&"; /* Logical AND */ +static const char *OP_or = "|"; /* Logical OR */ +static const char *OP_not = "!"; /* Logical NOT */ + +#define is_operator(c) ((c) == '|' || (c) == '&' || (c) == '!') +#define is_separator(c) (is_operator(c) || (c) == '(' || (c) == ')') + +static void strfilter_node__delete(struct strfilter_node *self) +{ + if (self) { + if (self->p && !is_operator(*self->p)) + free((char *)self->p); + strfilter_node__delete(self->l); + strfilter_node__delete(self->r); + free(self); + } +} + +void strfilter__delete(struct strfilter *self) +{ + if (self) { + strfilter_node__delete(self->root); + free(self); + } +} + +static const char *get_token(const char *s, const char **e) +{ + const char *p; + + while (isspace(*s)) /* Skip spaces */ + s++; + + if (*s == '\0') { + p = s; + goto end; + } + + p = s + 1; + if (!is_separator(*s)) { + /* End search */ +retry: + while (*p && !is_separator(*p) && !isspace(*p)) + p++; + /* Escape and special case: '!' is also used in glob pattern */ + if (*(p - 1) == '\\' || (*p == '!' && *(p - 1) == '[')) { + p++; + goto retry; + } + } +end: + *e = p; + return s; +} + +static struct strfilter_node *strfilter_node__alloc(const char *op, + struct strfilter_node *l, + struct strfilter_node *r) +{ + struct strfilter_node *ret = zalloc(sizeof(struct strfilter_node)); + + if (ret) { + ret->p = op; + ret->l = l; + ret->r = r; + } + + return ret; +} + +static struct strfilter_node *strfilter_node__new(const char *s, + const char **ep) +{ + struct strfilter_node root, *cur, *last_op; + const char *e; + + if (!s) + return NULL; + + memset(&root, 0, sizeof(root)); + last_op = cur = &root; + + s = get_token(s, &e); + while (*s != '\0' && *s != ')') { + switch (*s) { + case '&': /* Exchg last OP->r with AND */ + if (!cur->r || !last_op->r) + goto error; + cur = strfilter_node__alloc(OP_and, last_op->r, NULL); + if (!cur) + goto nomem; + last_op->r = cur; + last_op = cur; + break; + case '|': /* Exchg the root with OR */ + if (!cur->r || !root.r) + goto error; + cur = strfilter_node__alloc(OP_or, root.r, NULL); + if (!cur) + goto nomem; + root.r = cur; + last_op = cur; + break; + case '!': /* Add NOT as a leaf node */ + if (cur->r) + goto error; + cur->r = strfilter_node__alloc(OP_not, NULL, NULL); + if (!cur->r) + goto nomem; + cur = cur->r; + break; + case '(': /* Recursively parses inside the parenthesis */ + if (cur->r) + goto error; + cur->r = strfilter_node__new(s + 1, &s); + if (!s) + goto nomem; + if (!cur->r || *s != ')') + goto error; + e = s + 1; + break; + default: + if (cur->r) + goto error; + cur->r = strfilter_node__alloc(NULL, NULL, NULL); + if (!cur->r) + goto nomem; + cur->r->p = strndup(s, e - s); + if (!cur->r->p) + goto nomem; + } + s = get_token(e, &e); + } + if (!cur->r) + goto error; + *ep = s; + return root.r; +nomem: + s = NULL; +error: + *ep = s; + strfilter_node__delete(root.r); + return NULL; +} + +/* + * Parse filter rule and return new strfilter. + * Return NULL if fail, and *ep == NULL if memory allocation failed. + */ +struct strfilter *strfilter__new(const char *rules, const char **err) +{ + struct strfilter *ret = zalloc(sizeof(struct strfilter)); + const char *ep = NULL; + + if (ret) + ret->root = strfilter_node__new(rules, &ep); + + if (!ret || !ret->root || *ep != '\0') { + if (err) + *err = ep; + strfilter__delete(ret); + ret = NULL; + } + + return ret; +} + +static bool strfilter_node__compare(struct strfilter_node *self, + const char *str) +{ + if (!self || !self->p) + return false; + + switch (*self->p) { + case '|': /* OR */ + return strfilter_node__compare(self->l, str) || + strfilter_node__compare(self->r, str); + case '&': /* AND */ + return strfilter_node__compare(self->l, str) && + strfilter_node__compare(self->r, str); + case '!': /* NOT */ + return !strfilter_node__compare(self->r, str); + default: + return strglobmatch(str, self->p); + } +} + +/* Return true if STR matches the filter rules */ +bool strfilter__compare(struct strfilter *self, const char *str) +{ + if (!self) + return false; + return strfilter_node__compare(self->root, str); +} diff --git a/tools/perf/util/strfilter.h b/tools/perf/util/strfilter.h new file mode 100644 index 00000000..00f58a75 --- /dev/null +++ b/tools/perf/util/strfilter.h @@ -0,0 +1,48 @@ +#ifndef __PERF_STRFILTER_H +#define __PERF_STRFILTER_H +/* General purpose glob matching filter */ + +#include <linux/list.h> +#include <stdbool.h> + +/* A node of string filter */ +struct strfilter_node { + struct strfilter_node *l; /* Tree left branche (for &,|) */ + struct strfilter_node *r; /* Tree right branche (for !,&,|) */ + const char *p; /* Operator or rule */ +}; + +/* String filter */ +struct strfilter { + struct strfilter_node *root; +}; + +/** + * strfilter__new - Create a new string filter + * @rules: Filter rule, which is a combination of glob expressions. + * @err: Pointer which points an error detected on @rules + * + * Parse @rules and return new strfilter. Return NULL if an error detected. + * In that case, *@err will indicate where it is detected, and *@err is NULL + * if a memory allocation is failed. + */ +struct strfilter *strfilter__new(const char *rules, const char **err); + +/** + * strfilter__compare - compare given string and a string filter + * @self: String filter + * @str: target string + * + * Compare @str and @self. Return true if the str match the rule + */ +bool strfilter__compare(struct strfilter *self, const char *str); + +/** + * strfilter__delete - delete a string filter + * @self: String filter to delete + * + * Delete @self. + */ +void strfilter__delete(struct strfilter *self); + +#endif diff --git a/tools/perf/util/string.c b/tools/perf/util/string.c new file mode 100644 index 00000000..29c7b2cb --- /dev/null +++ b/tools/perf/util/string.c @@ -0,0 +1,389 @@ +#include "util.h" +#include "linux/string.h" + +#define K 1024LL +/* + * perf_atoll() + * Parse (\d+)(b|B|kb|KB|mb|MB|gb|GB|tb|TB) (e.g. "256MB") + * and return its numeric value + */ +s64 perf_atoll(const char *str) +{ + unsigned int i; + s64 length = -1, unit = 1; + + if (!isdigit(str[0])) + goto out_err; + + for (i = 1; i < strlen(str); i++) { + switch (str[i]) { + case 'B': + case 'b': + break; + case 'K': + if (str[i + 1] != 'B') + goto out_err; + else + goto kilo; + case 'k': + if (str[i + 1] != 'b') + goto out_err; +kilo: + unit = K; + break; + case 'M': + if (str[i + 1] != 'B') + goto out_err; + else + goto mega; + case 'm': + if (str[i + 1] != 'b') + goto out_err; +mega: + unit = K * K; + break; + case 'G': + if (str[i + 1] != 'B') + goto out_err; + else + goto giga; + case 'g': + if (str[i + 1] != 'b') + goto out_err; +giga: + unit = K * K * K; + break; + case 'T': + if (str[i + 1] != 'B') + goto out_err; + else + goto tera; + case 't': + if (str[i + 1] != 'b') + goto out_err; +tera: + unit = K * K * K * K; + break; + case '\0': /* only specified figures */ + unit = 1; + break; + default: + if (!isdigit(str[i])) + goto out_err; + break; + } + } + + length = atoll(str) * unit; + goto out; + +out_err: + length = -1; +out: + return length; +} + +/* + * Helper function for splitting a string into an argv-like array. + * originally copied from lib/argv_split.c + */ +static const char *skip_sep(const char *cp) +{ + while (*cp && isspace(*cp)) + cp++; + + return cp; +} + +static const char *skip_arg(const char *cp) +{ + while (*cp && !isspace(*cp)) + cp++; + + return cp; +} + +static int count_argc(const char *str) +{ + int count = 0; + + while (*str) { + str = skip_sep(str); + if (*str) { + count++; + str = skip_arg(str); + } + } + + return count; +} + +/** + * argv_free - free an argv + * @argv - the argument vector to be freed + * + * Frees an argv and the strings it points to. + */ +void argv_free(char **argv) +{ + char **p; + for (p = argv; *p; p++) + free(*p); + + free(argv); +} + +/** + * argv_split - split a string at whitespace, returning an argv + * @str: the string to be split + * @argcp: returned argument count + * + * Returns an array of pointers to strings which are split out from + * @str. This is performed by strictly splitting on white-space; no + * quote processing is performed. Multiple whitespace characters are + * considered to be a single argument separator. The returned array + * is always NULL-terminated. Returns NULL on memory allocation + * failure. + */ +char **argv_split(const char *str, int *argcp) +{ + int argc = count_argc(str); + char **argv = zalloc(sizeof(*argv) * (argc+1)); + char **argvp; + + if (argv == NULL) + goto out; + + if (argcp) + *argcp = argc; + + argvp = argv; + + while (*str) { + str = skip_sep(str); + + if (*str) { + const char *p = str; + char *t; + + str = skip_arg(str); + + t = strndup(p, str-p); + if (t == NULL) + goto fail; + *argvp++ = t; + } + } + *argvp = NULL; + +out: + return argv; + +fail: + argv_free(argv); + return NULL; +} + +/* Character class matching */ +static bool __match_charclass(const char *pat, char c, const char **npat) +{ + bool complement = false, ret = true; + + if (*pat == '!') { + complement = true; + pat++; + } + if (*pat++ == c) /* First character is special */ + goto end; + + while (*pat && *pat != ']') { /* Matching */ + if (*pat == '-' && *(pat + 1) != ']') { /* Range */ + if (*(pat - 1) <= c && c <= *(pat + 1)) + goto end; + if (*(pat - 1) > *(pat + 1)) + goto error; + pat += 2; + } else if (*pat++ == c) + goto end; + } + if (!*pat) + goto error; + ret = false; + +end: + while (*pat && *pat != ']') /* Searching closing */ + pat++; + if (!*pat) + goto error; + *npat = pat + 1; + return complement ? !ret : ret; + +error: + return false; +} + +/* Glob/lazy pattern matching */ +static bool __match_glob(const char *str, const char *pat, bool ignore_space) +{ + while (*str && *pat && *pat != '*') { + if (ignore_space) { + /* Ignore spaces for lazy matching */ + if (isspace(*str)) { + str++; + continue; + } + if (isspace(*pat)) { + pat++; + continue; + } + } + if (*pat == '?') { /* Matches any single character */ + str++; + pat++; + continue; + } else if (*pat == '[') /* Character classes/Ranges */ + if (__match_charclass(pat + 1, *str, &pat)) { + str++; + continue; + } else + return false; + else if (*pat == '\\') /* Escaped char match as normal char */ + pat++; + if (*str++ != *pat++) + return false; + } + /* Check wild card */ + if (*pat == '*') { + while (*pat == '*') + pat++; + if (!*pat) /* Tail wild card matches all */ + return true; + while (*str) + if (__match_glob(str++, pat, ignore_space)) + return true; + } + return !*str && !*pat; +} + +/** + * strglobmatch - glob expression pattern matching + * @str: the target string to match + * @pat: the pattern string to match + * + * This returns true if the @str matches @pat. @pat can includes wildcards + * ('*','?') and character classes ([CHARS], complementation and ranges are + * also supported). Also, this supports escape character ('\') to use special + * characters as normal character. + * + * Note: if @pat syntax is broken, this always returns false. + */ +bool strglobmatch(const char *str, const char *pat) +{ + return __match_glob(str, pat, false); +} + +/** + * strlazymatch - matching pattern strings lazily with glob pattern + * @str: the target string to match + * @pat: the pattern string to match + * + * This is similar to strglobmatch, except this ignores spaces in + * the target string. + */ +bool strlazymatch(const char *str, const char *pat) +{ + return __match_glob(str, pat, true); +} + +/** + * strtailcmp - Compare the tail of two strings + * @s1: 1st string to be compared + * @s2: 2nd string to be compared + * + * Return 0 if whole of either string is same as another's tail part. + */ +int strtailcmp(const char *s1, const char *s2) +{ + int i1 = strlen(s1); + int i2 = strlen(s2); + while (--i1 >= 0 && --i2 >= 0) { + if (s1[i1] != s2[i2]) + return s1[i1] - s2[i2]; + } + return 0; +} + +/** + * strxfrchar - Locate and replace character in @s + * @s: The string to be searched/changed. + * @from: Source character to be replaced. + * @to: Destination character. + * + * Return pointer to the changed string. + */ +char *strxfrchar(char *s, char from, char to) +{ + char *p = s; + + while ((p = strchr(p, from)) != NULL) + *p++ = to; + + return s; +} + +/** + * ltrim - Removes leading whitespace from @s. + * @s: The string to be stripped. + * + * Return pointer to the first non-whitespace character in @s. + */ +char *ltrim(char *s) +{ + int len = strlen(s); + + while (len && isspace(*s)) { + len--; + s++; + } + + return s; +} + +/** + * rtrim - Removes trailing whitespace from @s. + * @s: The string to be stripped. + * + * Note that the first trailing whitespace is replaced with a %NUL-terminator + * in the given string @s. Returns @s. + */ +char *rtrim(char *s) +{ + size_t size = strlen(s); + char *end; + + if (!size) + return s; + + end = s + size - 1; + while (end >= s && isspace(*end)) + end--; + *(end + 1) = '\0'; + + return s; +} + +/** + * memdup - duplicate region of memory + * @src: memory region to duplicate + * @len: memory region length + */ +void *memdup(const void *src, size_t len) +{ + void *p; + + p = malloc(len); + if (p) + memcpy(p, src, len); + + return p; +} diff --git a/tools/perf/util/strlist.c b/tools/perf/util/strlist.c new file mode 100644 index 00000000..55433aa4 --- /dev/null +++ b/tools/perf/util/strlist.c @@ -0,0 +1,172 @@ +/* + * (c) 2009 Arnaldo Carvalho de Melo <acme@redhat.com> + * + * Licensed under the GPLv2. + */ + +#include "strlist.h" +#include <errno.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +static +struct rb_node *strlist__node_new(struct rblist *rblist, const void *entry) +{ + const char *s = entry; + struct rb_node *rc = NULL; + struct strlist *strlist = container_of(rblist, struct strlist, rblist); + struct str_node *snode = malloc(sizeof(*snode)); + + if (snode != NULL) { + if (strlist->dupstr) { + s = strdup(s); + if (s == NULL) + goto out_delete; + } + snode->s = s; + rc = &snode->rb_node; + } + + return rc; + +out_delete: + free(snode); + return NULL; +} + +static void str_node__delete(struct str_node *snode, bool dupstr) +{ + if (dupstr) + free((void *)snode->s); + free(snode); +} + +static +void strlist__node_delete(struct rblist *rblist, struct rb_node *rb_node) +{ + struct strlist *slist = container_of(rblist, struct strlist, rblist); + struct str_node *snode = container_of(rb_node, struct str_node, rb_node); + + str_node__delete(snode, slist->dupstr); +} + +static int strlist__node_cmp(struct rb_node *rb_node, const void *entry) +{ + const char *str = entry; + struct str_node *snode = container_of(rb_node, struct str_node, rb_node); + + return strcmp(snode->s, str); +} + +int strlist__add(struct strlist *slist, const char *new_entry) +{ + return rblist__add_node(&slist->rblist, new_entry); +} + +int strlist__load(struct strlist *slist, const char *filename) +{ + char entry[1024]; + int err; + FILE *fp = fopen(filename, "r"); + + if (fp == NULL) + return errno; + + while (fgets(entry, sizeof(entry), fp) != NULL) { + const size_t len = strlen(entry); + + if (len == 0) + continue; + entry[len - 1] = '\0'; + + err = strlist__add(slist, entry); + if (err != 0) + goto out; + } + + err = 0; +out: + fclose(fp); + return err; +} + +void strlist__remove(struct strlist *slist, struct str_node *snode) +{ + rblist__remove_node(&slist->rblist, &snode->rb_node); +} + +struct str_node *strlist__find(struct strlist *slist, const char *entry) +{ + struct str_node *snode = NULL; + struct rb_node *rb_node = rblist__find(&slist->rblist, entry); + + if (rb_node) + snode = container_of(rb_node, struct str_node, rb_node); + + return snode; +} + +static int strlist__parse_list_entry(struct strlist *slist, const char *s) +{ + if (strncmp(s, "file://", 7) == 0) + return strlist__load(slist, s + 7); + + return strlist__add(slist, s); +} + +int strlist__parse_list(struct strlist *slist, const char *s) +{ + char *sep; + int err; + + while ((sep = strchr(s, ',')) != NULL) { + *sep = '\0'; + err = strlist__parse_list_entry(slist, s); + *sep = ','; + if (err != 0) + return err; + s = sep + 1; + } + + return *s ? strlist__parse_list_entry(slist, s) : 0; +} + +struct strlist *strlist__new(bool dupstr, const char *list) +{ + struct strlist *slist = malloc(sizeof(*slist)); + + if (slist != NULL) { + rblist__init(&slist->rblist); + slist->rblist.node_cmp = strlist__node_cmp; + slist->rblist.node_new = strlist__node_new; + slist->rblist.node_delete = strlist__node_delete; + + slist->dupstr = dupstr; + if (slist && strlist__parse_list(slist, list) != 0) + goto out_error; + } + + return slist; +out_error: + free(slist); + return NULL; +} + +void strlist__delete(struct strlist *slist) +{ + if (slist != NULL) + rblist__delete(&slist->rblist); +} + +struct str_node *strlist__entry(const struct strlist *slist, unsigned int idx) +{ + struct str_node *snode = NULL; + struct rb_node *rb_node; + + rb_node = rblist__entry(&slist->rblist, idx); + if (rb_node) + snode = container_of(rb_node, struct str_node, rb_node); + + return snode; +} diff --git a/tools/perf/util/strlist.h b/tools/perf/util/strlist.h new file mode 100644 index 00000000..5c7f8706 --- /dev/null +++ b/tools/perf/util/strlist.h @@ -0,0 +1,79 @@ +#ifndef __PERF_STRLIST_H +#define __PERF_STRLIST_H + +#include <linux/rbtree.h> +#include <stdbool.h> + +#include "rblist.h" + +struct str_node { + struct rb_node rb_node; + const char *s; +}; + +struct strlist { + struct rblist rblist; + bool dupstr; +}; + +struct strlist *strlist__new(bool dupstr, const char *slist); +void strlist__delete(struct strlist *slist); + +void strlist__remove(struct strlist *slist, struct str_node *sn); +int strlist__load(struct strlist *slist, const char *filename); +int strlist__add(struct strlist *slist, const char *str); + +struct str_node *strlist__entry(const struct strlist *slist, unsigned int idx); +struct str_node *strlist__find(struct strlist *slist, const char *entry); + +static inline bool strlist__has_entry(struct strlist *slist, const char *entry) +{ + return strlist__find(slist, entry) != NULL; +} + +static inline bool strlist__empty(const struct strlist *slist) +{ + return rblist__empty(&slist->rblist); +} + +static inline unsigned int strlist__nr_entries(const struct strlist *slist) +{ + return rblist__nr_entries(&slist->rblist); +} + +/* For strlist iteration */ +static inline struct str_node *strlist__first(struct strlist *slist) +{ + struct rb_node *rn = rb_first(&slist->rblist.entries); + return rn ? rb_entry(rn, struct str_node, rb_node) : NULL; +} +static inline struct str_node *strlist__next(struct str_node *sn) +{ + struct rb_node *rn; + if (!sn) + return NULL; + rn = rb_next(&sn->rb_node); + return rn ? rb_entry(rn, struct str_node, rb_node) : NULL; +} + +/** + * strlist_for_each - iterate over a strlist + * @pos: the &struct str_node to use as a loop cursor. + * @slist: the &struct strlist for loop. + */ +#define strlist__for_each(pos, slist) \ + for (pos = strlist__first(slist); pos; pos = strlist__next(pos)) + +/** + * strlist_for_each_safe - iterate over a strlist safe against removal of + * str_node + * @pos: the &struct str_node to use as a loop cursor. + * @n: another &struct str_node to use as temporary storage. + * @slist: the &struct strlist for loop. + */ +#define strlist__for_each_safe(pos, n, slist) \ + for (pos = strlist__first(slist), n = strlist__next(pos); pos;\ + pos = n, n = strlist__next(n)) + +int strlist__parse_list(struct strlist *slist, const char *s); +#endif /* __PERF_STRLIST_H */ diff --git a/tools/perf/util/svghelper.c b/tools/perf/util/svghelper.c new file mode 100644 index 00000000..96c86604 --- /dev/null +++ b/tools/perf/util/svghelper.c @@ -0,0 +1,501 @@ +/* + * svghelper.c - helper functions for outputting svg + * + * (C) Copyright 2009 Intel Corporation + * + * Authors: + * Arjan van de Ven <arjan@linux.intel.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 + * of the License. + */ + +#include <inttypes.h> +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <string.h> + +#include "svghelper.h" + +static u64 first_time, last_time; +static u64 turbo_frequency, max_freq; + + +#define SLOT_MULT 30.0 +#define SLOT_HEIGHT 25.0 + +int svg_page_width = 1000; + +#define MIN_TEXT_SIZE 0.01 + +static u64 total_height; +static FILE *svgfile; + +static double cpu2slot(int cpu) +{ + return 2 * cpu + 1; +} + +static double cpu2y(int cpu) +{ + return cpu2slot(cpu) * SLOT_MULT; +} + +static double time2pixels(u64 __time) +{ + double X; + + X = 1.0 * svg_page_width * (__time - first_time) / (last_time - first_time); + return X; +} + +/* + * Round text sizes so that the svg viewer only needs a discrete + * number of renderings of the font + */ +static double round_text_size(double size) +{ + int loop = 100; + double target = 10.0; + + if (size >= 10.0) + return size; + while (loop--) { + if (size >= target) + return target; + target = target / 2.0; + } + return size; +} + +void open_svg(const char *filename, int cpus, int rows, u64 start, u64 end) +{ + int new_width; + + svgfile = fopen(filename, "w"); + if (!svgfile) { + fprintf(stderr, "Cannot open %s for output\n", filename); + return; + } + first_time = start; + first_time = first_time / 100000000 * 100000000; + last_time = end; + + /* + * if the recording is short, we default to a width of 1000, but + * for longer recordings we want at least 200 units of width per second + */ + new_width = (last_time - first_time) / 5000000; + + if (new_width > svg_page_width) + svg_page_width = new_width; + + total_height = (1 + rows + cpu2slot(cpus)) * SLOT_MULT; + fprintf(svgfile, "<?xml version=\"1.0\" standalone=\"no\"?> \n"); + fprintf(svgfile, "<svg width=\"%i\" height=\"%" PRIu64 "\" version=\"1.1\" xmlns=\"http://www.w3.org/2000/svg\">\n", svg_page_width, total_height); + + fprintf(svgfile, "<defs>\n <style type=\"text/css\">\n <![CDATA[\n"); + + fprintf(svgfile, " rect { stroke-width: 1; }\n"); + fprintf(svgfile, " rect.process { fill:rgb(180,180,180); fill-opacity:0.9; stroke-width:1; stroke:rgb( 0, 0, 0); } \n"); + fprintf(svgfile, " rect.process2 { fill:rgb(180,180,180); fill-opacity:0.9; stroke-width:0; stroke:rgb( 0, 0, 0); } \n"); + fprintf(svgfile, " rect.sample { fill:rgb( 0, 0,255); fill-opacity:0.8; stroke-width:0; stroke:rgb( 0, 0, 0); } \n"); + fprintf(svgfile, " rect.blocked { fill:rgb(255, 0, 0); fill-opacity:0.5; stroke-width:0; stroke:rgb( 0, 0, 0); } \n"); + fprintf(svgfile, " rect.waiting { fill:rgb(224,214, 0); fill-opacity:0.8; stroke-width:0; stroke:rgb( 0, 0, 0); } \n"); + fprintf(svgfile, " rect.WAITING { fill:rgb(255,214, 48); fill-opacity:0.6; stroke-width:0; stroke:rgb( 0, 0, 0); } \n"); + fprintf(svgfile, " rect.cpu { fill:rgb(192,192,192); fill-opacity:0.2; stroke-width:0.5; stroke:rgb(128,128,128); } \n"); + fprintf(svgfile, " rect.pstate { fill:rgb(128,128,128); fill-opacity:0.8; stroke-width:0; } \n"); + fprintf(svgfile, " rect.c1 { fill:rgb(255,214,214); fill-opacity:0.5; stroke-width:0; } \n"); + fprintf(svgfile, " rect.c2 { fill:rgb(255,172,172); fill-opacity:0.5; stroke-width:0; } \n"); + fprintf(svgfile, " rect.c3 { fill:rgb(255,130,130); fill-opacity:0.5; stroke-width:0; } \n"); + fprintf(svgfile, " rect.c4 { fill:rgb(255, 88, 88); fill-opacity:0.5; stroke-width:0; } \n"); + fprintf(svgfile, " rect.c5 { fill:rgb(255, 44, 44); fill-opacity:0.5; stroke-width:0; } \n"); + fprintf(svgfile, " rect.c6 { fill:rgb(255, 0, 0); fill-opacity:0.5; stroke-width:0; } \n"); + fprintf(svgfile, " line.pstate { stroke:rgb(255,255, 0); stroke-opacity:0.8; stroke-width:2; } \n"); + + fprintf(svgfile, " ]]>\n </style>\n</defs>\n"); +} + +void svg_box(int Yslot, u64 start, u64 end, const char *type) +{ + if (!svgfile) + return; + + fprintf(svgfile, "<rect x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\" class=\"%s\"/>\n", + time2pixels(start), time2pixels(end)-time2pixels(start), Yslot * SLOT_MULT, SLOT_HEIGHT, type); +} + +void svg_sample(int Yslot, int cpu, u64 start, u64 end) +{ + double text_size; + if (!svgfile) + return; + + fprintf(svgfile, "<rect x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\" class=\"sample\"/>\n", + time2pixels(start), time2pixels(end)-time2pixels(start), Yslot * SLOT_MULT, SLOT_HEIGHT); + + text_size = (time2pixels(end)-time2pixels(start)); + if (cpu > 9) + text_size = text_size/2; + if (text_size > 1.25) + text_size = 1.25; + text_size = round_text_size(text_size); + + if (text_size > MIN_TEXT_SIZE) + fprintf(svgfile, "<text x=\"%1.8f\" y=\"%1.8f\" font-size=\"%1.8fpt\">%i</text>\n", + time2pixels(start), Yslot * SLOT_MULT + SLOT_HEIGHT - 1, text_size, cpu + 1); + +} + +static char *time_to_string(u64 duration) +{ + static char text[80]; + + text[0] = 0; + + if (duration < 1000) /* less than 1 usec */ + return text; + + if (duration < 1000 * 1000) { /* less than 1 msec */ + sprintf(text, "%4.1f us", duration / 1000.0); + return text; + } + sprintf(text, "%4.1f ms", duration / 1000.0 / 1000); + + return text; +} + +void svg_waiting(int Yslot, u64 start, u64 end) +{ + char *text; + const char *style; + double font_size; + + if (!svgfile) + return; + + style = "waiting"; + + if (end-start > 10 * 1000000) /* 10 msec */ + style = "WAITING"; + + text = time_to_string(end-start); + + font_size = 1.0 * (time2pixels(end)-time2pixels(start)); + + if (font_size > 3) + font_size = 3; + + font_size = round_text_size(font_size); + + fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\">\n", time2pixels(start), Yslot * SLOT_MULT); + fprintf(svgfile, "<rect x=\"0\" width=\"%4.8f\" y=\"0\" height=\"%4.1f\" class=\"%s\"/>\n", + time2pixels(end)-time2pixels(start), SLOT_HEIGHT, style); + if (font_size > MIN_TEXT_SIZE) + fprintf(svgfile, "<text transform=\"rotate(90)\" font-size=\"%1.8fpt\"> %s</text>\n", + font_size, text); + fprintf(svgfile, "</g>\n"); +} + +static char *cpu_model(void) +{ + static char cpu_m[255]; + char buf[256]; + FILE *file; + + cpu_m[0] = 0; + /* CPU type */ + file = fopen("/proc/cpuinfo", "r"); + if (file) { + while (fgets(buf, 255, file)) { + if (strstr(buf, "model name")) { + strncpy(cpu_m, &buf[13], 255); + break; + } + } + fclose(file); + } + + /* CPU type */ + file = fopen("/sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies", "r"); + if (file) { + while (fgets(buf, 255, file)) { + unsigned int freq; + freq = strtoull(buf, NULL, 10); + if (freq > max_freq) + max_freq = freq; + } + fclose(file); + } + return cpu_m; +} + +void svg_cpu_box(int cpu, u64 __max_freq, u64 __turbo_freq) +{ + char cpu_string[80]; + if (!svgfile) + return; + + max_freq = __max_freq; + turbo_frequency = __turbo_freq; + + fprintf(svgfile, "<rect x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\" class=\"cpu\"/>\n", + time2pixels(first_time), + time2pixels(last_time)-time2pixels(first_time), + cpu2y(cpu), SLOT_MULT+SLOT_HEIGHT); + + sprintf(cpu_string, "CPU %i", (int)cpu+1); + fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\">%s</text>\n", + 10+time2pixels(first_time), cpu2y(cpu) + SLOT_HEIGHT/2, cpu_string); + + fprintf(svgfile, "<text transform=\"translate(%4.8f,%4.8f)\" font-size=\"1.25pt\">%s</text>\n", + 10+time2pixels(first_time), cpu2y(cpu) + SLOT_MULT + SLOT_HEIGHT - 4, cpu_model()); +} + +void svg_process(int cpu, u64 start, u64 end, const char *type, const char *name) +{ + double width; + + if (!svgfile) + return; + + + fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\">\n", time2pixels(start), cpu2y(cpu)); + fprintf(svgfile, "<rect x=\"0\" width=\"%4.8f\" y=\"0\" height=\"%4.1f\" class=\"%s\"/>\n", + time2pixels(end)-time2pixels(start), SLOT_MULT+SLOT_HEIGHT, type); + width = time2pixels(end)-time2pixels(start); + if (width > 6) + width = 6; + + width = round_text_size(width); + + if (width > MIN_TEXT_SIZE) + fprintf(svgfile, "<text transform=\"rotate(90)\" font-size=\"%3.8fpt\">%s</text>\n", + width, name); + + fprintf(svgfile, "</g>\n"); +} + +void svg_cstate(int cpu, u64 start, u64 end, int type) +{ + double width; + char style[128]; + + if (!svgfile) + return; + + + if (type > 6) + type = 6; + sprintf(style, "c%i", type); + + fprintf(svgfile, "<rect class=\"%s\" x=\"%4.8f\" width=\"%4.8f\" y=\"%4.1f\" height=\"%4.1f\"/>\n", + style, + time2pixels(start), time2pixels(end)-time2pixels(start), + cpu2y(cpu), SLOT_MULT+SLOT_HEIGHT); + + width = (time2pixels(end)-time2pixels(start))/2.0; + if (width > 6) + width = 6; + + width = round_text_size(width); + + if (width > MIN_TEXT_SIZE) + fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\" font-size=\"%3.8fpt\">C%i</text>\n", + time2pixels(start), cpu2y(cpu)+width, width, type); +} + +static char *HzToHuman(unsigned long hz) +{ + static char buffer[1024]; + unsigned long long Hz; + + memset(buffer, 0, 1024); + + Hz = hz; + + /* default: just put the Number in */ + sprintf(buffer, "%9lli", Hz); + + if (Hz > 1000) + sprintf(buffer, " %6lli Mhz", (Hz+500)/1000); + + if (Hz > 1500000) + sprintf(buffer, " %6.2f Ghz", (Hz+5000.0)/1000000); + + if (Hz == turbo_frequency) + sprintf(buffer, "Turbo"); + + return buffer; +} + +void svg_pstate(int cpu, u64 start, u64 end, u64 freq) +{ + double height = 0; + + if (!svgfile) + return; + + if (max_freq) + height = freq * 1.0 / max_freq * (SLOT_HEIGHT + SLOT_MULT); + height = 1 + cpu2y(cpu) + SLOT_MULT + SLOT_HEIGHT - height; + fprintf(svgfile, "<line x1=\"%4.8f\" x2=\"%4.8f\" y1=\"%4.1f\" y2=\"%4.1f\" class=\"pstate\"/>\n", + time2pixels(start), time2pixels(end), height, height); + fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\" font-size=\"0.25pt\">%s</text>\n", + time2pixels(start), height+0.9, HzToHuman(freq)); + +} + + +void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2) +{ + double height; + + if (!svgfile) + return; + + + if (row1 < row2) { + if (row1) { + fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n", + time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT, time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/32); + if (desc2) + fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s ></text></g>\n", + time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT + SLOT_HEIGHT/48, desc2); + } + if (row2) { + fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n", + time2pixels(start), row2 * SLOT_MULT - SLOT_MULT/32, time2pixels(start), row2 * SLOT_MULT); + if (desc1) + fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s ></text></g>\n", + time2pixels(start), row2 * SLOT_MULT - SLOT_MULT/32, desc1); + } + } else { + if (row2) { + fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n", + time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT, time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/32); + if (desc1) + fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s <</text></g>\n", + time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/48, desc1); + } + if (row1) { + fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n", + time2pixels(start), row1 * SLOT_MULT - SLOT_MULT/32, time2pixels(start), row1 * SLOT_MULT); + if (desc2) + fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s <</text></g>\n", + time2pixels(start), row1 * SLOT_MULT - SLOT_HEIGHT/32, desc2); + } + } + height = row1 * SLOT_MULT; + if (row2 > row1) + height += SLOT_HEIGHT; + if (row1) + fprintf(svgfile, "<circle cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\" style=\"fill:rgb(32,255,32)\"/>\n", + time2pixels(start), height); +} + +void svg_wakeline(u64 start, int row1, int row2) +{ + double height; + + if (!svgfile) + return; + + + if (row1 < row2) + fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n", + time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT, time2pixels(start), row2 * SLOT_MULT); + else + fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%4.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n", + time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT, time2pixels(start), row1 * SLOT_MULT); + + height = row1 * SLOT_MULT; + if (row2 > row1) + height += SLOT_HEIGHT; + fprintf(svgfile, "<circle cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\" style=\"fill:rgb(32,255,32)\"/>\n", + time2pixels(start), height); +} + +void svg_interrupt(u64 start, int row) +{ + if (!svgfile) + return; + + fprintf(svgfile, "<circle cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\" style=\"fill:rgb(255,128,128)\"/>\n", + time2pixels(start), row * SLOT_MULT); + fprintf(svgfile, "<circle cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\" style=\"fill:rgb(255,128,128)\"/>\n", + time2pixels(start), row * SLOT_MULT + SLOT_HEIGHT); +} + +void svg_text(int Yslot, u64 start, const char *text) +{ + if (!svgfile) + return; + + fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\">%s</text>\n", + time2pixels(start), Yslot * SLOT_MULT+SLOT_HEIGHT/2, text); +} + +static void svg_legenda_box(int X, const char *text, const char *style) +{ + double boxsize; + boxsize = SLOT_HEIGHT / 2; + + fprintf(svgfile, "<rect x=\"%i\" width=\"%4.8f\" y=\"0\" height=\"%4.1f\" class=\"%s\"/>\n", + X, boxsize, boxsize, style); + fprintf(svgfile, "<text transform=\"translate(%4.8f, %4.8f)\" font-size=\"%4.8fpt\">%s</text>\n", + X + boxsize + 5, boxsize, 0.8 * boxsize, text); +} + +void svg_legenda(void) +{ + if (!svgfile) + return; + + svg_legenda_box(0, "Running", "sample"); + svg_legenda_box(100, "Idle","c1"); + svg_legenda_box(200, "Deeper Idle", "c3"); + svg_legenda_box(350, "Deepest Idle", "c6"); + svg_legenda_box(550, "Sleeping", "process2"); + svg_legenda_box(650, "Waiting for cpu", "waiting"); + svg_legenda_box(800, "Blocked on IO", "blocked"); +} + +void svg_time_grid(void) +{ + u64 i; + + if (!svgfile) + return; + + i = first_time; + while (i < last_time) { + int color = 220; + double thickness = 0.075; + if ((i % 100000000) == 0) { + thickness = 0.5; + color = 192; + } + if ((i % 1000000000) == 0) { + thickness = 2.0; + color = 128; + } + + fprintf(svgfile, "<line x1=\"%4.8f\" y1=\"%4.2f\" x2=\"%4.8f\" y2=\"%" PRIu64 "\" style=\"stroke:rgb(%i,%i,%i);stroke-width:%1.3f\"/>\n", + time2pixels(i), SLOT_MULT/2, time2pixels(i), total_height, color, color, color, thickness); + + i += 10000000; + } +} + +void svg_close(void) +{ + if (svgfile) { + fprintf(svgfile, "</svg>\n"); + fclose(svgfile); + svgfile = NULL; + } +} diff --git a/tools/perf/util/svghelper.h b/tools/perf/util/svghelper.h new file mode 100644 index 00000000..e0781989 --- /dev/null +++ b/tools/perf/util/svghelper.h @@ -0,0 +1,28 @@ +#ifndef __PERF_SVGHELPER_H +#define __PERF_SVGHELPER_H + +#include "types.h" + +extern void open_svg(const char *filename, int cpus, int rows, u64 start, u64 end); +extern void svg_box(int Yslot, u64 start, u64 end, const char *type); +extern void svg_sample(int Yslot, int cpu, u64 start, u64 end); +extern void svg_waiting(int Yslot, u64 start, u64 end); +extern void svg_cpu_box(int cpu, u64 max_frequency, u64 turbo_frequency); + + +extern void svg_process(int cpu, u64 start, u64 end, const char *type, const char *name); +extern void svg_cstate(int cpu, u64 start, u64 end, int type); +extern void svg_pstate(int cpu, u64 start, u64 end, u64 freq); + + +extern void svg_time_grid(void); +extern void svg_legenda(void); +extern void svg_wakeline(u64 start, int row1, int row2); +extern void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2); +extern void svg_interrupt(u64 start, int row); +extern void svg_text(int Yslot, u64 start, const char *text); +extern void svg_close(void); + +extern int svg_page_width; + +#endif /* __PERF_SVGHELPER_H */ diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c new file mode 100644 index 00000000..54efcb56 --- /dev/null +++ b/tools/perf/util/symbol-elf.c @@ -0,0 +1,849 @@ +#include <fcntl.h> +#include <stdio.h> +#include <errno.h> +#include <string.h> +#include <unistd.h> +#include <inttypes.h> + +#include "symbol.h" +#include "debug.h" + +#ifndef NT_GNU_BUILD_ID +#define NT_GNU_BUILD_ID 3 +#endif + +/** + * elf_symtab__for_each_symbol - iterate thru all the symbols + * + * @syms: struct elf_symtab instance to iterate + * @idx: uint32_t idx + * @sym: GElf_Sym iterator + */ +#define elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) \ + for (idx = 0, gelf_getsym(syms, idx, &sym);\ + idx < nr_syms; \ + idx++, gelf_getsym(syms, idx, &sym)) + +static inline uint8_t elf_sym__type(const GElf_Sym *sym) +{ + 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; +} + +static inline bool elf_sym__is_object(const GElf_Sym *sym) +{ + return elf_sym__type(sym) == STT_OBJECT && + sym->st_name != 0 && + sym->st_shndx != SHN_UNDEF; +} + +static inline int elf_sym__is_label(const GElf_Sym *sym) +{ + return elf_sym__type(sym) == STT_NOTYPE && + sym->st_name != 0 && + sym->st_shndx != SHN_UNDEF && + sym->st_shndx != SHN_ABS; +} + +static bool elf_sym__is_a(GElf_Sym *sym, enum map_type type) +{ + switch (type) { + case MAP__FUNCTION: + return elf_sym__is_function(sym); + case MAP__VARIABLE: + return elf_sym__is_object(sym); + default: + return false; + } +} + +static inline const char *elf_sym__name(const GElf_Sym *sym, + const Elf_Data *symstrs) +{ + return symstrs->d_buf + sym->st_name; +} + +static inline const char *elf_sec__name(const GElf_Shdr *shdr, + const Elf_Data *secstrs) +{ + return secstrs->d_buf + shdr->sh_name; +} + +static inline int elf_sec__is_text(const GElf_Shdr *shdr, + const Elf_Data *secstrs) +{ + return strstr(elf_sec__name(shdr, secstrs), "text") != NULL; +} + +static inline bool elf_sec__is_data(const GElf_Shdr *shdr, + const Elf_Data *secstrs) +{ + return strstr(elf_sec__name(shdr, secstrs), "data") != NULL; +} + +static bool elf_sec__is_a(GElf_Shdr *shdr, Elf_Data *secstrs, + enum map_type type) +{ + switch (type) { + case MAP__FUNCTION: + return elf_sec__is_text(shdr, secstrs); + case MAP__VARIABLE: + return elf_sec__is_data(shdr, secstrs); + default: + return false; + } +} + +static size_t elf_addr_to_index(Elf *elf, GElf_Addr addr) +{ + Elf_Scn *sec = NULL; + GElf_Shdr shdr; + size_t cnt = 1; + + while ((sec = elf_nextscn(elf, sec)) != NULL) { + gelf_getshdr(sec, &shdr); + + if ((addr >= shdr.sh_addr) && + (addr < (shdr.sh_addr + shdr.sh_size))) + return cnt; + + ++cnt; + } + + return -1; +} + +static Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep, + GElf_Shdr *shp, const char *name, + size_t *idx) +{ + Elf_Scn *sec = NULL; + size_t cnt = 1; + + /* Elf is corrupted/truncated, avoid calling elf_strptr. */ + if (!elf_rawdata(elf_getscn(elf, ep->e_shstrndx), NULL)) + return NULL; + + 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 (idx) + *idx = 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)) + +/* + * We need to check if we have a .dynsym, so that we can handle the + * .plt, synthesizing its symbols, that aren't on the symtabs (be it + * .dynsym or .symtab). + * And always look at the original dso, not at debuginfo packages, that + * have the PLT data stripped out (shdr_rel_plt.sh_type == SHT_NOBITS). + */ +int dso__synthesize_plt_symbols(struct dso *dso, struct symsrc *ss, struct map *map, + symbol_filter_t filter) +{ + uint32_t nr_rel_entries, idx; + GElf_Sym sym; + u64 plt_offset; + GElf_Shdr shdr_plt; + struct symbol *f; + GElf_Shdr shdr_rel_plt, shdr_dynsym; + Elf_Data *reldata, *syms, *symstrs; + Elf_Scn *scn_plt_rel, *scn_symstrs, *scn_dynsym; + size_t dynsym_idx; + GElf_Ehdr ehdr; + char sympltname[1024]; + Elf *elf; + int nr = 0, symidx, err = 0; + + if (!ss->dynsym) + return 0; + + elf = ss->elf; + ehdr = ss->ehdr; + + scn_dynsym = ss->dynsym; + shdr_dynsym = ss->dynshdr; + dynsym_idx = ss->dynsym_idx; + + if (scn_dynsym == NULL) + goto out_elf_end; + + scn_plt_rel = elf_section_by_name(elf, &ehdr, &shdr_rel_plt, + ".rela.plt", NULL); + if (scn_plt_rel == NULL) { + scn_plt_rel = elf_section_by_name(elf, &ehdr, &shdr_rel_plt, + ".rel.plt", NULL); + if (scn_plt_rel == NULL) + goto out_elf_end; + } + + err = -1; + + if (shdr_rel_plt.sh_link != dynsym_idx) + goto out_elf_end; + + if (elf_section_by_name(elf, &ehdr, &shdr_plt, ".plt", NULL) == NULL) + goto out_elf_end; + + /* + * Fetch the relocation section to find the idxes to the GOT + * and the symbols in the .dynsym they refer to. + */ + reldata = elf_getdata(scn_plt_rel, NULL); + if (reldata == NULL) + goto out_elf_end; + + syms = elf_getdata(scn_dynsym, NULL); + if (syms == NULL) + goto out_elf_end; + + scn_symstrs = elf_getscn(elf, shdr_dynsym.sh_link); + if (scn_symstrs == NULL) + goto out_elf_end; + + symstrs = elf_getdata(scn_symstrs, NULL); + if (symstrs == NULL) + goto out_elf_end; + + if (symstrs->d_size == 0) + goto out_elf_end; + + 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, + STB_GLOBAL, sympltname); + if (!f) + goto out_elf_end; + + if (filter && filter(map, f)) + symbol__delete(f); + else { + symbols__insert(&dso->symbols[map->type], 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, + STB_GLOBAL, sympltname); + if (!f) + goto out_elf_end; + + if (filter && filter(map, f)) + symbol__delete(f); + else { + symbols__insert(&dso->symbols[map->type], f); + ++nr; + } + } + } + + err = 0; +out_elf_end: + if (err == 0) + return nr; + pr_debug("%s: problems reading %s PLT info.\n", + __func__, dso->long_name); + return 0; +} + +/* + * Align offset to 4 bytes as needed for note name and descriptor data. + */ +#define NOTE_ALIGN(n) (((n) + 3) & -4U) + +static int elf_read_build_id(Elf *elf, void *bf, size_t size) +{ + int err = -1; + GElf_Ehdr ehdr; + GElf_Shdr shdr; + Elf_Data *data; + Elf_Scn *sec; + Elf_Kind ek; + void *ptr; + + if (size < BUILD_ID_SIZE) + goto out; + + ek = elf_kind(elf); + if (ek != ELF_K_ELF) + goto out; + + if (gelf_getehdr(elf, &ehdr) == NULL) { + pr_err("%s: cannot get elf header.\n", __func__); + goto out; + } + + /* + * Check following sections for notes: + * '.note.gnu.build-id' + * '.notes' + * '.note' (VDSO specific) + */ + do { + sec = elf_section_by_name(elf, &ehdr, &shdr, + ".note.gnu.build-id", NULL); + if (sec) + break; + + sec = elf_section_by_name(elf, &ehdr, &shdr, + ".notes", NULL); + if (sec) + break; + + sec = elf_section_by_name(elf, &ehdr, &shdr, + ".note", NULL); + if (sec) + break; + + return err; + + } while (0); + + data = elf_getdata(sec, NULL); + if (data == NULL) + goto out; + + ptr = data->d_buf; + while (ptr < (data->d_buf + data->d_size)) { + GElf_Nhdr *nhdr = ptr; + size_t namesz = NOTE_ALIGN(nhdr->n_namesz), + descsz = NOTE_ALIGN(nhdr->n_descsz); + const char *name; + + ptr += sizeof(*nhdr); + name = ptr; + ptr += namesz; + if (nhdr->n_type == NT_GNU_BUILD_ID && + nhdr->n_namesz == sizeof("GNU")) { + if (memcmp(name, "GNU", sizeof("GNU")) == 0) { + size_t sz = min(size, descsz); + memcpy(bf, ptr, sz); + memset(bf + sz, 0, size - sz); + err = descsz; + break; + } + } + ptr += descsz; + } + +out: + return err; +} + +int filename__read_build_id(const char *filename, void *bf, size_t size) +{ + int fd, err = -1; + Elf *elf; + + if (size < BUILD_ID_SIZE) + goto out; + + fd = open(filename, O_RDONLY); + if (fd < 0) + goto out; + + elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL); + if (elf == NULL) { + pr_debug2("%s: cannot read %s ELF file.\n", __func__, filename); + goto out_close; + } + + err = elf_read_build_id(elf, bf, size); + + elf_end(elf); +out_close: + close(fd); +out: + return err; +} + +int sysfs__read_build_id(const char *filename, void *build_id, size_t size) +{ + int fd, err = -1; + + if (size < BUILD_ID_SIZE) + goto out; + + fd = open(filename, O_RDONLY); + if (fd < 0) + goto out; + + while (1) { + char bf[BUFSIZ]; + GElf_Nhdr nhdr; + size_t namesz, descsz; + + if (read(fd, &nhdr, sizeof(nhdr)) != sizeof(nhdr)) + break; + + namesz = NOTE_ALIGN(nhdr.n_namesz); + descsz = NOTE_ALIGN(nhdr.n_descsz); + if (nhdr.n_type == NT_GNU_BUILD_ID && + nhdr.n_namesz == sizeof("GNU")) { + if (read(fd, bf, namesz) != (ssize_t)namesz) + break; + if (memcmp(bf, "GNU", sizeof("GNU")) == 0) { + size_t sz = min(descsz, size); + if (read(fd, build_id, sz) == (ssize_t)sz) { + memset(build_id + sz, 0, size - sz); + err = 0; + break; + } + } else if (read(fd, bf, descsz) != (ssize_t)descsz) + break; + } else { + int n = namesz + descsz; + if (read(fd, bf, n) != n) + break; + } + } + close(fd); +out: + return err; +} + +int filename__read_debuglink(const char *filename, char *debuglink, + size_t size) +{ + int fd, err = -1; + Elf *elf; + GElf_Ehdr ehdr; + GElf_Shdr shdr; + Elf_Data *data; + Elf_Scn *sec; + Elf_Kind ek; + + fd = open(filename, O_RDONLY); + if (fd < 0) + goto out; + + elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL); + if (elf == NULL) { + pr_debug2("%s: cannot read %s ELF file.\n", __func__, filename); + goto out_close; + } + + ek = elf_kind(elf); + if (ek != ELF_K_ELF) + goto out_close; + + if (gelf_getehdr(elf, &ehdr) == NULL) { + pr_err("%s: cannot get elf header.\n", __func__); + goto out_close; + } + + sec = elf_section_by_name(elf, &ehdr, &shdr, + ".gnu_debuglink", NULL); + if (sec == NULL) + goto out_close; + + data = elf_getdata(sec, NULL); + if (data == NULL) + goto out_close; + + /* the start of this section is a zero-terminated string */ + strncpy(debuglink, data->d_buf, size); + + elf_end(elf); + +out_close: + close(fd); +out: + return err; +} + +static int dso__swap_init(struct dso *dso, unsigned char eidata) +{ + static unsigned int const endian = 1; + + dso->needs_swap = DSO_SWAP__NO; + + switch (eidata) { + case ELFDATA2LSB: + /* We are big endian, DSO is little endian. */ + if (*(unsigned char const *)&endian != 1) + dso->needs_swap = DSO_SWAP__YES; + break; + + case ELFDATA2MSB: + /* We are little endian, DSO is big endian. */ + if (*(unsigned char const *)&endian != 0) + dso->needs_swap = DSO_SWAP__YES; + break; + + default: + pr_err("unrecognized DSO data encoding %d\n", eidata); + return -EINVAL; + } + + return 0; +} + +bool symsrc__possibly_runtime(struct symsrc *ss) +{ + return ss->dynsym || ss->opdsec; +} + +bool symsrc__has_symtab(struct symsrc *ss) +{ + return ss->symtab != NULL; +} + +void symsrc__destroy(struct symsrc *ss) +{ + free(ss->name); + elf_end(ss->elf); + close(ss->fd); +} + +int symsrc__init(struct symsrc *ss, struct dso *dso, const char *name, + enum dso_binary_type type) +{ + int err = -1; + GElf_Ehdr ehdr; + Elf *elf; + int fd; + + fd = open(name, O_RDONLY); + if (fd < 0) + return -1; + + elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL); + if (elf == NULL) { + pr_debug("%s: cannot read %s ELF file.\n", __func__, name); + goto out_close; + } + + if (gelf_getehdr(elf, &ehdr) == NULL) { + pr_debug("%s: cannot get elf header.\n", __func__); + goto out_elf_end; + } + + if (dso__swap_init(dso, ehdr.e_ident[EI_DATA])) + goto out_elf_end; + + /* Always reject images with a mismatched build-id: */ + if (dso->has_build_id) { + u8 build_id[BUILD_ID_SIZE]; + + if (elf_read_build_id(elf, build_id, BUILD_ID_SIZE) < 0) + goto out_elf_end; + + if (!dso__build_id_equal(dso, build_id)) + goto out_elf_end; + } + + ss->symtab = elf_section_by_name(elf, &ehdr, &ss->symshdr, ".symtab", + NULL); + if (ss->symshdr.sh_type != SHT_SYMTAB) + ss->symtab = NULL; + + ss->dynsym_idx = 0; + ss->dynsym = elf_section_by_name(elf, &ehdr, &ss->dynshdr, ".dynsym", + &ss->dynsym_idx); + if (ss->dynshdr.sh_type != SHT_DYNSYM) + ss->dynsym = NULL; + + ss->opdidx = 0; + ss->opdsec = elf_section_by_name(elf, &ehdr, &ss->opdshdr, ".opd", + &ss->opdidx); + if (ss->opdshdr.sh_type != SHT_PROGBITS) + ss->opdsec = NULL; + + if (dso->kernel == DSO_TYPE_USER) { + GElf_Shdr shdr; + ss->adjust_symbols = (ehdr.e_type == ET_EXEC || + elf_section_by_name(elf, &ehdr, &shdr, + ".gnu.prelink_undo", + NULL) != NULL); + } else { + ss->adjust_symbols = 0; + } + + ss->name = strdup(name); + if (!ss->name) + goto out_elf_end; + + ss->elf = elf; + ss->fd = fd; + ss->ehdr = ehdr; + ss->type = type; + + return 0; + +out_elf_end: + elf_end(elf); +out_close: + close(fd); + return err; +} + +int dso__load_sym(struct dso *dso, struct map *map, + struct symsrc *syms_ss, struct symsrc *runtime_ss, + symbol_filter_t filter, int kmodule) +{ + struct kmap *kmap = dso->kernel ? map__kmap(map) : NULL; + struct map *curr_map = map; + struct dso *curr_dso = dso; + Elf_Data *symstrs, *secstrs; + uint32_t nr_syms; + int err = -1; + uint32_t idx; + GElf_Ehdr ehdr; + GElf_Shdr shdr; + Elf_Data *syms, *opddata = NULL; + GElf_Sym sym; + Elf_Scn *sec, *sec_strndx; + Elf *elf; + int nr = 0; + + dso->symtab_type = syms_ss->type; + + if (!syms_ss->symtab) { + syms_ss->symtab = syms_ss->dynsym; + syms_ss->symshdr = syms_ss->dynshdr; + } + + elf = syms_ss->elf; + ehdr = syms_ss->ehdr; + sec = syms_ss->symtab; + shdr = syms_ss->symshdr; + + if (runtime_ss->opdsec) + opddata = elf_rawdata(runtime_ss->opdsec, NULL); + + 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; + + sec_strndx = elf_getscn(elf, ehdr.e_shstrndx); + if (sec_strndx == NULL) + goto out_elf_end; + + secstrs = elf_getdata(sec_strndx, NULL); + if (secstrs == NULL) + goto out_elf_end; + + nr_syms = shdr.sh_size / shdr.sh_entsize; + + memset(&sym, 0, sizeof(sym)); + dso->adjust_symbols = runtime_ss->adjust_symbols; + elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) { + struct symbol *f; + const char *elf_name = elf_sym__name(&sym, symstrs); + char *demangled = NULL; + int is_label = elf_sym__is_label(&sym); + const char *section_name; + bool used_opd = false; + + if (kmap && kmap->ref_reloc_sym && kmap->ref_reloc_sym->name && + strcmp(elf_name, kmap->ref_reloc_sym->name) == 0) + kmap->ref_reloc_sym->unrelocated_addr = sym.st_value; + + if (!is_label && !elf_sym__is_a(&sym, map->type)) + continue; + + /* Reject ARM ELF "mapping symbols": these aren't unique and + * don't identify functions, so will confuse the profile + * output: */ + if (ehdr.e_machine == EM_ARM) { + if (!strcmp(elf_name, "$a") || + !strcmp(elf_name, "$d") || + !strcmp(elf_name, "$t")) + continue; + } + + if (runtime_ss->opdsec && sym.st_shndx == runtime_ss->opdidx) { + u32 offset = sym.st_value - syms_ss->opdshdr.sh_addr; + u64 *opd = opddata->d_buf + offset; + sym.st_value = DSO__SWAP(dso, u64, *opd); + sym.st_shndx = elf_addr_to_index(runtime_ss->elf, + sym.st_value); + used_opd = true; + } + /* + * When loading symbols in a data mapping, ABS symbols (which + * has a value of SHN_ABS in its st_shndx) failed at + * elf_getscn(). And it marks the loading as a failure so + * already loaded symbols cannot be fixed up. + * + * I'm not sure what should be done. Just ignore them for now. + * - Namhyung Kim + */ + if (sym.st_shndx == SHN_ABS) + continue; + + sec = elf_getscn(runtime_ss->elf, sym.st_shndx); + if (!sec) + goto out_elf_end; + + gelf_getshdr(sec, &shdr); + + if (is_label && !elf_sec__is_a(&shdr, secstrs, map->type)) + continue; + + section_name = elf_sec__name(&shdr, secstrs); + + /* On ARM, symbols for thumb functions have 1 added to + * the symbol address as a flag - remove it */ + if ((ehdr.e_machine == EM_ARM) && + (map->type == MAP__FUNCTION) && + (sym.st_value & 1)) + --sym.st_value; + + if (dso->kernel != DSO_TYPE_USER || kmodule) { + char dso_name[PATH_MAX]; + + if (strcmp(section_name, + (curr_dso->short_name + + dso->short_name_len)) == 0) + goto new_symbol; + + if (strcmp(section_name, ".text") == 0) { + curr_map = map; + curr_dso = dso; + goto new_symbol; + } + + snprintf(dso_name, sizeof(dso_name), + "%s%s", dso->short_name, section_name); + + curr_map = map_groups__find_by_name(kmap->kmaps, map->type, dso_name); + if (curr_map == NULL) { + u64 start = sym.st_value; + + if (kmodule) + start += map->start + shdr.sh_offset; + + curr_dso = dso__new(dso_name); + if (curr_dso == NULL) + goto out_elf_end; + curr_dso->kernel = dso->kernel; + curr_dso->long_name = dso->long_name; + curr_dso->long_name_len = dso->long_name_len; + curr_map = map__new2(start, curr_dso, + map->type); + if (curr_map == NULL) { + dso__delete(curr_dso); + goto out_elf_end; + } + curr_map->map_ip = identity__map_ip; + curr_map->unmap_ip = identity__map_ip; + curr_dso->symtab_type = dso->symtab_type; + map_groups__insert(kmap->kmaps, curr_map); + dsos__add(&dso->node, curr_dso); + dso__set_loaded(curr_dso, map->type); + } else + curr_dso = curr_map->dso; + + goto new_symbol; + } + + if ((used_opd && runtime_ss->adjust_symbols) + || (!used_opd && syms_ss->adjust_symbols)) { + pr_debug4("%s: adjusting symbol: st_value: %#" PRIx64 " " + "sh_addr: %#" PRIx64 " sh_offset: %#" PRIx64 "\n", __func__, + (u64)sym.st_value, (u64)shdr.sh_addr, + (u64)shdr.sh_offset); + sym.st_value -= shdr.sh_addr - shdr.sh_offset; + } + /* + * We need to figure out if the object was created from C++ sources + * DWARF DW_compile_unit has this, but we don't always have access + * to it... + */ + demangled = bfd_demangle(NULL, elf_name, DMGL_PARAMS | DMGL_ANSI); + if (demangled != NULL) + elf_name = demangled; +new_symbol: + f = symbol__new(sym.st_value, sym.st_size, + GELF_ST_BIND(sym.st_info), elf_name); + free(demangled); + if (!f) + goto out_elf_end; + + if (filter && filter(curr_map, f)) + symbol__delete(f); + else { + symbols__insert(&curr_dso->symbols[curr_map->type], f); + nr++; + } + } + + /* + * For misannotated, zeroed, ASM function sizes. + */ + if (nr > 0) { + symbols__fixup_duplicate(&dso->symbols[map->type]); + symbols__fixup_end(&dso->symbols[map->type]); + if (kmap) { + /* + * We need to fixup this here too because we create new + * maps here, for things like vsyscall sections. + */ + __map_groups__fixup_end(kmap->kmaps, map->type); + } + } + err = nr; +out_elf_end: + return err; +} + +void symbol__elf_init(void) +{ + elf_version(EV_CURRENT); +} diff --git a/tools/perf/util/symbol-minimal.c b/tools/perf/util/symbol-minimal.c new file mode 100644 index 00000000..a7390cde --- /dev/null +++ b/tools/perf/util/symbol-minimal.c @@ -0,0 +1,306 @@ +#include "symbol.h" + +#include <stdio.h> +#include <fcntl.h> +#include <string.h> +#include <byteswap.h> +#include <sys/stat.h> + + +static bool check_need_swap(int file_endian) +{ + const int data = 1; + u8 *check = (u8 *)&data; + int host_endian; + + if (check[0] == 1) + host_endian = ELFDATA2LSB; + else + host_endian = ELFDATA2MSB; + + return host_endian != file_endian; +} + +#define NOTE_ALIGN(sz) (((sz) + 3) & ~3) + +#define NT_GNU_BUILD_ID 3 + +static int read_build_id(void *note_data, size_t note_len, void *bf, + size_t size, bool need_swap) +{ + struct { + u32 n_namesz; + u32 n_descsz; + u32 n_type; + } *nhdr; + void *ptr; + + ptr = note_data; + while (ptr < (note_data + note_len)) { + const char *name; + size_t namesz, descsz; + + nhdr = ptr; + if (need_swap) { + nhdr->n_namesz = bswap_32(nhdr->n_namesz); + nhdr->n_descsz = bswap_32(nhdr->n_descsz); + nhdr->n_type = bswap_32(nhdr->n_type); + } + + namesz = NOTE_ALIGN(nhdr->n_namesz); + descsz = NOTE_ALIGN(nhdr->n_descsz); + + ptr += sizeof(*nhdr); + name = ptr; + ptr += namesz; + if (nhdr->n_type == NT_GNU_BUILD_ID && + nhdr->n_namesz == sizeof("GNU")) { + if (memcmp(name, "GNU", sizeof("GNU")) == 0) { + size_t sz = min(size, descsz); + memcpy(bf, ptr, sz); + memset(bf + sz, 0, size - sz); + return 0; + } + } + ptr += descsz; + } + + return -1; +} + +int filename__read_debuglink(const char *filename __maybe_unused, + char *debuglink __maybe_unused, + size_t size __maybe_unused) +{ + return -1; +} + +/* + * Just try PT_NOTE header otherwise fails + */ +int filename__read_build_id(const char *filename, void *bf, size_t size) +{ + FILE *fp; + int ret = -1; + bool need_swap = false; + u8 e_ident[EI_NIDENT]; + size_t buf_size; + void *buf; + int i; + + fp = fopen(filename, "r"); + if (fp == NULL) + return -1; + + if (fread(e_ident, sizeof(e_ident), 1, fp) != 1) + goto out; + + if (memcmp(e_ident, ELFMAG, SELFMAG) || + e_ident[EI_VERSION] != EV_CURRENT) + goto out; + + need_swap = check_need_swap(e_ident[EI_DATA]); + + /* for simplicity */ + fseek(fp, 0, SEEK_SET); + + if (e_ident[EI_CLASS] == ELFCLASS32) { + Elf32_Ehdr ehdr; + Elf32_Phdr *phdr; + + if (fread(&ehdr, sizeof(ehdr), 1, fp) != 1) + goto out; + + if (need_swap) { + ehdr.e_phoff = bswap_32(ehdr.e_phoff); + ehdr.e_phentsize = bswap_16(ehdr.e_phentsize); + ehdr.e_phnum = bswap_16(ehdr.e_phnum); + } + + buf_size = ehdr.e_phentsize * ehdr.e_phnum; + buf = malloc(buf_size); + if (buf == NULL) + goto out; + + fseek(fp, ehdr.e_phoff, SEEK_SET); + if (fread(buf, buf_size, 1, fp) != 1) + goto out_free; + + for (i = 0, phdr = buf; i < ehdr.e_phnum; i++, phdr++) { + void *tmp; + + if (need_swap) { + phdr->p_type = bswap_32(phdr->p_type); + phdr->p_offset = bswap_32(phdr->p_offset); + phdr->p_filesz = bswap_32(phdr->p_filesz); + } + + if (phdr->p_type != PT_NOTE) + continue; + + buf_size = phdr->p_filesz; + tmp = realloc(buf, buf_size); + if (tmp == NULL) + goto out_free; + + buf = tmp; + fseek(fp, phdr->p_offset, SEEK_SET); + if (fread(buf, buf_size, 1, fp) != 1) + goto out_free; + + ret = read_build_id(buf, buf_size, bf, size, need_swap); + if (ret == 0) + ret = size; + break; + } + } else { + Elf64_Ehdr ehdr; + Elf64_Phdr *phdr; + + if (fread(&ehdr, sizeof(ehdr), 1, fp) != 1) + goto out; + + if (need_swap) { + ehdr.e_phoff = bswap_64(ehdr.e_phoff); + ehdr.e_phentsize = bswap_16(ehdr.e_phentsize); + ehdr.e_phnum = bswap_16(ehdr.e_phnum); + } + + buf_size = ehdr.e_phentsize * ehdr.e_phnum; + buf = malloc(buf_size); + if (buf == NULL) + goto out; + + fseek(fp, ehdr.e_phoff, SEEK_SET); + if (fread(buf, buf_size, 1, fp) != 1) + goto out_free; + + for (i = 0, phdr = buf; i < ehdr.e_phnum; i++, phdr++) { + void *tmp; + + if (need_swap) { + phdr->p_type = bswap_32(phdr->p_type); + phdr->p_offset = bswap_64(phdr->p_offset); + phdr->p_filesz = bswap_64(phdr->p_filesz); + } + + if (phdr->p_type != PT_NOTE) + continue; + + buf_size = phdr->p_filesz; + tmp = realloc(buf, buf_size); + if (tmp == NULL) + goto out_free; + + buf = tmp; + fseek(fp, phdr->p_offset, SEEK_SET); + if (fread(buf, buf_size, 1, fp) != 1) + goto out_free; + + ret = read_build_id(buf, buf_size, bf, size, need_swap); + if (ret == 0) + ret = size; + break; + } + } +out_free: + free(buf); +out: + fclose(fp); + return ret; +} + +int sysfs__read_build_id(const char *filename, void *build_id, size_t size) +{ + int fd; + int ret = -1; + struct stat stbuf; + size_t buf_size; + void *buf; + + fd = open(filename, O_RDONLY); + if (fd < 0) + return -1; + + if (fstat(fd, &stbuf) < 0) + goto out; + + buf_size = stbuf.st_size; + buf = malloc(buf_size); + if (buf == NULL) + goto out; + + if (read(fd, buf, buf_size) != (ssize_t) buf_size) + goto out_free; + + ret = read_build_id(buf, buf_size, build_id, size, false); +out_free: + free(buf); +out: + close(fd); + return ret; +} + +int symsrc__init(struct symsrc *ss, struct dso *dso __maybe_unused, + const char *name, + enum dso_binary_type type) +{ + int fd = open(name, O_RDONLY); + if (fd < 0) + return -1; + + ss->name = strdup(name); + if (!ss->name) + goto out_close; + + ss->type = type; + + return 0; +out_close: + close(fd); + return -1; +} + +bool symsrc__possibly_runtime(struct symsrc *ss __maybe_unused) +{ + /* Assume all sym sources could be a runtime image. */ + return true; +} + +bool symsrc__has_symtab(struct symsrc *ss __maybe_unused) +{ + return false; +} + +void symsrc__destroy(struct symsrc *ss) +{ + free(ss->name); + close(ss->fd); +} + +int dso__synthesize_plt_symbols(struct dso *dso __maybe_unused, + struct symsrc *ss __maybe_unused, + struct map *map __maybe_unused, + symbol_filter_t filter __maybe_unused) +{ + return 0; +} + +int dso__load_sym(struct dso *dso, struct map *map __maybe_unused, + struct symsrc *ss, + struct symsrc *runtime_ss __maybe_unused, + symbol_filter_t filter __maybe_unused, + int kmodule __maybe_unused) +{ + unsigned char *build_id[BUILD_ID_SIZE]; + + if (filename__read_build_id(ss->name, build_id, BUILD_ID_SIZE) > 0) { + dso__set_build_id(dso, build_id); + return 1; + } + return 0; +} + +void symbol__elf_init(void) +{ +} diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c new file mode 100644 index 00000000..e6432d85 --- /dev/null +++ b/tools/perf/util/symbol.c @@ -0,0 +1,1268 @@ +#include <dirent.h> +#include <errno.h> +#include <stdlib.h> +#include <stdio.h> +#include <string.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/param.h> +#include <fcntl.h> +#include <unistd.h> +#include <inttypes.h> +#include "build-id.h" +#include "util.h" +#include "debug.h" +#include "machine.h" +#include "symbol.h" +#include "strlist.h" + +#include <elf.h> +#include <limits.h> +#include <sys/utsname.h> + +#ifndef KSYM_NAME_LEN +#define KSYM_NAME_LEN 256 +#endif + +static int dso__load_kernel_sym(struct dso *dso, struct map *map, + symbol_filter_t filter); +static int dso__load_guest_kernel_sym(struct dso *dso, struct map *map, + symbol_filter_t filter); +int vmlinux_path__nr_entries; +char **vmlinux_path; + +struct symbol_conf symbol_conf = { + .exclude_other = true, + .use_modules = true, + .try_vmlinux_path = true, + .annotate_src = true, + .symfs = "", +}; + +static enum dso_binary_type binary_type_symtab[] = { + DSO_BINARY_TYPE__KALLSYMS, + DSO_BINARY_TYPE__GUEST_KALLSYMS, + DSO_BINARY_TYPE__JAVA_JIT, + DSO_BINARY_TYPE__DEBUGLINK, + DSO_BINARY_TYPE__BUILD_ID_CACHE, + DSO_BINARY_TYPE__FEDORA_DEBUGINFO, + DSO_BINARY_TYPE__UBUNTU_DEBUGINFO, + DSO_BINARY_TYPE__BUILDID_DEBUGINFO, + DSO_BINARY_TYPE__SYSTEM_PATH_DSO, + DSO_BINARY_TYPE__GUEST_KMODULE, + DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE, + DSO_BINARY_TYPE__NOT_FOUND, +}; + +#define DSO_BINARY_TYPE__SYMTAB_CNT ARRAY_SIZE(binary_type_symtab) + +bool symbol_type__is_a(char symbol_type, enum map_type map_type) +{ + symbol_type = toupper(symbol_type); + + switch (map_type) { + case MAP__FUNCTION: + return symbol_type == 'T' || symbol_type == 'W'; + case MAP__VARIABLE: + return symbol_type == 'D'; + default: + return false; + } +} + +static int prefix_underscores_count(const char *str) +{ + const char *tail = str; + + while (*tail == '_') + tail++; + + return tail - str; +} + +#define SYMBOL_A 0 +#define SYMBOL_B 1 + +static int choose_best_symbol(struct symbol *syma, struct symbol *symb) +{ + s64 a; + s64 b; + + /* Prefer a symbol with non zero length */ + a = syma->end - syma->start; + b = symb->end - symb->start; + if ((b == 0) && (a > 0)) + return SYMBOL_A; + else if ((a == 0) && (b > 0)) + return SYMBOL_B; + + /* Prefer a non weak symbol over a weak one */ + a = syma->binding == STB_WEAK; + b = symb->binding == STB_WEAK; + if (b && !a) + return SYMBOL_A; + if (a && !b) + return SYMBOL_B; + + /* Prefer a global symbol over a non global one */ + a = syma->binding == STB_GLOBAL; + b = symb->binding == STB_GLOBAL; + if (a && !b) + return SYMBOL_A; + if (b && !a) + return SYMBOL_B; + + /* Prefer a symbol with less underscores */ + a = prefix_underscores_count(syma->name); + b = prefix_underscores_count(symb->name); + if (b > a) + return SYMBOL_A; + else if (a > b) + return SYMBOL_B; + + /* If all else fails, choose the symbol with the longest name */ + if (strlen(syma->name) >= strlen(symb->name)) + return SYMBOL_A; + else + return SYMBOL_B; +} + +void symbols__fixup_duplicate(struct rb_root *symbols) +{ + struct rb_node *nd; + struct symbol *curr, *next; + + nd = rb_first(symbols); + + while (nd) { + curr = rb_entry(nd, struct symbol, rb_node); +again: + nd = rb_next(&curr->rb_node); + next = rb_entry(nd, struct symbol, rb_node); + + if (!nd) + break; + + if (curr->start != next->start) + continue; + + if (choose_best_symbol(curr, next) == SYMBOL_A) { + rb_erase(&next->rb_node, symbols); + goto again; + } else { + nd = rb_next(&curr->rb_node); + rb_erase(&curr->rb_node, symbols); + } + } +} + +void symbols__fixup_end(struct rb_root *symbols) +{ + struct rb_node *nd, *prevnd = rb_first(symbols); + struct symbol *curr, *prev; + + if (prevnd == NULL) + return; + + curr = rb_entry(prevnd, struct symbol, rb_node); + + for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) { + prev = curr; + curr = rb_entry(nd, struct symbol, rb_node); + + if (prev->end == prev->start && prev->end != curr->start) + prev->end = curr->start - 1; + } + + /* Last entry */ + if (curr->end == curr->start) + curr->end = roundup(curr->start, 4096); +} + +void __map_groups__fixup_end(struct map_groups *mg, enum map_type type) +{ + struct map *prev, *curr; + struct rb_node *nd, *prevnd = rb_first(&mg->maps[type]); + + if (prevnd == NULL) + return; + + curr = rb_entry(prevnd, struct map, rb_node); + + for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) { + prev = curr; + curr = rb_entry(nd, struct map, rb_node); + prev->end = curr->start - 1; + } + + /* + * We still haven't the actual symbols, so guess the + * last map final address. + */ + curr->end = ~0ULL; +} + +struct symbol *symbol__new(u64 start, u64 len, u8 binding, const char *name) +{ + size_t namelen = strlen(name) + 1; + struct symbol *sym = calloc(1, (symbol_conf.priv_size + + sizeof(*sym) + namelen)); + if (sym == NULL) + return NULL; + + if (symbol_conf.priv_size) + sym = ((void *)sym) + symbol_conf.priv_size; + + sym->start = start; + sym->end = len ? start + len - 1 : start; + sym->binding = binding; + sym->namelen = namelen - 1; + + pr_debug4("%s: %s %#" PRIx64 "-%#" PRIx64 "\n", + __func__, name, start, sym->end); + memcpy(sym->name, name, namelen); + + return sym; +} + +void symbol__delete(struct symbol *sym) +{ + free(((void *)sym) - symbol_conf.priv_size); +} + +size_t symbol__fprintf(struct symbol *sym, FILE *fp) +{ + return fprintf(fp, " %" PRIx64 "-%" PRIx64 " %c %s\n", + sym->start, sym->end, + sym->binding == STB_GLOBAL ? 'g' : + sym->binding == STB_LOCAL ? 'l' : 'w', + sym->name); +} + +size_t symbol__fprintf_symname_offs(const struct symbol *sym, + const struct addr_location *al, FILE *fp) +{ + unsigned long offset; + size_t length; + + if (sym && sym->name) { + length = fprintf(fp, "%s", sym->name); + if (al) { + offset = al->addr - sym->start; + length += fprintf(fp, "+0x%lx", offset); + } + return length; + } else + return fprintf(fp, "[unknown]"); +} + +size_t symbol__fprintf_symname(const struct symbol *sym, FILE *fp) +{ + return symbol__fprintf_symname_offs(sym, NULL, fp); +} + +void symbols__delete(struct rb_root *symbols) +{ + struct symbol *pos; + struct rb_node *next = rb_first(symbols); + + while (next) { + pos = rb_entry(next, struct symbol, rb_node); + next = rb_next(&pos->rb_node); + rb_erase(&pos->rb_node, symbols); + symbol__delete(pos); + } +} + +void symbols__insert(struct rb_root *symbols, struct symbol *sym) +{ + struct rb_node **p = &symbols->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, symbols); +} + +static struct symbol *symbols__find(struct rb_root *symbols, u64 ip) +{ + struct rb_node *n; + + if (symbols == NULL) + return NULL; + + n = symbols->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; +} + +struct symbol_name_rb_node { + struct rb_node rb_node; + struct symbol sym; +}; + +static void symbols__insert_by_name(struct rb_root *symbols, struct symbol *sym) +{ + struct rb_node **p = &symbols->rb_node; + struct rb_node *parent = NULL; + struct symbol_name_rb_node *symn, *s; + + symn = container_of(sym, struct symbol_name_rb_node, sym); + + while (*p != NULL) { + parent = *p; + s = rb_entry(parent, struct symbol_name_rb_node, rb_node); + if (strcmp(sym->name, s->sym.name) < 0) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; + } + rb_link_node(&symn->rb_node, parent, p); + rb_insert_color(&symn->rb_node, symbols); +} + +static void symbols__sort_by_name(struct rb_root *symbols, + struct rb_root *source) +{ + struct rb_node *nd; + + for (nd = rb_first(source); nd; nd = rb_next(nd)) { + struct symbol *pos = rb_entry(nd, struct symbol, rb_node); + symbols__insert_by_name(symbols, pos); + } +} + +static struct symbol *symbols__find_by_name(struct rb_root *symbols, + const char *name) +{ + struct rb_node *n; + + if (symbols == NULL) + return NULL; + + n = symbols->rb_node; + + while (n) { + struct symbol_name_rb_node *s; + int cmp; + + s = rb_entry(n, struct symbol_name_rb_node, rb_node); + cmp = strcmp(name, s->sym.name); + + if (cmp < 0) + n = n->rb_left; + else if (cmp > 0) + n = n->rb_right; + else + return &s->sym; + } + + return NULL; +} + +struct symbol *dso__find_symbol(struct dso *dso, + enum map_type type, u64 addr) +{ + return symbols__find(&dso->symbols[type], addr); +} + +struct symbol *dso__find_symbol_by_name(struct dso *dso, enum map_type type, + const char *name) +{ + return symbols__find_by_name(&dso->symbol_names[type], name); +} + +void dso__sort_by_name(struct dso *dso, enum map_type type) +{ + dso__set_sorted_by_name(dso, type); + return symbols__sort_by_name(&dso->symbol_names[type], + &dso->symbols[type]); +} + +size_t dso__fprintf_symbols_by_name(struct dso *dso, + enum map_type type, FILE *fp) +{ + size_t ret = 0; + struct rb_node *nd; + struct symbol_name_rb_node *pos; + + for (nd = rb_first(&dso->symbol_names[type]); nd; nd = rb_next(nd)) { + pos = rb_entry(nd, struct symbol_name_rb_node, rb_node); + fprintf(fp, "%s\n", pos->sym.name); + } + + return ret; +} + +int kallsyms__parse(const char *filename, void *arg, + int (*process_symbol)(void *arg, const char *name, + char type, u64 start)) +{ + char *line = NULL; + size_t n; + int err = -1; + FILE *file = fopen(filename, "r"); + + if (file == NULL) + goto out_failure; + + err = 0; + + while (!feof(file)) { + u64 start; + int line_len, len; + char symbol_type; + char *symbol_name; + + line_len = getline(&line, &n, file); + if (line_len < 0 || !line) + break; + + line[--line_len] = '\0'; /* \n */ + + len = hex2u64(line, &start); + + len++; + if (len + 2 >= line_len) + continue; + + symbol_type = line[len]; + len += 2; + symbol_name = line + len; + len = line_len - len; + + if (len >= KSYM_NAME_LEN) { + err = -1; + break; + } + + err = process_symbol(arg, symbol_name, + symbol_type, start); + if (err) + break; + } + + free(line); + fclose(file); + return err; + +out_failure: + return -1; +} + +struct process_kallsyms_args { + struct map *map; + struct dso *dso; +}; + +static u8 kallsyms2elf_type(char type) +{ + if (type == 'W') + return STB_WEAK; + + return isupper(type) ? STB_GLOBAL : STB_LOCAL; +} + +static int map__process_kallsym_symbol(void *arg, const char *name, + char type, u64 start) +{ + struct symbol *sym; + struct process_kallsyms_args *a = arg; + struct rb_root *root = &a->dso->symbols[a->map->type]; + + if (!symbol_type__is_a(type, a->map->type)) + return 0; + + /* + * module symbols are not sorted so we add all + * symbols, setting length to 0, and rely on + * symbols__fixup_end() to fix it up. + */ + sym = symbol__new(start, 0, kallsyms2elf_type(type), name); + if (sym == NULL) + return -ENOMEM; + /* + * We will pass the symbols to the filter later, in + * map__split_kallsyms, when we have split the maps per module + */ + symbols__insert(root, sym); + + return 0; +} + +/* + * Loads the function entries in /proc/kallsyms into kernel_map->dso, + * so that we can in the next step set the symbol ->end address and then + * call kernel_maps__split_kallsyms. + */ +static int dso__load_all_kallsyms(struct dso *dso, const char *filename, + struct map *map) +{ + struct process_kallsyms_args args = { .map = map, .dso = dso, }; + return kallsyms__parse(filename, &args, map__process_kallsym_symbol); +} + +/* + * Split the symbols into maps, making sure there are no overlaps, i.e. the + * kernel range is broken in several maps, named [kernel].N, as we don't have + * the original ELF section names vmlinux have. + */ +static int dso__split_kallsyms(struct dso *dso, struct map *map, + symbol_filter_t filter) +{ + struct map_groups *kmaps = map__kmap(map)->kmaps; + struct machine *machine = kmaps->machine; + struct map *curr_map = map; + struct symbol *pos; + int count = 0, moved = 0; + struct rb_root *root = &dso->symbols[map->type]; + struct rb_node *next = rb_first(root); + int kernel_range = 0; + + while (next) { + char *module; + + pos = rb_entry(next, struct symbol, rb_node); + next = rb_next(&pos->rb_node); + + module = strchr(pos->name, '\t'); + if (module) { + if (!symbol_conf.use_modules) + goto discard_symbol; + + *module++ = '\0'; + + if (strcmp(curr_map->dso->short_name, module)) { + if (curr_map != map && + dso->kernel == DSO_TYPE_GUEST_KERNEL && + machine__is_default_guest(machine)) { + /* + * We assume all symbols of a module are + * continuous in * kallsyms, so curr_map + * points to a module and all its + * symbols are in its kmap. Mark it as + * loaded. + */ + dso__set_loaded(curr_map->dso, + curr_map->type); + } + + curr_map = map_groups__find_by_name(kmaps, + map->type, module); + if (curr_map == NULL) { + pr_debug("%s/proc/{kallsyms,modules} " + "inconsistency while looking " + "for \"%s\" module!\n", + machine->root_dir, module); + curr_map = map; + goto discard_symbol; + } + + if (curr_map->dso->loaded && + !machine__is_default_guest(machine)) + goto discard_symbol; + } + /* + * So that we look just like we get from .ko files, + * i.e. not prelinked, relative to map->start. + */ + pos->start = curr_map->map_ip(curr_map, pos->start); + pos->end = curr_map->map_ip(curr_map, pos->end); + } else if (curr_map != map) { + char dso_name[PATH_MAX]; + struct dso *ndso; + + if (count == 0) { + curr_map = map; + goto filter_symbol; + } + + if (dso->kernel == DSO_TYPE_GUEST_KERNEL) + snprintf(dso_name, sizeof(dso_name), + "[guest.kernel].%d", + kernel_range++); + else + snprintf(dso_name, sizeof(dso_name), + "[kernel].%d", + kernel_range++); + + ndso = dso__new(dso_name); + if (ndso == NULL) + return -1; + + ndso->kernel = dso->kernel; + + curr_map = map__new2(pos->start, ndso, map->type); + if (curr_map == NULL) { + dso__delete(ndso); + return -1; + } + + curr_map->map_ip = curr_map->unmap_ip = identity__map_ip; + map_groups__insert(kmaps, curr_map); + ++kernel_range; + } +filter_symbol: + if (filter && filter(curr_map, pos)) { +discard_symbol: rb_erase(&pos->rb_node, root); + symbol__delete(pos); + } else { + if (curr_map != map) { + rb_erase(&pos->rb_node, root); + symbols__insert(&curr_map->dso->symbols[curr_map->type], pos); + ++moved; + } else + ++count; + } + } + + if (curr_map != map && + dso->kernel == DSO_TYPE_GUEST_KERNEL && + machine__is_default_guest(kmaps->machine)) { + dso__set_loaded(curr_map->dso, curr_map->type); + } + + return count + moved; +} + +bool symbol__restricted_filename(const char *filename, + const char *restricted_filename) +{ + bool restricted = false; + + if (symbol_conf.kptr_restrict) { + char *r = realpath(filename, NULL); + + if (r != NULL) { + restricted = strcmp(r, restricted_filename) == 0; + free(r); + return restricted; + } + } + + return restricted; +} + +int dso__load_kallsyms(struct dso *dso, const char *filename, + struct map *map, symbol_filter_t filter) +{ + if (symbol__restricted_filename(filename, "/proc/kallsyms")) + return -1; + + if (dso__load_all_kallsyms(dso, filename, map) < 0) + return -1; + + symbols__fixup_duplicate(&dso->symbols[map->type]); + symbols__fixup_end(&dso->symbols[map->type]); + + if (dso->kernel == DSO_TYPE_GUEST_KERNEL) + dso->symtab_type = DSO_BINARY_TYPE__GUEST_KALLSYMS; + else + dso->symtab_type = DSO_BINARY_TYPE__KALLSYMS; + + return dso__split_kallsyms(dso, map, filter); +} + +static int dso__load_perf_map(struct dso *dso, struct map *map, + symbol_filter_t filter) +{ + char *line = NULL; + size_t n; + FILE *file; + int nr_syms = 0; + + file = fopen(dso->long_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, STB_GLOBAL, line + len); + + if (sym == NULL) + goto out_delete_line; + + if (filter && filter(map, sym)) + symbol__delete(sym); + else { + symbols__insert(&dso->symbols[map->type], sym); + nr_syms++; + } + } + + free(line); + fclose(file); + + return nr_syms; + +out_delete_line: + free(line); +out_failure: + return -1; +} + +int dso__load(struct dso *dso, struct map *map, symbol_filter_t filter) +{ + char *name; + int ret = -1; + u_int i; + struct machine *machine; + char *root_dir = (char *) ""; + int ss_pos = 0; + struct symsrc ss_[2]; + struct symsrc *syms_ss = NULL, *runtime_ss = NULL; + + dso__set_loaded(dso, map->type); + + if (dso->kernel == DSO_TYPE_KERNEL) + return dso__load_kernel_sym(dso, map, filter); + else if (dso->kernel == DSO_TYPE_GUEST_KERNEL) + return dso__load_guest_kernel_sym(dso, map, filter); + + if (map->groups && map->groups->machine) + machine = map->groups->machine; + else + machine = NULL; + + dso->adjust_symbols = 0; + + if (strncmp(dso->name, "/tmp/perf-", 10) == 0) { + struct stat st; + + if (lstat(dso->name, &st) < 0) + return -1; + + if (st.st_uid && (st.st_uid != geteuid())) { + pr_warning("File %s not owned by current user or root, " + "ignoring it.\n", dso->name); + return -1; + } + + ret = dso__load_perf_map(dso, map, filter); + dso->symtab_type = ret > 0 ? DSO_BINARY_TYPE__JAVA_JIT : + DSO_BINARY_TYPE__NOT_FOUND; + return ret; + } + + if (machine) + root_dir = machine->root_dir; + + name = malloc(PATH_MAX); + if (!name) + return -1; + + /* Iterate over candidate debug images. + * Keep track of "interesting" ones (those which have a symtab, dynsym, + * and/or opd section) for processing. + */ + for (i = 0; i < DSO_BINARY_TYPE__SYMTAB_CNT; i++) { + struct symsrc *ss = &ss_[ss_pos]; + bool next_slot = false; + + enum dso_binary_type symtab_type = binary_type_symtab[i]; + + if (dso__binary_type_file(dso, symtab_type, + root_dir, name, PATH_MAX)) + continue; + + /* Name is now the name of the next image to try */ + if (symsrc__init(ss, dso, name, symtab_type) < 0) + continue; + + if (!syms_ss && symsrc__has_symtab(ss)) { + syms_ss = ss; + next_slot = true; + } + + if (!runtime_ss && symsrc__possibly_runtime(ss)) { + runtime_ss = ss; + next_slot = true; + } + + if (next_slot) { + ss_pos++; + + if (syms_ss && runtime_ss) + break; + } + + } + + if (!runtime_ss && !syms_ss) + goto out_free; + + if (runtime_ss && !syms_ss) { + syms_ss = runtime_ss; + } + + /* We'll have to hope for the best */ + if (!runtime_ss && syms_ss) + runtime_ss = syms_ss; + + if (syms_ss) + ret = dso__load_sym(dso, map, syms_ss, runtime_ss, filter, 0); + else + ret = -1; + + if (ret > 0) { + int nr_plt; + + nr_plt = dso__synthesize_plt_symbols(dso, runtime_ss, map, filter); + if (nr_plt > 0) + ret += nr_plt; + } + + for (; ss_pos > 0; ss_pos--) + symsrc__destroy(&ss_[ss_pos - 1]); +out_free: + free(name); + if (ret < 0 && strstr(dso->name, " (deleted)") != NULL) + return 0; + return ret; +} + +struct map *map_groups__find_by_name(struct map_groups *mg, + enum map_type type, const char *name) +{ + struct rb_node *nd; + + for (nd = rb_first(&mg->maps[type]); nd; nd = rb_next(nd)) { + struct map *map = rb_entry(nd, struct map, rb_node); + + if (map->dso && strcmp(map->dso->short_name, name) == 0) + return map; + } + + return NULL; +} + +int dso__load_vmlinux(struct dso *dso, struct map *map, + const char *vmlinux, symbol_filter_t filter) +{ + int err = -1; + struct symsrc ss; + char symfs_vmlinux[PATH_MAX]; + enum dso_binary_type symtab_type; + + snprintf(symfs_vmlinux, sizeof(symfs_vmlinux), "%s%s", + symbol_conf.symfs, vmlinux); + + if (dso->kernel == DSO_TYPE_GUEST_KERNEL) + symtab_type = DSO_BINARY_TYPE__GUEST_VMLINUX; + else + symtab_type = DSO_BINARY_TYPE__VMLINUX; + + if (symsrc__init(&ss, dso, symfs_vmlinux, symtab_type)) + return -1; + + err = dso__load_sym(dso, map, &ss, &ss, filter, 0); + symsrc__destroy(&ss); + + if (err > 0) { + dso__set_long_name(dso, (char *)vmlinux); + dso__set_loaded(dso, map->type); + pr_debug("Using %s for symbols\n", symfs_vmlinux); + } + + return err; +} + +int dso__load_vmlinux_path(struct dso *dso, struct map *map, + symbol_filter_t filter) +{ + int i, err = 0; + char *filename; + + pr_debug("Looking at the vmlinux_path (%d entries long)\n", + vmlinux_path__nr_entries + 1); + + filename = dso__build_id_filename(dso, NULL, 0); + if (filename != NULL) { + err = dso__load_vmlinux(dso, map, filename, filter); + if (err > 0) { + dso->lname_alloc = 1; + goto out; + } + free(filename); + } + + for (i = 0; i < vmlinux_path__nr_entries; ++i) { + err = dso__load_vmlinux(dso, map, vmlinux_path[i], filter); + if (err > 0) { + dso__set_long_name(dso, strdup(vmlinux_path[i])); + dso->lname_alloc = 1; + break; + } + } +out: + return err; +} + +static int dso__load_kernel_sym(struct dso *dso, struct map *map, + symbol_filter_t filter) +{ + int err; + const char *kallsyms_filename = NULL; + char *kallsyms_allocated_filename = NULL; + /* + * Step 1: if the user specified a kallsyms or vmlinux filename, use + * it and only it, reporting errors to the user if it cannot be used. + * + * For instance, try to analyse an ARM perf.data file _without_ a + * build-id, or if the user specifies the wrong path to the right + * vmlinux file, obviously we can't fallback to another vmlinux (a + * x86_86 one, on the machine where analysis is being performed, say), + * or worse, /proc/kallsyms. + * + * If the specified file _has_ a build-id and there is a build-id + * section in the perf.data file, we will still do the expected + * validation in dso__load_vmlinux and will bail out if they don't + * match. + */ + if (symbol_conf.kallsyms_name != NULL) { + kallsyms_filename = symbol_conf.kallsyms_name; + goto do_kallsyms; + } + + if (symbol_conf.vmlinux_name != NULL) { + err = dso__load_vmlinux(dso, map, + symbol_conf.vmlinux_name, filter); + if (err > 0) { + dso__set_long_name(dso, + strdup(symbol_conf.vmlinux_name)); + dso->lname_alloc = 1; + goto out_fixup; + } + return err; + } + + if (vmlinux_path != NULL) { + err = dso__load_vmlinux_path(dso, map, filter); + if (err > 0) + goto out_fixup; + } + + /* do not try local files if a symfs was given */ + if (symbol_conf.symfs[0] != 0) + return -1; + + /* + * Say the kernel DSO was created when processing the build-id header table, + * we have a build-id, so check if it is the same as the running kernel, + * using it if it is. + */ + if (dso->has_build_id) { + u8 kallsyms_build_id[BUILD_ID_SIZE]; + char sbuild_id[BUILD_ID_SIZE * 2 + 1]; + + if (sysfs__read_build_id("/sys/kernel/notes", kallsyms_build_id, + sizeof(kallsyms_build_id)) == 0) { + if (dso__build_id_equal(dso, kallsyms_build_id)) { + kallsyms_filename = "/proc/kallsyms"; + goto do_kallsyms; + } + } + /* + * Now look if we have it on the build-id cache in + * $HOME/.debug/[kernel.kallsyms]. + */ + build_id__sprintf(dso->build_id, sizeof(dso->build_id), + sbuild_id); + + if (asprintf(&kallsyms_allocated_filename, + "%s/.debug/[kernel.kallsyms]/%s", + getenv("HOME"), sbuild_id) == -1) { + pr_err("Not enough memory for kallsyms file lookup\n"); + return -1; + } + + kallsyms_filename = kallsyms_allocated_filename; + + if (access(kallsyms_filename, F_OK)) { + pr_err("No kallsyms or vmlinux with build-id %s " + "was found\n", sbuild_id); + free(kallsyms_allocated_filename); + return -1; + } + } else { + /* + * Last resort, if we don't have a build-id and couldn't find + * any vmlinux file, try the running kernel kallsyms table. + */ + kallsyms_filename = "/proc/kallsyms"; + } + +do_kallsyms: + err = dso__load_kallsyms(dso, kallsyms_filename, map, filter); + if (err > 0) + pr_debug("Using %s for symbols\n", kallsyms_filename); + free(kallsyms_allocated_filename); + + if (err > 0) { + dso__set_long_name(dso, strdup("[kernel.kallsyms]")); +out_fixup: + map__fixup_start(map); + map__fixup_end(map); + } + + return err; +} + +static int dso__load_guest_kernel_sym(struct dso *dso, struct map *map, + symbol_filter_t filter) +{ + int err; + const char *kallsyms_filename = NULL; + struct machine *machine; + char path[PATH_MAX]; + + if (!map->groups) { + pr_debug("Guest kernel map hasn't the point to groups\n"); + return -1; + } + machine = map->groups->machine; + + if (machine__is_default_guest(machine)) { + /* + * if the user specified a vmlinux filename, use it and only + * it, reporting errors to the user if it cannot be used. + * Or use file guest_kallsyms inputted by user on commandline + */ + if (symbol_conf.default_guest_vmlinux_name != NULL) { + err = dso__load_vmlinux(dso, map, + symbol_conf.default_guest_vmlinux_name, filter); + goto out_try_fixup; + } + + kallsyms_filename = symbol_conf.default_guest_kallsyms; + if (!kallsyms_filename) + return -1; + } else { + sprintf(path, "%s/proc/kallsyms", machine->root_dir); + kallsyms_filename = path; + } + + err = dso__load_kallsyms(dso, kallsyms_filename, map, filter); + if (err > 0) + pr_debug("Using %s for symbols\n", kallsyms_filename); + +out_try_fixup: + if (err > 0) { + if (kallsyms_filename != NULL) { + machine__mmap_name(machine, path, sizeof(path)); + dso__set_long_name(dso, strdup(path)); + } + map__fixup_start(map); + map__fixup_end(map); + } + + return err; +} + +static void vmlinux_path__exit(void) +{ + while (--vmlinux_path__nr_entries >= 0) { + free(vmlinux_path[vmlinux_path__nr_entries]); + vmlinux_path[vmlinux_path__nr_entries] = NULL; + } + + free(vmlinux_path); + vmlinux_path = NULL; +} + +static int vmlinux_path__init(void) +{ + struct utsname uts; + char bf[PATH_MAX]; + + vmlinux_path = malloc(sizeof(char *) * 5); + if (vmlinux_path == NULL) + return -1; + + vmlinux_path[vmlinux_path__nr_entries] = strdup("vmlinux"); + if (vmlinux_path[vmlinux_path__nr_entries] == NULL) + goto out_fail; + ++vmlinux_path__nr_entries; + vmlinux_path[vmlinux_path__nr_entries] = strdup("/boot/vmlinux"); + if (vmlinux_path[vmlinux_path__nr_entries] == NULL) + goto out_fail; + ++vmlinux_path__nr_entries; + + /* only try running kernel version if no symfs was given */ + if (symbol_conf.symfs[0] != 0) + return 0; + + if (uname(&uts) < 0) + return -1; + + snprintf(bf, sizeof(bf), "/boot/vmlinux-%s", uts.release); + vmlinux_path[vmlinux_path__nr_entries] = strdup(bf); + if (vmlinux_path[vmlinux_path__nr_entries] == NULL) + goto out_fail; + ++vmlinux_path__nr_entries; + snprintf(bf, sizeof(bf), "/lib/modules/%s/build/vmlinux", uts.release); + vmlinux_path[vmlinux_path__nr_entries] = strdup(bf); + if (vmlinux_path[vmlinux_path__nr_entries] == NULL) + goto out_fail; + ++vmlinux_path__nr_entries; + snprintf(bf, sizeof(bf), "/usr/lib/debug/lib/modules/%s/vmlinux", + uts.release); + vmlinux_path[vmlinux_path__nr_entries] = strdup(bf); + if (vmlinux_path[vmlinux_path__nr_entries] == NULL) + goto out_fail; + ++vmlinux_path__nr_entries; + + return 0; + +out_fail: + vmlinux_path__exit(); + return -1; +} + +static int setup_list(struct strlist **list, const char *list_str, + const char *list_name) +{ + if (list_str == NULL) + return 0; + + *list = strlist__new(true, list_str); + if (!*list) { + pr_err("problems parsing %s list\n", list_name); + return -1; + } + return 0; +} + +static bool symbol__read_kptr_restrict(void) +{ + bool value = false; + + if (geteuid() != 0) { + FILE *fp = fopen("/proc/sys/kernel/kptr_restrict", "r"); + if (fp != NULL) { + char line[8]; + + if (fgets(line, sizeof(line), fp) != NULL) + value = atoi(line) != 0; + + fclose(fp); + } + } + + return value; +} + +int symbol__init(void) +{ + const char *symfs; + + if (symbol_conf.initialized) + return 0; + + symbol_conf.priv_size = PERF_ALIGN(symbol_conf.priv_size, sizeof(u64)); + + symbol__elf_init(); + + if (symbol_conf.sort_by_name) + symbol_conf.priv_size += (sizeof(struct symbol_name_rb_node) - + sizeof(struct symbol)); + + if (symbol_conf.try_vmlinux_path && vmlinux_path__init() < 0) + return -1; + + if (symbol_conf.field_sep && *symbol_conf.field_sep == '.') { + pr_err("'.' is the only non valid --field-separator argument\n"); + return -1; + } + + if (setup_list(&symbol_conf.dso_list, + symbol_conf.dso_list_str, "dso") < 0) + return -1; + + if (setup_list(&symbol_conf.comm_list, + symbol_conf.comm_list_str, "comm") < 0) + goto out_free_dso_list; + + if (setup_list(&symbol_conf.sym_list, + symbol_conf.sym_list_str, "symbol") < 0) + goto out_free_comm_list; + + /* + * A path to symbols of "/" is identical to "" + * reset here for simplicity. + */ + symfs = realpath(symbol_conf.symfs, NULL); + if (symfs == NULL) + symfs = symbol_conf.symfs; + if (strcmp(symfs, "/") == 0) + symbol_conf.symfs = ""; + if (symfs != symbol_conf.symfs) + free((void *)symfs); + + symbol_conf.kptr_restrict = symbol__read_kptr_restrict(); + + symbol_conf.initialized = true; + return 0; + +out_free_comm_list: + strlist__delete(symbol_conf.comm_list); +out_free_dso_list: + strlist__delete(symbol_conf.dso_list); + return -1; +} + +void symbol__exit(void) +{ + if (!symbol_conf.initialized) + return; + strlist__delete(symbol_conf.sym_list); + strlist__delete(symbol_conf.dso_list); + strlist__delete(symbol_conf.comm_list); + vmlinux_path__exit(); + symbol_conf.sym_list = symbol_conf.dso_list = symbol_conf.comm_list = NULL; + symbol_conf.initialized = false; +} diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h new file mode 100644 index 00000000..b62ca37c --- /dev/null +++ b/tools/perf/util/symbol.h @@ -0,0 +1,243 @@ +#ifndef __PERF_SYMBOL +#define __PERF_SYMBOL 1 + +#include <linux/types.h> +#include <stdbool.h> +#include <stdint.h> +#include "map.h" +#include "../perf.h" +#include <linux/list.h> +#include <linux/rbtree.h> +#include <stdio.h> +#include <byteswap.h> +#include <libgen.h> +#include "build-id.h" + +#ifdef LIBELF_SUPPORT +#include <libelf.h> +#include <gelf.h> +#endif +#include <elf.h> + +#include "dso.h" + +#ifdef HAVE_CPLUS_DEMANGLE +extern char *cplus_demangle(const char *, int); + +static inline char *bfd_demangle(void __maybe_unused *v, const char *c, int i) +{ + return cplus_demangle(c, i); +} +#else +#ifdef NO_DEMANGLE +static inline char *bfd_demangle(void __maybe_unused *v, + const char __maybe_unused *c, + int __maybe_unused i) +{ + return NULL; +} +#else +#define PACKAGE 'perf' +#include <bfd.h> +#endif +#endif + +/* + * libelf 0.8.x and earlier do not support ELF_C_READ_MMAP; + * for newer versions we can use mmap to reduce memory usage: + */ +#ifdef LIBELF_MMAP +# define PERF_ELF_C_READ_MMAP ELF_C_READ_MMAP +#else +# define PERF_ELF_C_READ_MMAP ELF_C_READ +#endif + +#ifndef DMGL_PARAMS +#define DMGL_PARAMS (1 << 0) /* Include function args */ +#define DMGL_ANSI (1 << 1) /* Include const, volatile, etc */ +#endif + +/** struct symbol - symtab entry + * + * @ignore - resolvable but tools ignore it (e.g. idle routines) + */ +struct symbol { + struct rb_node rb_node; + u64 start; + u64 end; + u16 namelen; + u8 binding; + bool ignore; + char name[0]; +}; + +void symbol__delete(struct symbol *sym); +void symbols__delete(struct rb_root *symbols); + +static inline size_t symbol__size(const struct symbol *sym) +{ + return sym->end - sym->start + 1; +} + +struct strlist; + +struct symbol_conf { + unsigned short priv_size; + unsigned short nr_events; + bool try_vmlinux_path, + show_kernel_path, + use_modules, + sort_by_name, + show_nr_samples, + show_total_period, + use_callchain, + exclude_other, + show_cpu_utilization, + initialized, + kptr_restrict, + annotate_asm_raw, + annotate_src, + event_group; + const char *vmlinux_name, + *kallsyms_name, + *source_prefix, + *field_sep; + const char *default_guest_vmlinux_name, + *default_guest_kallsyms, + *default_guest_modules; + const char *guestmount; + const char *dso_list_str, + *comm_list_str, + *sym_list_str, + *col_width_list_str; + struct strlist *dso_list, + *comm_list, + *sym_list, + *dso_from_list, + *dso_to_list, + *sym_from_list, + *sym_to_list; + const char *symfs; +}; + +extern struct symbol_conf symbol_conf; +extern int vmlinux_path__nr_entries; +extern char **vmlinux_path; + +static inline void *symbol__priv(struct symbol *sym) +{ + return ((void *)sym) - symbol_conf.priv_size; +} + +struct ref_reloc_sym { + const char *name; + u64 addr; + u64 unrelocated_addr; +}; + +struct map_symbol { + struct map *map; + struct symbol *sym; + bool unfolded; + bool has_children; +}; + +struct addr_map_symbol { + struct map *map; + struct symbol *sym; + u64 addr; + u64 al_addr; +}; + +struct branch_info { + struct addr_map_symbol from; + struct addr_map_symbol to; + struct branch_flags flags; +}; + +struct addr_location { + struct thread *thread; + struct map *map; + struct symbol *sym; + u64 addr; + char level; + bool filtered; + u8 cpumode; + s32 cpu; +}; + +struct symsrc { + char *name; + int fd; + enum dso_binary_type type; + +#ifdef LIBELF_SUPPORT + Elf *elf; + GElf_Ehdr ehdr; + + Elf_Scn *opdsec; + size_t opdidx; + GElf_Shdr opdshdr; + + Elf_Scn *symtab; + GElf_Shdr symshdr; + + Elf_Scn *dynsym; + size_t dynsym_idx; + GElf_Shdr dynshdr; + + bool adjust_symbols; +#endif +}; + +void symsrc__destroy(struct symsrc *ss); +int symsrc__init(struct symsrc *ss, struct dso *dso, const char *name, + enum dso_binary_type type); +bool symsrc__has_symtab(struct symsrc *ss); +bool symsrc__possibly_runtime(struct symsrc *ss); + +int dso__load(struct dso *dso, struct map *map, symbol_filter_t filter); +int dso__load_vmlinux(struct dso *dso, struct map *map, + const char *vmlinux, symbol_filter_t filter); +int dso__load_vmlinux_path(struct dso *dso, struct map *map, + symbol_filter_t filter); +int dso__load_kallsyms(struct dso *dso, const char *filename, struct map *map, + symbol_filter_t filter); + +struct symbol *dso__find_symbol(struct dso *dso, enum map_type type, + u64 addr); +struct symbol *dso__find_symbol_by_name(struct dso *dso, enum map_type type, + const char *name); + +int filename__read_build_id(const char *filename, void *bf, size_t size); +int sysfs__read_build_id(const char *filename, void *bf, size_t size); +int kallsyms__parse(const char *filename, void *arg, + int (*process_symbol)(void *arg, const char *name, + char type, u64 start)); +int filename__read_debuglink(const char *filename, char *debuglink, + size_t size); + +int symbol__init(void); +void symbol__exit(void); +void symbol__elf_init(void); +struct symbol *symbol__new(u64 start, u64 len, u8 binding, const char *name); +size_t symbol__fprintf_symname_offs(const struct symbol *sym, + const struct addr_location *al, FILE *fp); +size_t symbol__fprintf_symname(const struct symbol *sym, FILE *fp); +size_t symbol__fprintf(struct symbol *sym, FILE *fp); +bool symbol_type__is_a(char symbol_type, enum map_type map_type); +bool symbol__restricted_filename(const char *filename, + const char *restricted_filename); + +int dso__load_sym(struct dso *dso, struct map *map, struct symsrc *syms_ss, + struct symsrc *runtime_ss, symbol_filter_t filter, + int kmodule); +int dso__synthesize_plt_symbols(struct dso *dso, struct symsrc *ss, + struct map *map, symbol_filter_t filter); + +void symbols__insert(struct rb_root *symbols, struct symbol *sym); +void symbols__fixup_duplicate(struct rb_root *symbols); +void symbols__fixup_end(struct rb_root *symbols); +void __map_groups__fixup_end(struct map_groups *mg, enum map_type type); + +#endif /* __PERF_SYMBOL */ diff --git a/tools/perf/util/sysfs.c b/tools/perf/util/sysfs.c new file mode 100644 index 00000000..f71e9eaf --- /dev/null +++ b/tools/perf/util/sysfs.c @@ -0,0 +1,60 @@ + +#include "util.h" +#include "sysfs.h" + +static const char * const sysfs_known_mountpoints[] = { + "/sys", + 0, +}; + +static int sysfs_found; +char sysfs_mountpoint[PATH_MAX + 1]; + +static int sysfs_valid_mountpoint(const char *sysfs) +{ + struct statfs st_fs; + + if (statfs(sysfs, &st_fs) < 0) + return -ENOENT; + else if (st_fs.f_type != (long) SYSFS_MAGIC) + return -ENOENT; + + return 0; +} + +const char *sysfs_find_mountpoint(void) +{ + const char * const *ptr; + char type[100]; + FILE *fp; + + if (sysfs_found) + return (const char *) sysfs_mountpoint; + + ptr = sysfs_known_mountpoints; + while (*ptr) { + if (sysfs_valid_mountpoint(*ptr) == 0) { + sysfs_found = 1; + strcpy(sysfs_mountpoint, *ptr); + return sysfs_mountpoint; + } + ptr++; + } + + /* give up and parse /proc/mounts */ + fp = fopen("/proc/mounts", "r"); + if (fp == NULL) + return NULL; + + while (!sysfs_found && + fscanf(fp, "%*s %" STR(PATH_MAX) "s %99s %*s %*d %*d\n", + sysfs_mountpoint, type) == 2) { + + if (strcmp(type, "sysfs") == 0) + sysfs_found = 1; + } + + fclose(fp); + + return sysfs_found ? sysfs_mountpoint : NULL; +} diff --git a/tools/perf/util/sysfs.h b/tools/perf/util/sysfs.h new file mode 100644 index 00000000..a813b720 --- /dev/null +++ b/tools/perf/util/sysfs.h @@ -0,0 +1,6 @@ +#ifndef __SYSFS_H__ +#define __SYSFS_H__ + +const char *sysfs_find_mountpoint(void); + +#endif /* __DEBUGFS_H__ */ diff --git a/tools/perf/util/target.c b/tools/perf/util/target.c new file mode 100644 index 00000000..065528b7 --- /dev/null +++ b/tools/perf/util/target.c @@ -0,0 +1,151 @@ +/* + * Helper functions for handling target threads/cpus + * + * Copyright (C) 2012, LG Electronics, Namhyung Kim <namhyung.kim@lge.com> + * + * Released under the GPL v2. + */ + +#include "target.h" +#include "debug.h" + +#include <pwd.h> +#include <string.h> + + +enum perf_target_errno perf_target__validate(struct perf_target *target) +{ + enum perf_target_errno ret = PERF_ERRNO_TARGET__SUCCESS; + + if (target->pid) + target->tid = target->pid; + + /* CPU and PID are mutually exclusive */ + if (target->tid && target->cpu_list) { + target->cpu_list = NULL; + if (ret == PERF_ERRNO_TARGET__SUCCESS) + ret = PERF_ERRNO_TARGET__PID_OVERRIDE_CPU; + } + + /* UID and PID are mutually exclusive */ + if (target->tid && target->uid_str) { + target->uid_str = NULL; + if (ret == PERF_ERRNO_TARGET__SUCCESS) + ret = PERF_ERRNO_TARGET__PID_OVERRIDE_UID; + } + + /* UID and CPU are mutually exclusive */ + if (target->uid_str && target->cpu_list) { + target->cpu_list = NULL; + if (ret == PERF_ERRNO_TARGET__SUCCESS) + ret = PERF_ERRNO_TARGET__UID_OVERRIDE_CPU; + } + + /* PID and SYSTEM are mutually exclusive */ + if (target->tid && target->system_wide) { + target->system_wide = false; + if (ret == PERF_ERRNO_TARGET__SUCCESS) + ret = PERF_ERRNO_TARGET__PID_OVERRIDE_SYSTEM; + } + + /* UID and SYSTEM are mutually exclusive */ + if (target->uid_str && target->system_wide) { + target->system_wide = false; + if (ret == PERF_ERRNO_TARGET__SUCCESS) + ret = PERF_ERRNO_TARGET__UID_OVERRIDE_SYSTEM; + } + + return ret; +} + +enum perf_target_errno perf_target__parse_uid(struct perf_target *target) +{ + struct passwd pwd, *result; + char buf[1024]; + const char *str = target->uid_str; + + target->uid = UINT_MAX; + if (str == NULL) + return PERF_ERRNO_TARGET__SUCCESS; + + /* Try user name first */ + getpwnam_r(str, &pwd, buf, sizeof(buf), &result); + + if (result == NULL) { + /* + * The user name not found. Maybe it's a UID number. + */ + char *endptr; + int uid = strtol(str, &endptr, 10); + + if (*endptr != '\0') + return PERF_ERRNO_TARGET__INVALID_UID; + + getpwuid_r(uid, &pwd, buf, sizeof(buf), &result); + + if (result == NULL) + return PERF_ERRNO_TARGET__USER_NOT_FOUND; + } + + target->uid = result->pw_uid; + return PERF_ERRNO_TARGET__SUCCESS; +} + +/* + * This must have a same ordering as the enum perf_target_errno. + */ +static const char *perf_target__error_str[] = { + "PID/TID switch overriding CPU", + "PID/TID switch overriding UID", + "UID switch overriding CPU", + "PID/TID switch overriding SYSTEM", + "UID switch overriding SYSTEM", + "Invalid User: %s", + "Problems obtaining information for user %s", +}; + +int perf_target__strerror(struct perf_target *target, int errnum, + char *buf, size_t buflen) +{ + int idx; + const char *msg; + + BUG_ON(buflen == 0); + + if (errnum >= 0) { + const char *err = strerror_r(errnum, buf, buflen); + + if (err != buf) { + size_t len = strlen(err); + memcpy(buf, err, min(buflen - 1, len)); + *(buf + min(buflen - 1, len)) = '\0'; + } + + return 0; + } + + if (errnum < __PERF_ERRNO_TARGET__START || + errnum >= __PERF_ERRNO_TARGET__END) + return -1; + + idx = errnum - __PERF_ERRNO_TARGET__START; + msg = perf_target__error_str[idx]; + + switch (errnum) { + case PERF_ERRNO_TARGET__PID_OVERRIDE_CPU + ... PERF_ERRNO_TARGET__UID_OVERRIDE_SYSTEM: + snprintf(buf, buflen, "%s", msg); + break; + + case PERF_ERRNO_TARGET__INVALID_UID: + case PERF_ERRNO_TARGET__USER_NOT_FOUND: + snprintf(buf, buflen, msg, target->uid_str); + break; + + default: + /* cannot reach here */ + break; + } + + return 0; +} diff --git a/tools/perf/util/target.h b/tools/perf/util/target.h new file mode 100644 index 00000000..a4be8575 --- /dev/null +++ b/tools/perf/util/target.h @@ -0,0 +1,65 @@ +#ifndef _PERF_TARGET_H +#define _PERF_TARGET_H + +#include <stdbool.h> +#include <sys/types.h> + +struct perf_target { + const char *pid; + const char *tid; + const char *cpu_list; + const char *uid_str; + uid_t uid; + bool system_wide; + bool uses_mmap; +}; + +enum perf_target_errno { + PERF_ERRNO_TARGET__SUCCESS = 0, + + /* + * Choose an arbitrary negative big number not to clash with standard + * errno since SUS requires the errno has distinct positive values. + * See 'Issue 6' in the link below. + * + * http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/errno.h.html + */ + __PERF_ERRNO_TARGET__START = -10000, + + + /* for perf_target__validate() */ + PERF_ERRNO_TARGET__PID_OVERRIDE_CPU = __PERF_ERRNO_TARGET__START, + PERF_ERRNO_TARGET__PID_OVERRIDE_UID, + PERF_ERRNO_TARGET__UID_OVERRIDE_CPU, + PERF_ERRNO_TARGET__PID_OVERRIDE_SYSTEM, + PERF_ERRNO_TARGET__UID_OVERRIDE_SYSTEM, + + /* for perf_target__parse_uid() */ + PERF_ERRNO_TARGET__INVALID_UID, + PERF_ERRNO_TARGET__USER_NOT_FOUND, + + __PERF_ERRNO_TARGET__END, +}; + +enum perf_target_errno perf_target__validate(struct perf_target *target); +enum perf_target_errno perf_target__parse_uid(struct perf_target *target); + +int perf_target__strerror(struct perf_target *target, int errnum, char *buf, + size_t buflen); + +static inline bool perf_target__has_task(struct perf_target *target) +{ + return target->tid || target->pid || target->uid_str; +} + +static inline bool perf_target__has_cpu(struct perf_target *target) +{ + return target->system_wide || target->cpu_list; +} + +static inline bool perf_target__none(struct perf_target *target) +{ + return !perf_target__has_task(target) && !perf_target__has_cpu(target); +} + +#endif /* _PERF_TARGET_H */ diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c new file mode 100644 index 00000000..632e40e5 --- /dev/null +++ b/tools/perf/util/thread.c @@ -0,0 +1,86 @@ +#include "../perf.h" +#include <stdlib.h> +#include <stdio.h> +#include <string.h> +#include "session.h" +#include "thread.h" +#include "util.h" +#include "debug.h" + +struct thread *thread__new(pid_t pid) +{ + struct thread *self = zalloc(sizeof(*self)); + + if (self != NULL) { + map_groups__init(&self->mg); + self->pid = pid; + self->comm = malloc(32); + if (self->comm) + snprintf(self->comm, 32, ":%d", self->pid); + } + + return self; +} + +void thread__delete(struct thread *self) +{ + map_groups__exit(&self->mg); + free(self->comm); + free(self); +} + +int thread__set_comm(struct thread *self, const char *comm) +{ + int err; + + if (self->comm) + free(self->comm); + self->comm = strdup(comm); + err = self->comm == NULL ? -ENOMEM : 0; + if (!err) { + self->comm_set = true; + } + return err; +} + +int thread__comm_len(struct thread *self) +{ + if (!self->comm_len) { + if (!self->comm) + return 0; + self->comm_len = strlen(self->comm); + } + + return self->comm_len; +} + +size_t thread__fprintf(struct thread *thread, FILE *fp) +{ + return fprintf(fp, "Thread %d %s\n", thread->pid, thread->comm) + + map_groups__fprintf(&thread->mg, verbose, fp); +} + +void thread__insert_map(struct thread *self, struct map *map) +{ + map_groups__fixup_overlappings(&self->mg, map, verbose, stderr); + map_groups__insert(&self->mg, map); +} + +int thread__fork(struct thread *self, struct thread *parent) +{ + int i; + + if (parent->comm_set) { + if (self->comm) + free(self->comm); + self->comm = strdup(parent->comm); + if (!self->comm) + return -ENOMEM; + self->comm_set = true; + } + + for (i = 0; i < MAP__NR_TYPES; ++i) + if (map_groups__clone(&self->mg, &parent->mg, i) < 0) + return -ENOMEM; + return 0; +} diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h new file mode 100644 index 00000000..5ad26640 --- /dev/null +++ b/tools/perf/util/thread.h @@ -0,0 +1,49 @@ +#ifndef __PERF_THREAD_H +#define __PERF_THREAD_H + +#include <linux/rbtree.h> +#include <unistd.h> +#include <sys/types.h> +#include "symbol.h" + +struct thread { + union { + struct rb_node rb_node; + struct list_head node; + }; + struct map_groups mg; + pid_t pid; + char shortname[3]; + bool comm_set; + char *comm; + int comm_len; + + void *priv; +}; + +struct machine; + +struct thread *thread__new(pid_t pid); +void thread__delete(struct thread *self); + +int thread__set_comm(struct thread *self, const char *comm); +int thread__comm_len(struct thread *self); +void thread__insert_map(struct thread *self, struct map *map); +int thread__fork(struct thread *self, struct thread *parent); +size_t thread__fprintf(struct thread *thread, FILE *fp); + +static inline struct map *thread__find_map(struct thread *self, + enum map_type type, u64 addr) +{ + return self ? map_groups__find(&self->mg, type, addr) : NULL; +} + +void thread__find_addr_map(struct thread *thread, struct machine *machine, + u8 cpumode, enum map_type type, u64 addr, + struct addr_location *al); + +void thread__find_addr_location(struct thread *thread, struct machine *machine, + u8 cpumode, enum map_type type, u64 addr, + struct addr_location *al, + symbol_filter_t filter); +#endif /* __PERF_THREAD_H */ diff --git a/tools/perf/util/thread_map.c b/tools/perf/util/thread_map.c new file mode 100644 index 00000000..9b5f856c --- /dev/null +++ b/tools/perf/util/thread_map.c @@ -0,0 +1,296 @@ +#include <dirent.h> +#include <limits.h> +#include <stdbool.h> +#include <stdlib.h> +#include <stdio.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <unistd.h> +#include "strlist.h" +#include <string.h> +#include "thread_map.h" + +/* Skip "." and ".." directories */ +static int filter(const struct dirent *dir) +{ + if (dir->d_name[0] == '.') + return 0; + else + return 1; +} + +struct thread_map *thread_map__new_by_pid(pid_t pid) +{ + struct thread_map *threads; + char name[256]; + int items; + struct dirent **namelist = NULL; + int i; + + sprintf(name, "/proc/%d/task", pid); + items = scandir(name, &namelist, filter, NULL); + if (items <= 0) + return NULL; + + threads = malloc(sizeof(*threads) + sizeof(pid_t) * items); + if (threads != NULL) { + for (i = 0; i < items; i++) + threads->map[i] = atoi(namelist[i]->d_name); + threads->nr = items; + } + + for (i=0; i<items; i++) + free(namelist[i]); + free(namelist); + + return threads; +} + +struct thread_map *thread_map__new_by_tid(pid_t tid) +{ + struct thread_map *threads = malloc(sizeof(*threads) + sizeof(pid_t)); + + if (threads != NULL) { + threads->map[0] = tid; + threads->nr = 1; + } + + return threads; +} + +struct thread_map *thread_map__new_by_uid(uid_t uid) +{ + DIR *proc; + int max_threads = 32, items, i; + char path[256]; + struct dirent dirent, *next, **namelist = NULL; + struct thread_map *threads = malloc(sizeof(*threads) + + max_threads * sizeof(pid_t)); + if (threads == NULL) + goto out; + + proc = opendir("/proc"); + if (proc == NULL) + goto out_free_threads; + + threads->nr = 0; + + while (!readdir_r(proc, &dirent, &next) && next) { + char *end; + bool grow = false; + struct stat st; + pid_t pid = strtol(dirent.d_name, &end, 10); + + if (*end) /* only interested in proper numerical dirents */ + continue; + + snprintf(path, sizeof(path), "/proc/%s", dirent.d_name); + + if (stat(path, &st) != 0) + continue; + + if (st.st_uid != uid) + continue; + + snprintf(path, sizeof(path), "/proc/%d/task", pid); + items = scandir(path, &namelist, filter, NULL); + if (items <= 0) + goto out_free_closedir; + + while (threads->nr + items >= max_threads) { + max_threads *= 2; + grow = true; + } + + if (grow) { + struct thread_map *tmp; + + tmp = realloc(threads, (sizeof(*threads) + + max_threads * sizeof(pid_t))); + if (tmp == NULL) + goto out_free_namelist; + + threads = tmp; + } + + for (i = 0; i < items; i++) + threads->map[threads->nr + i] = atoi(namelist[i]->d_name); + + for (i = 0; i < items; i++) + free(namelist[i]); + free(namelist); + + threads->nr += items; + } + +out_closedir: + closedir(proc); +out: + return threads; + +out_free_threads: + free(threads); + return NULL; + +out_free_namelist: + for (i = 0; i < items; i++) + free(namelist[i]); + free(namelist); + +out_free_closedir: + free(threads); + threads = NULL; + goto out_closedir; +} + +struct thread_map *thread_map__new(pid_t pid, pid_t tid, uid_t uid) +{ + if (pid != -1) + return thread_map__new_by_pid(pid); + + if (tid == -1 && uid != UINT_MAX) + return thread_map__new_by_uid(uid); + + return thread_map__new_by_tid(tid); +} + +static struct thread_map *thread_map__new_by_pid_str(const char *pid_str) +{ + struct thread_map *threads = NULL, *nt; + char name[256]; + int items, total_tasks = 0; + struct dirent **namelist = NULL; + int i, j = 0; + pid_t pid, prev_pid = INT_MAX; + char *end_ptr; + struct str_node *pos; + struct strlist *slist = strlist__new(false, pid_str); + + if (!slist) + return NULL; + + strlist__for_each(pos, slist) { + pid = strtol(pos->s, &end_ptr, 10); + + if (pid == INT_MIN || pid == INT_MAX || + (*end_ptr != '\0' && *end_ptr != ',')) + goto out_free_threads; + + if (pid == prev_pid) + continue; + + sprintf(name, "/proc/%d/task", pid); + items = scandir(name, &namelist, filter, NULL); + if (items <= 0) + goto out_free_threads; + + total_tasks += items; + nt = realloc(threads, (sizeof(*threads) + + sizeof(pid_t) * total_tasks)); + if (nt == NULL) + goto out_free_namelist; + + threads = nt; + + for (i = 0; i < items; i++) { + threads->map[j++] = atoi(namelist[i]->d_name); + free(namelist[i]); + } + threads->nr = total_tasks; + free(namelist); + } + +out: + strlist__delete(slist); + return threads; + +out_free_namelist: + for (i = 0; i < items; i++) + free(namelist[i]); + free(namelist); + +out_free_threads: + free(threads); + threads = NULL; + goto out; +} + +static struct thread_map *thread_map__new_by_tid_str(const char *tid_str) +{ + struct thread_map *threads = NULL, *nt; + int ntasks = 0; + pid_t tid, prev_tid = INT_MAX; + char *end_ptr; + struct str_node *pos; + struct strlist *slist; + + /* perf-stat expects threads to be generated even if tid not given */ + if (!tid_str) { + threads = malloc(sizeof(*threads) + sizeof(pid_t)); + if (threads != NULL) { + threads->map[0] = -1; + threads->nr = 1; + } + return threads; + } + + slist = strlist__new(false, tid_str); + if (!slist) + return NULL; + + strlist__for_each(pos, slist) { + tid = strtol(pos->s, &end_ptr, 10); + + if (tid == INT_MIN || tid == INT_MAX || + (*end_ptr != '\0' && *end_ptr != ',')) + goto out_free_threads; + + if (tid == prev_tid) + continue; + + ntasks++; + nt = realloc(threads, sizeof(*threads) + sizeof(pid_t) * ntasks); + + if (nt == NULL) + goto out_free_threads; + + threads = nt; + threads->map[ntasks - 1] = tid; + threads->nr = ntasks; + } +out: + return threads; + +out_free_threads: + free(threads); + threads = NULL; + goto out; +} + +struct thread_map *thread_map__new_str(const char *pid, const char *tid, + uid_t uid) +{ + if (pid) + return thread_map__new_by_pid_str(pid); + + if (!tid && uid != UINT_MAX) + return thread_map__new_by_uid(uid); + + return thread_map__new_by_tid_str(tid); +} + +void thread_map__delete(struct thread_map *threads) +{ + free(threads); +} + +size_t thread_map__fprintf(struct thread_map *threads, FILE *fp) +{ + int i; + size_t printed = fprintf(fp, "%d thread%s: ", + threads->nr, threads->nr > 1 ? "s" : ""); + for (i = 0; i < threads->nr; ++i) + printed += fprintf(fp, "%s%d", i ? ", " : "", threads->map[i]); + + return printed + fprintf(fp, "\n"); +} diff --git a/tools/perf/util/thread_map.h b/tools/perf/util/thread_map.h new file mode 100644 index 00000000..f718df8a --- /dev/null +++ b/tools/perf/util/thread_map.h @@ -0,0 +1,24 @@ +#ifndef __PERF_THREAD_MAP_H +#define __PERF_THREAD_MAP_H + +#include <sys/types.h> +#include <stdio.h> + +struct thread_map { + int nr; + pid_t map[]; +}; + +struct thread_map *thread_map__new_by_pid(pid_t pid); +struct thread_map *thread_map__new_by_tid(pid_t tid); +struct thread_map *thread_map__new_by_uid(uid_t uid); +struct thread_map *thread_map__new(pid_t pid, pid_t tid, uid_t uid); + +struct thread_map *thread_map__new_str(const char *pid, + const char *tid, uid_t uid); + +void thread_map__delete(struct thread_map *threads); + +size_t thread_map__fprintf(struct thread_map *threads, FILE *fp); + +#endif /* __PERF_THREAD_MAP_H */ diff --git a/tools/perf/util/tool.h b/tools/perf/util/tool.h new file mode 100644 index 00000000..b0e1aadb --- /dev/null +++ b/tools/perf/util/tool.h @@ -0,0 +1,50 @@ +#ifndef __PERF_TOOL_H +#define __PERF_TOOL_H + +#include <stdbool.h> + +struct perf_session; +union perf_event; +struct perf_evlist; +struct perf_evsel; +struct perf_sample; +struct perf_tool; +struct machine; + +typedef int (*event_sample)(struct perf_tool *tool, union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, struct machine *machine); + +typedef int (*event_op)(struct perf_tool *tool, union perf_event *event, + struct perf_sample *sample, struct machine *machine); + +typedef int (*event_attr_op)(union perf_event *event, + struct perf_evlist **pevlist); +typedef int (*event_simple_op)(struct perf_tool *tool, union perf_event *event); + +typedef int (*event_synth_op)(union perf_event *event, + struct perf_session *session); + +typedef int (*event_op2)(struct perf_tool *tool, union perf_event *event, + struct perf_session *session); + +struct perf_tool { + event_sample sample, + read; + event_op mmap, + comm, + fork, + exit, + lost, + throttle, + unthrottle; + event_attr_op attr; + event_synth_op tracing_data; + event_simple_op event_type; + event_op2 finished_round, + build_id; + bool ordered_samples; + bool ordering_requires_timestamps; +}; + +#endif /* __PERF_TOOL_H */ diff --git a/tools/perf/util/top.c b/tools/perf/util/top.c new file mode 100644 index 00000000..54d37a47 --- /dev/null +++ b/tools/perf/util/top.c @@ -0,0 +1,106 @@ +/* + * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com> + * + * Refactored from builtin-top.c, see that files for further copyright notes. + * + * Released under the GPL v2. (and only v2, not any later version) + */ + +#include "cpumap.h" +#include "event.h" +#include "evlist.h" +#include "evsel.h" +#include "parse-events.h" +#include "symbol.h" +#include "top.h" +#include <inttypes.h> + +#define SNPRINTF(buf, size, fmt, args...) \ +({ \ + size_t r = snprintf(buf, size, fmt, ## args); \ + r > size ? size : r; \ +}) + +size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size) +{ + float samples_per_sec = top->samples / top->delay_secs; + float ksamples_per_sec = top->kernel_samples / top->delay_secs; + float esamples_percent = (100.0 * top->exact_samples) / top->samples; + struct perf_record_opts *opts = &top->record_opts; + struct perf_target *target = &opts->target; + size_t ret = 0; + + if (!perf_guest) { + ret = SNPRINTF(bf, size, + " PerfTop:%8.0f irqs/sec kernel:%4.1f%%" + " exact: %4.1f%% [", samples_per_sec, + 100.0 - (100.0 * ((samples_per_sec - ksamples_per_sec) / + samples_per_sec)), + esamples_percent); + } else { + float us_samples_per_sec = top->us_samples / top->delay_secs; + float guest_kernel_samples_per_sec = top->guest_kernel_samples / top->delay_secs; + float guest_us_samples_per_sec = top->guest_us_samples / top->delay_secs; + + ret = SNPRINTF(bf, size, + " PerfTop:%8.0f irqs/sec kernel:%4.1f%% us:%4.1f%%" + " guest kernel:%4.1f%% guest us:%4.1f%%" + " exact: %4.1f%% [", samples_per_sec, + 100.0 - (100.0 * ((samples_per_sec - ksamples_per_sec) / + samples_per_sec)), + 100.0 - (100.0 * ((samples_per_sec - us_samples_per_sec) / + samples_per_sec)), + 100.0 - (100.0 * ((samples_per_sec - + guest_kernel_samples_per_sec) / + samples_per_sec)), + 100.0 - (100.0 * ((samples_per_sec - + guest_us_samples_per_sec) / + samples_per_sec)), + esamples_percent); + } + + if (top->evlist->nr_entries == 1) { + struct perf_evsel *first = perf_evlist__first(top->evlist); + ret += SNPRINTF(bf + ret, size - ret, "%" PRIu64 "%s ", + (uint64_t)first->attr.sample_period, + opts->freq ? "Hz" : ""); + } + + ret += SNPRINTF(bf + ret, size - ret, "%s", perf_evsel__name(top->sym_evsel)); + + ret += SNPRINTF(bf + ret, size - ret, "], "); + + if (target->pid) + ret += SNPRINTF(bf + ret, size - ret, " (target_pid: %s", + target->pid); + else if (target->tid) + ret += SNPRINTF(bf + ret, size - ret, " (target_tid: %s", + target->tid); + else if (target->uid_str != NULL) + ret += SNPRINTF(bf + ret, size - ret, " (uid: %s", + target->uid_str); + else + ret += SNPRINTF(bf + ret, size - ret, " (all"); + + if (target->cpu_list) + ret += SNPRINTF(bf + ret, size - ret, ", CPU%s: %s)", + top->evlist->cpus->nr > 1 ? "s" : "", + target->cpu_list); + else { + if (target->tid) + ret += SNPRINTF(bf + ret, size - ret, ")"); + else + ret += SNPRINTF(bf + ret, size - ret, ", %d CPU%s)", + top->evlist->cpus->nr, + top->evlist->cpus->nr > 1 ? "s" : ""); + } + + return ret; +} + +void perf_top__reset_sample_counters(struct perf_top *top) +{ + top->samples = top->us_samples = top->kernel_samples = + top->exact_samples = top->guest_kernel_samples = + top->guest_us_samples = 0; +} diff --git a/tools/perf/util/top.h b/tools/perf/util/top.h new file mode 100644 index 00000000..7ebf357d --- /dev/null +++ b/tools/perf/util/top.h @@ -0,0 +1,44 @@ +#ifndef __PERF_TOP_H +#define __PERF_TOP_H 1 + +#include "tool.h" +#include "types.h" +#include <stddef.h> +#include <stdbool.h> +#include <termios.h> + +struct perf_evlist; +struct perf_evsel; +struct perf_session; + +struct perf_top { + struct perf_tool tool; + struct perf_evlist *evlist; + struct perf_record_opts record_opts; + /* + * Symbols will be added here in perf_event__process_sample and will + * get out after decayed. + */ + u64 samples; + u64 kernel_samples, us_samples; + u64 exact_samples; + u64 guest_us_samples, guest_kernel_samples; + int print_entries, count_filter, delay_secs; + bool hide_kernel_symbols, hide_user_symbols, zero; + bool use_tui, use_stdio; + bool sort_has_symbols; + bool kptr_restrict_warned; + bool vmlinux_warned; + bool dump_symtab; + struct hist_entry *sym_filter_entry; + struct perf_evsel *sym_evsel; + struct perf_session *session; + struct winsize winsize; + int realtime_prio; + int sym_pcnt_filter; + const char *sym_filter; +}; + +size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size); +void perf_top__reset_sample_counters(struct perf_top *top); +#endif /* __PERF_TOP_H */ diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c new file mode 100644 index 00000000..a8d81c35 --- /dev/null +++ b/tools/perf/util/trace-event-info.c @@ -0,0 +1,544 @@ +/* + * Copyright (C) 2008,2009, Steven Rostedt <srostedt@redhat.com> + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License (not later!) + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + */ +#include "util.h" +#include <dirent.h> +#include <mntent.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <stdarg.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/wait.h> +#include <pthread.h> +#include <fcntl.h> +#include <unistd.h> +#include <errno.h> +#include <stdbool.h> +#include <linux/list.h> +#include <linux/kernel.h> + +#include "../perf.h" +#include "trace-event.h" +#include "debugfs.h" +#include "evsel.h" + +#define VERSION "0.5" + +#define TRACE_CTRL "tracing_on" +#define TRACE "trace" +#define AVAILABLE "available_tracers" +#define CURRENT "current_tracer" +#define ITER_CTRL "trace_options" +#define MAX_LATENCY "tracing_max_latency" + +unsigned int page_size; + +static const char *output_file = "trace.info"; +static int output_fd; + +struct event_list { + struct event_list *next; + const char *event; +}; + +struct events { + struct events *sibling; + struct events *children; + struct events *next; + char *name; +}; + + +static void *malloc_or_die(unsigned int size) +{ + void *data; + + data = malloc(size); + if (!data) + die("malloc"); + return data; +} + +static const char *find_debugfs(void) +{ + const char *path = debugfs_mount(NULL); + + if (!path) + die("Your kernel not support debugfs filesystem"); + + return path; +} + +/* + * Finds the path to the debugfs/tracing + * Allocates the string and stores it. + */ +static const char *find_tracing_dir(void) +{ + static char *tracing; + static int tracing_found; + const char *debugfs; + + if (tracing_found) + return tracing; + + debugfs = find_debugfs(); + + tracing = malloc_or_die(strlen(debugfs) + 9); + + sprintf(tracing, "%s/tracing", debugfs); + + tracing_found = 1; + return tracing; +} + +static char *get_tracing_file(const char *name) +{ + const char *tracing; + char *file; + + tracing = find_tracing_dir(); + if (!tracing) + return NULL; + + file = malloc_or_die(strlen(tracing) + strlen(name) + 2); + + sprintf(file, "%s/%s", tracing, name); + return file; +} + +static void put_tracing_file(char *file) +{ + free(file); +} + +static ssize_t calc_data_size; + +static ssize_t write_or_die(const void *buf, size_t len) +{ + int ret; + + if (calc_data_size) { + calc_data_size += len; + return len; + } + + ret = write(output_fd, buf, len); + if (ret < 0) + die("writing to '%s'", output_file); + + return ret; +} + +int bigendian(void) +{ + unsigned char str[] = { 0x1, 0x2, 0x3, 0x4, 0x0, 0x0, 0x0, 0x0}; + unsigned int *ptr; + + ptr = (unsigned int *)(void *)str; + return *ptr == 0x01020304; +} + +/* unfortunately, you can not stat debugfs or proc files for size */ +static void record_file(const char *file, size_t hdr_sz) +{ + unsigned long long size = 0; + char buf[BUFSIZ], *sizep; + off_t hdr_pos = lseek(output_fd, 0, SEEK_CUR); + int r, fd; + + fd = open(file, O_RDONLY); + if (fd < 0) + die("Can't read '%s'", file); + + /* put in zeros for file size, then fill true size later */ + if (hdr_sz) + write_or_die(&size, hdr_sz); + + do { + r = read(fd, buf, BUFSIZ); + if (r > 0) { + size += r; + write_or_die(buf, r); + } + } while (r > 0); + close(fd); + + /* ugh, handle big-endian hdr_size == 4 */ + sizep = (char*)&size; + if (bigendian()) + sizep += sizeof(u64) - hdr_sz; + + if (hdr_sz && pwrite(output_fd, sizep, hdr_sz, hdr_pos) < 0) + die("writing to %s", output_file); +} + +static void read_header_files(void) +{ + char *path; + struct stat st; + + path = get_tracing_file("events/header_page"); + if (stat(path, &st) < 0) + die("can't read '%s'", path); + + write_or_die("header_page", 12); + record_file(path, 8); + put_tracing_file(path); + + path = get_tracing_file("events/header_event"); + if (stat(path, &st) < 0) + die("can't read '%s'", path); + + write_or_die("header_event", 13); + record_file(path, 8); + put_tracing_file(path); +} + +static bool name_in_tp_list(char *sys, struct tracepoint_path *tps) +{ + while (tps) { + if (!strcmp(sys, tps->name)) + return true; + tps = tps->next; + } + + return false; +} + +static void copy_event_system(const char *sys, struct tracepoint_path *tps) +{ + struct dirent *dent; + struct stat st; + char *format; + DIR *dir; + int count = 0; + int ret; + + dir = opendir(sys); + if (!dir) + die("can't read directory '%s'", sys); + + while ((dent = readdir(dir))) { + if (dent->d_type != DT_DIR || + strcmp(dent->d_name, ".") == 0 || + strcmp(dent->d_name, "..") == 0 || + !name_in_tp_list(dent->d_name, tps)) + continue; + format = malloc_or_die(strlen(sys) + strlen(dent->d_name) + 10); + sprintf(format, "%s/%s/format", sys, dent->d_name); + ret = stat(format, &st); + free(format); + if (ret < 0) + continue; + count++; + } + + write_or_die(&count, 4); + + rewinddir(dir); + while ((dent = readdir(dir))) { + if (dent->d_type != DT_DIR || + strcmp(dent->d_name, ".") == 0 || + strcmp(dent->d_name, "..") == 0 || + !name_in_tp_list(dent->d_name, tps)) + continue; + format = malloc_or_die(strlen(sys) + strlen(dent->d_name) + 10); + sprintf(format, "%s/%s/format", sys, dent->d_name); + ret = stat(format, &st); + + if (ret >= 0) + record_file(format, 8); + + free(format); + } + closedir(dir); +} + +static void read_ftrace_files(struct tracepoint_path *tps) +{ + char *path; + + path = get_tracing_file("events/ftrace"); + + copy_event_system(path, tps); + + put_tracing_file(path); +} + +static bool system_in_tp_list(char *sys, struct tracepoint_path *tps) +{ + while (tps) { + if (!strcmp(sys, tps->system)) + return true; + tps = tps->next; + } + + return false; +} + +static void read_event_files(struct tracepoint_path *tps) +{ + struct dirent *dent; + struct stat st; + char *path; + char *sys; + DIR *dir; + int count = 0; + int ret; + + path = get_tracing_file("events"); + + dir = opendir(path); + if (!dir) + die("can't read directory '%s'", path); + + while ((dent = readdir(dir))) { + if (dent->d_type != DT_DIR || + strcmp(dent->d_name, ".") == 0 || + strcmp(dent->d_name, "..") == 0 || + strcmp(dent->d_name, "ftrace") == 0 || + !system_in_tp_list(dent->d_name, tps)) + continue; + count++; + } + + write_or_die(&count, 4); + + rewinddir(dir); + while ((dent = readdir(dir))) { + if (dent->d_type != DT_DIR || + strcmp(dent->d_name, ".") == 0 || + strcmp(dent->d_name, "..") == 0 || + strcmp(dent->d_name, "ftrace") == 0 || + !system_in_tp_list(dent->d_name, tps)) + continue; + sys = malloc_or_die(strlen(path) + strlen(dent->d_name) + 2); + sprintf(sys, "%s/%s", path, dent->d_name); + ret = stat(sys, &st); + if (ret >= 0) { + write_or_die(dent->d_name, strlen(dent->d_name) + 1); + copy_event_system(sys, tps); + } + free(sys); + } + + closedir(dir); + put_tracing_file(path); +} + +static void read_proc_kallsyms(void) +{ + unsigned int size; + const char *path = "/proc/kallsyms"; + struct stat st; + int ret; + + ret = stat(path, &st); + if (ret < 0) { + /* not found */ + size = 0; + write_or_die(&size, 4); + return; + } + record_file(path, 4); +} + +static void read_ftrace_printk(void) +{ + unsigned int size; + char *path; + struct stat st; + int ret; + + path = get_tracing_file("printk_formats"); + ret = stat(path, &st); + if (ret < 0) { + /* not found */ + size = 0; + write_or_die(&size, 4); + goto out; + } + record_file(path, 4); + +out: + put_tracing_file(path); +} + +static struct tracepoint_path * +get_tracepoints_path(struct list_head *pattrs) +{ + struct tracepoint_path path, *ppath = &path; + struct perf_evsel *pos; + int nr_tracepoints = 0; + + list_for_each_entry(pos, pattrs, node) { + if (pos->attr.type != PERF_TYPE_TRACEPOINT) + continue; + ++nr_tracepoints; + ppath->next = tracepoint_id_to_path(pos->attr.config); + if (!ppath->next) + die("%s\n", "No memory to alloc tracepoints list"); + ppath = ppath->next; + } + + return nr_tracepoints > 0 ? path.next : NULL; +} + +static void +put_tracepoints_path(struct tracepoint_path *tps) +{ + while (tps) { + struct tracepoint_path *t = tps; + + tps = tps->next; + free(t->name); + free(t->system); + free(t); + } +} + +bool have_tracepoints(struct list_head *pattrs) +{ + struct perf_evsel *pos; + + list_for_each_entry(pos, pattrs, node) + if (pos->attr.type == PERF_TYPE_TRACEPOINT) + return true; + + return false; +} + +static void tracing_data_header(void) +{ + char buf[20]; + + /* just guessing this is someone's birthday.. ;) */ + buf[0] = 23; + buf[1] = 8; + buf[2] = 68; + memcpy(buf + 3, "tracing", 7); + + write_or_die(buf, 10); + + write_or_die(VERSION, strlen(VERSION) + 1); + + /* save endian */ + if (bigendian()) + buf[0] = 1; + else + buf[0] = 0; + + read_trace_init(buf[0], buf[0]); + + write_or_die(buf, 1); + + /* save size of long */ + buf[0] = sizeof(long); + write_or_die(buf, 1); + + /* save page_size */ + page_size = sysconf(_SC_PAGESIZE); + write_or_die(&page_size, 4); +} + +struct tracing_data *tracing_data_get(struct list_head *pattrs, + int fd, bool temp) +{ + struct tracepoint_path *tps; + struct tracing_data *tdata; + + output_fd = fd; + + tps = get_tracepoints_path(pattrs); + if (!tps) + return NULL; + + tdata = malloc_or_die(sizeof(*tdata)); + tdata->temp = temp; + tdata->size = 0; + + if (temp) { + int temp_fd; + + snprintf(tdata->temp_file, sizeof(tdata->temp_file), + "/tmp/perf-XXXXXX"); + if (!mkstemp(tdata->temp_file)) + die("Can't make temp file"); + + temp_fd = open(tdata->temp_file, O_RDWR); + if (temp_fd < 0) + die("Can't read '%s'", tdata->temp_file); + + /* + * Set the temp file the default output, so all the + * tracing data are stored into it. + */ + output_fd = temp_fd; + } + + tracing_data_header(); + read_header_files(); + read_ftrace_files(tps); + read_event_files(tps); + read_proc_kallsyms(); + read_ftrace_printk(); + + /* + * All tracing data are stored by now, we can restore + * the default output file in case we used temp file. + */ + if (temp) { + tdata->size = lseek(output_fd, 0, SEEK_CUR); + close(output_fd); + output_fd = fd; + } + + put_tracepoints_path(tps); + return tdata; +} + +void tracing_data_put(struct tracing_data *tdata) +{ + if (tdata->temp) { + record_file(tdata->temp_file, 0); + unlink(tdata->temp_file); + } + + free(tdata); +} + +int read_tracing_data(int fd, struct list_head *pattrs) +{ + struct tracing_data *tdata; + + /* + * We work over the real file, so we can write data + * directly, no temp file is needed. + */ + tdata = tracing_data_get(pattrs, fd, false); + if (!tdata) + return -ENOMEM; + + tracing_data_put(tdata); + return 0; +} diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c new file mode 100644 index 00000000..3aabcd68 --- /dev/null +++ b/tools/perf/util/trace-event-parse.c @@ -0,0 +1,354 @@ +/* + * Copyright (C) 2009, Steven Rostedt <srostedt@redhat.com> + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License (not later!) + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + */ +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <ctype.h> +#include <errno.h> + +#include "../perf.h" +#include "util.h" +#include "trace-event.h" + +int header_page_size_size; +int header_page_ts_size; +int header_page_data_offset; + +bool latency_format; + +struct pevent *read_trace_init(int file_bigendian, int host_bigendian) +{ + struct pevent *pevent = pevent_alloc(); + + if (pevent != NULL) { + pevent_set_flag(pevent, PEVENT_NSEC_OUTPUT); + pevent_set_file_bigendian(pevent, file_bigendian); + pevent_set_host_bigendian(pevent, host_bigendian); + } + + return pevent; +} + +static int get_common_field(struct scripting_context *context, + int *offset, int *size, const char *type) +{ + struct pevent *pevent = context->pevent; + struct event_format *event; + struct format_field *field; + + if (!*size) { + if (!pevent->events) + return 0; + + event = pevent->events[0]; + field = pevent_find_common_field(event, type); + if (!field) + return 0; + *offset = field->offset; + *size = field->size; + } + + return pevent_read_number(pevent, context->event_data + *offset, *size); +} + +int common_lock_depth(struct scripting_context *context) +{ + static int offset; + static int size; + int ret; + + ret = get_common_field(context, &size, &offset, + "common_lock_depth"); + if (ret < 0) + return -1; + + return ret; +} + +int common_flags(struct scripting_context *context) +{ + static int offset; + static int size; + int ret; + + ret = get_common_field(context, &size, &offset, + "common_flags"); + if (ret < 0) + return -1; + + return ret; +} + +int common_pc(struct scripting_context *context) +{ + static int offset; + static int size; + int ret; + + ret = get_common_field(context, &size, &offset, + "common_preempt_count"); + if (ret < 0) + return -1; + + return ret; +} + +unsigned long long +raw_field_value(struct event_format *event, const char *name, void *data) +{ + struct format_field *field; + unsigned long long val; + + field = pevent_find_any_field(event, name); + if (!field) + return 0ULL; + + pevent_read_number_field(field, data, &val); + + return val; +} + +void *raw_field_ptr(struct event_format *event, const char *name, void *data) +{ + struct format_field *field; + + field = pevent_find_any_field(event, name); + if (!field) + return NULL; + + if (field->flags & FIELD_IS_DYNAMIC) { + int offset; + + offset = *(int *)(data + field->offset); + offset &= 0xffff; + + return data + offset; + } + + return data + field->offset; +} + +int trace_parse_common_type(struct pevent *pevent, void *data) +{ + struct pevent_record record; + + record.data = data; + return pevent_data_type(pevent, &record); +} + +int trace_parse_common_pid(struct pevent *pevent, void *data) +{ + struct pevent_record record; + + record.data = data; + return pevent_data_pid(pevent, &record); +} + +unsigned long long read_size(struct event_format *event, void *ptr, int size) +{ + return pevent_read_number(event->pevent, ptr, size); +} + +void event_format__print(struct event_format *event, + int cpu, void *data, int size) +{ + struct pevent_record record; + struct trace_seq s; + + memset(&record, 0, sizeof(record)); + record.cpu = cpu; + record.size = size; + record.data = data; + + trace_seq_init(&s); + pevent_event_info(&s, event, &record); + trace_seq_do_printf(&s); +} + +void print_trace_event(struct pevent *pevent, int cpu, void *data, int size) +{ + int type = trace_parse_common_type(pevent, data); + struct event_format *event = pevent_find_event(pevent, type); + + if (!event) { + warning("ug! no event found for type %d", type); + return; + } + + event_format__print(event, cpu, data, size); +} + +void print_event(struct pevent *pevent, int cpu, void *data, int size, + unsigned long long nsecs, char *comm) +{ + struct pevent_record record; + struct trace_seq s; + int pid; + + pevent->latency_format = latency_format; + + record.ts = nsecs; + record.cpu = cpu; + record.size = size; + record.data = data; + pid = pevent_data_pid(pevent, &record); + + if (!pevent_pid_is_registered(pevent, pid)) + pevent_register_comm(pevent, comm, pid); + + trace_seq_init(&s); + pevent_print_event(pevent, &s, &record); + trace_seq_do_printf(&s); + printf("\n"); +} + +void parse_proc_kallsyms(struct pevent *pevent, + char *file, unsigned int size __maybe_unused) +{ + unsigned long long addr; + char *func; + char *line; + char *next = NULL; + char *addr_str; + char *mod; + char *fmt; + + line = strtok_r(file, "\n", &next); + while (line) { + mod = NULL; + addr_str = strtok_r(line, " ", &fmt); + addr = strtoull(addr_str, NULL, 16); + /* skip character */ + strtok_r(NULL, " ", &fmt); + func = strtok_r(NULL, "\t", &fmt); + mod = strtok_r(NULL, "]", &fmt); + /* truncate the extra '[' */ + if (mod) + mod = mod + 1; + + pevent_register_function(pevent, func, addr, mod); + + line = strtok_r(NULL, "\n", &next); + } +} + +void parse_ftrace_printk(struct pevent *pevent, + char *file, unsigned int size __maybe_unused) +{ + unsigned long long addr; + char *printk; + char *line; + char *next = NULL; + char *addr_str; + char *fmt; + + line = strtok_r(file, "\n", &next); + while (line) { + addr_str = strtok_r(line, ":", &fmt); + if (!addr_str) { + warning("printk format with empty entry"); + break; + } + addr = strtoull(addr_str, NULL, 16); + /* fmt still has a space, skip it */ + printk = strdup(fmt+1); + line = strtok_r(NULL, "\n", &next); + pevent_register_print_string(pevent, printk, addr); + } +} + +int parse_ftrace_file(struct pevent *pevent, char *buf, unsigned long size) +{ + return pevent_parse_event(pevent, buf, size, "ftrace"); +} + +int parse_event_file(struct pevent *pevent, + char *buf, unsigned long size, char *sys) +{ + return pevent_parse_event(pevent, buf, size, sys); +} + +struct event_format *trace_find_next_event(struct pevent *pevent, + struct event_format *event) +{ + static int idx; + + if (!pevent || !pevent->events) + return NULL; + + if (!event) { + idx = 0; + return pevent->events[0]; + } + + if (idx < pevent->nr_events && event == pevent->events[idx]) { + idx++; + if (idx == pevent->nr_events) + return NULL; + return pevent->events[idx]; + } + + for (idx = 1; idx < pevent->nr_events; idx++) { + if (event == pevent->events[idx - 1]) + return pevent->events[idx]; + } + return NULL; +} + +struct flag { + const char *name; + unsigned long long value; +}; + +static const struct flag flags[] = { + { "HI_SOFTIRQ", 0 }, + { "TIMER_SOFTIRQ", 1 }, + { "NET_TX_SOFTIRQ", 2 }, + { "NET_RX_SOFTIRQ", 3 }, + { "BLOCK_SOFTIRQ", 4 }, + { "BLOCK_IOPOLL_SOFTIRQ", 5 }, + { "TASKLET_SOFTIRQ", 6 }, + { "SCHED_SOFTIRQ", 7 }, + { "HRTIMER_SOFTIRQ", 8 }, + { "RCU_SOFTIRQ", 9 }, + + { "HRTIMER_NORESTART", 0 }, + { "HRTIMER_RESTART", 1 }, +}; + +unsigned long long eval_flag(const char *flag) +{ + int i; + + /* + * Some flags in the format files do not get converted. + * If the flag is not numeric, see if it is something that + * we already know about. + */ + if (isdigit(flag[0])) + return strtoull(flag, NULL, 0); + + for (i = 0; i < (int)(sizeof(flags)/sizeof(flags[0])); i++) + if (strcmp(flags[i].name, flag) == 0) + return flags[i].value; + + return 0; +} diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c new file mode 100644 index 00000000..37415726 --- /dev/null +++ b/tools/perf/util/trace-event-read.c @@ -0,0 +1,551 @@ +/* + * Copyright (C) 2009, Steven Rostedt <srostedt@redhat.com> + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License (not later!) + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + */ +#define _FILE_OFFSET_BITS 64 + +#include <dirent.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <getopt.h> +#include <stdarg.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/wait.h> +#include <sys/mman.h> +#include <pthread.h> +#include <fcntl.h> +#include <unistd.h> +#include <errno.h> + +#include "../perf.h" +#include "util.h" +#include "trace-event.h" + +static int input_fd; + +static int read_page; + +int file_bigendian; +int host_bigendian; +static int long_size; + +static ssize_t calc_data_size; +static bool repipe; + +static void *malloc_or_die(int size) +{ + void *ret; + + ret = malloc(size); + if (!ret) + die("malloc"); + return ret; +} + +static int do_read(int fd, void *buf, int size) +{ + int rsize = size; + + while (size) { + int ret = read(fd, buf, size); + + if (ret <= 0) + return -1; + + if (repipe) { + int retw = write(STDOUT_FILENO, buf, ret); + + if (retw <= 0 || retw != ret) + die("repiping input file"); + } + + size -= ret; + buf += ret; + } + + return rsize; +} + +static int read_or_die(void *data, int size) +{ + int r; + + r = do_read(input_fd, data, size); + if (r <= 0) + die("reading input file (size expected=%d received=%d)", + size, r); + + if (calc_data_size) + calc_data_size += r; + + return r; +} + +/* If it fails, the next read will report it */ +static void skip(int size) +{ + char buf[BUFSIZ]; + int r; + + while (size) { + r = size > BUFSIZ ? BUFSIZ : size; + read_or_die(buf, r); + size -= r; + }; +} + +static unsigned int read4(struct pevent *pevent) +{ + unsigned int data; + + read_or_die(&data, 4); + return __data2host4(pevent, data); +} + +static unsigned long long read8(struct pevent *pevent) +{ + unsigned long long data; + + read_or_die(&data, 8); + return __data2host8(pevent, data); +} + +static char *read_string(void) +{ + char buf[BUFSIZ]; + char *str = NULL; + int size = 0; + off_t r; + char c; + + for (;;) { + r = read(input_fd, &c, 1); + if (r < 0) + die("reading input file"); + + if (!r) + die("no data"); + + if (repipe) { + int retw = write(STDOUT_FILENO, &c, 1); + + if (retw <= 0 || retw != r) + die("repiping input file string"); + } + + buf[size++] = c; + + if (!c) + break; + } + + if (calc_data_size) + calc_data_size += size; + + str = malloc_or_die(size); + memcpy(str, buf, size); + + return str; +} + +static void read_proc_kallsyms(struct pevent *pevent) +{ + unsigned int size; + char *buf; + + size = read4(pevent); + if (!size) + return; + + buf = malloc_or_die(size + 1); + read_or_die(buf, size); + buf[size] = '\0'; + + parse_proc_kallsyms(pevent, buf, size); + + free(buf); +} + +static void read_ftrace_printk(struct pevent *pevent) +{ + unsigned int size; + char *buf; + + size = read4(pevent); + if (!size) + return; + + buf = malloc_or_die(size); + read_or_die(buf, size); + + parse_ftrace_printk(pevent, buf, size); + + free(buf); +} + +static void read_header_files(struct pevent *pevent) +{ + unsigned long long size; + char *header_event; + char buf[BUFSIZ]; + + read_or_die(buf, 12); + + if (memcmp(buf, "header_page", 12) != 0) + die("did not read header page"); + + size = read8(pevent); + skip(size); + + /* + * The size field in the page is of type long, + * use that instead, since it represents the kernel. + */ + long_size = header_page_size_size; + + read_or_die(buf, 13); + if (memcmp(buf, "header_event", 13) != 0) + die("did not read header event"); + + size = read8(pevent); + header_event = malloc_or_die(size); + read_or_die(header_event, size); + free(header_event); +} + +static void read_ftrace_file(struct pevent *pevent, unsigned long long size) +{ + char *buf; + + buf = malloc_or_die(size); + read_or_die(buf, size); + parse_ftrace_file(pevent, buf, size); + free(buf); +} + +static void read_event_file(struct pevent *pevent, char *sys, + unsigned long long size) +{ + char *buf; + + buf = malloc_or_die(size); + read_or_die(buf, size); + parse_event_file(pevent, buf, size, sys); + free(buf); +} + +static void read_ftrace_files(struct pevent *pevent) +{ + unsigned long long size; + int count; + int i; + + count = read4(pevent); + + for (i = 0; i < count; i++) { + size = read8(pevent); + read_ftrace_file(pevent, size); + } +} + +static void read_event_files(struct pevent *pevent) +{ + unsigned long long size; + char *sys; + int systems; + int count; + int i,x; + + systems = read4(pevent); + + for (i = 0; i < systems; i++) { + sys = read_string(); + + count = read4(pevent); + for (x=0; x < count; x++) { + size = read8(pevent); + read_event_file(pevent, sys, size); + } + } +} + +struct cpu_data { + unsigned long long offset; + unsigned long long size; + unsigned long long timestamp; + struct pevent_record *next; + char *page; + int cpu; + int index; + int page_size; +}; + +static struct cpu_data *cpu_data; + +static void update_cpu_data_index(int cpu) +{ + cpu_data[cpu].offset += page_size; + cpu_data[cpu].size -= page_size; + cpu_data[cpu].index = 0; +} + +static void get_next_page(int cpu) +{ + off_t save_seek; + off_t ret; + + if (!cpu_data[cpu].page) + return; + + if (read_page) { + if (cpu_data[cpu].size <= page_size) { + free(cpu_data[cpu].page); + cpu_data[cpu].page = NULL; + return; + } + + update_cpu_data_index(cpu); + + /* other parts of the code may expect the pointer to not move */ + save_seek = lseek(input_fd, 0, SEEK_CUR); + + ret = lseek(input_fd, cpu_data[cpu].offset, SEEK_SET); + if (ret == (off_t)-1) + die("failed to lseek"); + ret = read(input_fd, cpu_data[cpu].page, page_size); + if (ret < 0) + die("failed to read page"); + + /* reset the file pointer back */ + lseek(input_fd, save_seek, SEEK_SET); + + return; + } + + munmap(cpu_data[cpu].page, page_size); + cpu_data[cpu].page = NULL; + + if (cpu_data[cpu].size <= page_size) + return; + + update_cpu_data_index(cpu); + + cpu_data[cpu].page = mmap(NULL, page_size, PROT_READ, MAP_PRIVATE, + input_fd, cpu_data[cpu].offset); + if (cpu_data[cpu].page == MAP_FAILED) + die("failed to mmap cpu %d at offset 0x%llx", + cpu, cpu_data[cpu].offset); +} + +static unsigned int type_len4host(unsigned int type_len_ts) +{ + if (file_bigendian) + return (type_len_ts >> 27) & ((1 << 5) - 1); + else + return type_len_ts & ((1 << 5) - 1); +} + +static unsigned int ts4host(unsigned int type_len_ts) +{ + if (file_bigendian) + return type_len_ts & ((1 << 27) - 1); + else + return type_len_ts >> 5; +} + +static int calc_index(void *ptr, int cpu) +{ + return (unsigned long)ptr - (unsigned long)cpu_data[cpu].page; +} + +struct pevent_record *trace_peek_data(struct pevent *pevent, int cpu) +{ + struct pevent_record *data; + void *page = cpu_data[cpu].page; + int idx = cpu_data[cpu].index; + void *ptr = page + idx; + unsigned long long extend; + unsigned int type_len_ts; + unsigned int type_len; + unsigned int delta; + unsigned int length = 0; + + if (cpu_data[cpu].next) + return cpu_data[cpu].next; + + if (!page) + return NULL; + + if (!idx) { + /* FIXME: handle header page */ + if (header_page_ts_size != 8) + die("expected a long long type for timestamp"); + cpu_data[cpu].timestamp = data2host8(pevent, ptr); + ptr += 8; + switch (header_page_size_size) { + case 4: + cpu_data[cpu].page_size = data2host4(pevent, ptr); + ptr += 4; + break; + case 8: + cpu_data[cpu].page_size = data2host8(pevent, ptr); + ptr += 8; + break; + default: + die("bad long size"); + } + ptr = cpu_data[cpu].page + header_page_data_offset; + } + +read_again: + idx = calc_index(ptr, cpu); + + if (idx >= cpu_data[cpu].page_size) { + get_next_page(cpu); + return trace_peek_data(pevent, cpu); + } + + type_len_ts = data2host4(pevent, ptr); + ptr += 4; + + type_len = type_len4host(type_len_ts); + delta = ts4host(type_len_ts); + + switch (type_len) { + case RINGBUF_TYPE_PADDING: + if (!delta) + die("error, hit unexpected end of page"); + length = data2host4(pevent, ptr); + ptr += 4; + length *= 4; + ptr += length; + goto read_again; + + case RINGBUF_TYPE_TIME_EXTEND: + extend = data2host4(pevent, ptr); + ptr += 4; + extend <<= TS_SHIFT; + extend += delta; + cpu_data[cpu].timestamp += extend; + goto read_again; + + case RINGBUF_TYPE_TIME_STAMP: + ptr += 12; + break; + case 0: + length = data2host4(pevent, ptr); + ptr += 4; + die("here! length=%d", length); + break; + default: + length = type_len * 4; + break; + } + + cpu_data[cpu].timestamp += delta; + + data = malloc_or_die(sizeof(*data)); + memset(data, 0, sizeof(*data)); + + data->ts = cpu_data[cpu].timestamp; + data->size = length; + data->data = ptr; + ptr += length; + + cpu_data[cpu].index = calc_index(ptr, cpu); + cpu_data[cpu].next = data; + + return data; +} + +struct pevent_record *trace_read_data(struct pevent *pevent, int cpu) +{ + struct pevent_record *data; + + data = trace_peek_data(pevent, cpu); + cpu_data[cpu].next = NULL; + + return data; +} + +ssize_t trace_report(int fd, struct pevent **ppevent, bool __repipe) +{ + char buf[BUFSIZ]; + char test[] = { 23, 8, 68 }; + char *version; + int show_version = 0; + int show_funcs = 0; + int show_printk = 0; + ssize_t size; + + calc_data_size = 1; + repipe = __repipe; + + input_fd = fd; + + read_or_die(buf, 3); + if (memcmp(buf, test, 3) != 0) + die("no trace data in the file"); + + read_or_die(buf, 7); + if (memcmp(buf, "tracing", 7) != 0) + die("not a trace file (missing 'tracing' tag)"); + + version = read_string(); + if (show_version) + printf("version = %s\n", version); + free(version); + + read_or_die(buf, 1); + file_bigendian = buf[0]; + host_bigendian = bigendian(); + + *ppevent = read_trace_init(file_bigendian, host_bigendian); + if (*ppevent == NULL) + die("read_trace_init failed"); + + read_or_die(buf, 1); + long_size = buf[0]; + + page_size = read4(*ppevent); + + read_header_files(*ppevent); + + read_ftrace_files(*ppevent); + read_event_files(*ppevent); + read_proc_kallsyms(*ppevent); + read_ftrace_printk(*ppevent); + + size = calc_data_size - 1; + calc_data_size = 0; + repipe = false; + + if (show_funcs) { + pevent_print_funcs(*ppevent); + return size; + } + if (show_printk) { + pevent_print_printk(*ppevent); + return size; + } + + return size; +} diff --git a/tools/perf/util/trace-event-scripting.c b/tools/perf/util/trace-event-scripting.c new file mode 100644 index 00000000..8715a100 --- /dev/null +++ b/tools/perf/util/trace-event-scripting.c @@ -0,0 +1,171 @@ +/* + * trace-event-scripting. Scripting engine common and initialization code. + * + * Copyright (C) 2009-2010 Tom Zanussi <tzanussi@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <errno.h> + +#include "../perf.h" +#include "util.h" +#include "trace-event.h" + +struct scripting_context *scripting_context; + +static int stop_script_unsupported(void) +{ + return 0; +} + +static void process_event_unsupported(union perf_event *event __maybe_unused, + struct perf_sample *sample __maybe_unused, + struct perf_evsel *evsel __maybe_unused, + struct machine *machine __maybe_unused, + struct addr_location *al __maybe_unused) +{ +} + +static void print_python_unsupported_msg(void) +{ + fprintf(stderr, "Python scripting not supported." + " Install libpython and rebuild perf to enable it.\n" + "For example:\n # apt-get install python-dev (ubuntu)" + "\n # yum install python-devel (Fedora)" + "\n etc.\n"); +} + +static int python_start_script_unsupported(const char *script __maybe_unused, + int argc __maybe_unused, + const char **argv __maybe_unused) +{ + print_python_unsupported_msg(); + + return -1; +} + +static int python_generate_script_unsupported(struct pevent *pevent + __maybe_unused, + const char *outfile + __maybe_unused) +{ + print_python_unsupported_msg(); + + return -1; +} + +struct scripting_ops python_scripting_unsupported_ops = { + .name = "Python", + .start_script = python_start_script_unsupported, + .stop_script = stop_script_unsupported, + .process_event = process_event_unsupported, + .generate_script = python_generate_script_unsupported, +}; + +static void register_python_scripting(struct scripting_ops *scripting_ops) +{ + int err; + err = script_spec_register("Python", scripting_ops); + if (err) + die("error registering Python script extension"); + + err = script_spec_register("py", scripting_ops); + if (err) + die("error registering py script extension"); + + scripting_context = malloc(sizeof(struct scripting_context)); +} + +#ifdef NO_LIBPYTHON +void setup_python_scripting(void) +{ + register_python_scripting(&python_scripting_unsupported_ops); +} +#else +extern struct scripting_ops python_scripting_ops; + +void setup_python_scripting(void) +{ + register_python_scripting(&python_scripting_ops); +} +#endif + +static void print_perl_unsupported_msg(void) +{ + fprintf(stderr, "Perl scripting not supported." + " Install libperl and rebuild perf to enable it.\n" + "For example:\n # apt-get install libperl-dev (ubuntu)" + "\n # yum install 'perl(ExtUtils::Embed)' (Fedora)" + "\n etc.\n"); +} + +static int perl_start_script_unsupported(const char *script __maybe_unused, + int argc __maybe_unused, + const char **argv __maybe_unused) +{ + print_perl_unsupported_msg(); + + return -1; +} + +static int perl_generate_script_unsupported(struct pevent *pevent + __maybe_unused, + const char *outfile __maybe_unused) +{ + print_perl_unsupported_msg(); + + return -1; +} + +struct scripting_ops perl_scripting_unsupported_ops = { + .name = "Perl", + .start_script = perl_start_script_unsupported, + .stop_script = stop_script_unsupported, + .process_event = process_event_unsupported, + .generate_script = perl_generate_script_unsupported, +}; + +static void register_perl_scripting(struct scripting_ops *scripting_ops) +{ + int err; + err = script_spec_register("Perl", scripting_ops); + if (err) + die("error registering Perl script extension"); + + err = script_spec_register("pl", scripting_ops); + if (err) + die("error registering pl script extension"); + + scripting_context = malloc(sizeof(struct scripting_context)); +} + +#ifdef NO_LIBPERL +void setup_perl_scripting(void) +{ + register_perl_scripting(&perl_scripting_unsupported_ops); +} +#else +extern struct scripting_ops perl_scripting_ops; + +void setup_perl_scripting(void) +{ + register_perl_scripting(&perl_scripting_ops); +} +#endif diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h new file mode 100644 index 00000000..a55fd37f --- /dev/null +++ b/tools/perf/util/trace-event.h @@ -0,0 +1,108 @@ +#ifndef _PERF_UTIL_TRACE_EVENT_H +#define _PERF_UTIL_TRACE_EVENT_H + +#include "parse-events.h" +#include "event-parse.h" +#include "session.h" + +struct machine; +struct perf_sample; +union perf_event; +struct perf_tool; + +extern int header_page_size_size; +extern int header_page_ts_size; +extern int header_page_data_offset; + +extern bool latency_format; +extern struct pevent *perf_pevent; + +enum { + RINGBUF_TYPE_PADDING = 29, + RINGBUF_TYPE_TIME_EXTEND = 30, + RINGBUF_TYPE_TIME_STAMP = 31, +}; + +#ifndef TS_SHIFT +#define TS_SHIFT 27 +#endif + +int bigendian(void); + +struct pevent *read_trace_init(int file_bigendian, int host_bigendian); +void print_trace_event(struct pevent *pevent, int cpu, void *data, int size); +void event_format__print(struct event_format *event, + int cpu, void *data, int size); + +void print_event(struct pevent *pevent, int cpu, void *data, int size, + unsigned long long nsecs, char *comm); + +int parse_ftrace_file(struct pevent *pevent, char *buf, unsigned long size); +int parse_event_file(struct pevent *pevent, + char *buf, unsigned long size, char *sys); + +struct pevent_record *trace_peek_data(struct pevent *pevent, int cpu); + +unsigned long long +raw_field_value(struct event_format *event, const char *name, void *data); +void *raw_field_ptr(struct event_format *event, const char *name, void *data); + +void parse_proc_kallsyms(struct pevent *pevent, char *file, unsigned int size); +void parse_ftrace_printk(struct pevent *pevent, char *file, unsigned int size); + +ssize_t trace_report(int fd, struct pevent **pevent, bool repipe); + +int trace_parse_common_type(struct pevent *pevent, void *data); +int trace_parse_common_pid(struct pevent *pevent, void *data); + +struct event_format *trace_find_next_event(struct pevent *pevent, + struct event_format *event); +unsigned long long read_size(struct event_format *event, void *ptr, int size); +unsigned long long eval_flag(const char *flag); + +struct pevent_record *trace_read_data(struct pevent *pevent, int cpu); +int read_tracing_data(int fd, struct list_head *pattrs); + +struct tracing_data { + /* size is only valid if temp is 'true' */ + ssize_t size; + bool temp; + char temp_file[50]; +}; + +struct tracing_data *tracing_data_get(struct list_head *pattrs, + int fd, bool temp); +void tracing_data_put(struct tracing_data *tdata); + + +struct addr_location; + +struct perf_session; + +struct scripting_ops { + const char *name; + int (*start_script) (const char *script, int argc, const char **argv); + int (*stop_script) (void); + void (*process_event) (union perf_event *event, + struct perf_sample *sample, + struct perf_evsel *evsel, + struct machine *machine, + struct addr_location *al); + int (*generate_script) (struct pevent *pevent, const char *outfile); +}; + +int script_spec_register(const char *spec, struct scripting_ops *ops); + +void setup_perl_scripting(void); +void setup_python_scripting(void); + +struct scripting_context { + struct pevent *pevent; + void *event_data; +}; + +int common_pc(struct scripting_context *context); +int common_flags(struct scripting_context *context); +int common_lock_depth(struct scripting_context *context); + +#endif /* _PERF_UTIL_TRACE_EVENT_H */ diff --git a/tools/perf/util/types.h b/tools/perf/util/types.h new file mode 100644 index 00000000..c51fa6b7 --- /dev/null +++ b/tools/perf/util/types.h @@ -0,0 +1,24 @@ +#ifndef __PERF_TYPES_H +#define __PERF_TYPES_H + +#include <stdint.h> + +/* + * We define u64 as uint64_t for every architecture + * so that we can print it with "%"PRIx64 without getting warnings. + */ +typedef uint64_t u64; +typedef int64_t s64; +typedef unsigned int u32; +typedef signed int s32; +typedef unsigned short u16; +typedef signed short s16; +typedef unsigned char u8; +typedef signed char s8; + +union u64_swap { + u64 val64; + u32 val32[2]; +}; + +#endif /* __PERF_TYPES_H */ diff --git a/tools/perf/util/unwind.c b/tools/perf/util/unwind.c new file mode 100644 index 00000000..958723ba --- /dev/null +++ b/tools/perf/util/unwind.c @@ -0,0 +1,571 @@ +/* + * Post mortem Dwarf CFI based unwinding on top of regs and stack dumps. + * + * Lots of this code have been borrowed or heavily inspired from parts of + * the libunwind 0.99 code which are (amongst other contributors I may have + * forgotten): + * + * Copyright (C) 2002-2007 Hewlett-Packard Co + * Contributed by David Mosberger-Tang <davidm@hpl.hp.com> + * + * And the bugs have been added by: + * + * Copyright (C) 2010, Frederic Weisbecker <fweisbec@gmail.com> + * Copyright (C) 2012, Jiri Olsa <jolsa@redhat.com> + * + */ + +#include <elf.h> +#include <gelf.h> +#include <fcntl.h> +#include <string.h> +#include <unistd.h> +#include <sys/mman.h> +#include <linux/list.h> +#include <libunwind.h> +#include <libunwind-ptrace.h> +#include "thread.h" +#include "session.h" +#include "perf_regs.h" +#include "unwind.h" +#include "util.h" + +extern int +UNW_OBJ(dwarf_search_unwind_table) (unw_addr_space_t as, + unw_word_t ip, + unw_dyn_info_t *di, + unw_proc_info_t *pi, + int need_unwind_info, void *arg); + +#define dwarf_search_unwind_table UNW_OBJ(dwarf_search_unwind_table) + +#define DW_EH_PE_FORMAT_MASK 0x0f /* format of the encoded value */ +#define DW_EH_PE_APPL_MASK 0x70 /* how the value is to be applied */ + +/* Pointer-encoding formats: */ +#define DW_EH_PE_omit 0xff +#define DW_EH_PE_ptr 0x00 /* pointer-sized unsigned value */ +#define DW_EH_PE_udata4 0x03 /* unsigned 32-bit value */ +#define DW_EH_PE_udata8 0x04 /* unsigned 64-bit value */ +#define DW_EH_PE_sdata4 0x0b /* signed 32-bit value */ +#define DW_EH_PE_sdata8 0x0c /* signed 64-bit value */ + +/* Pointer-encoding application: */ +#define DW_EH_PE_absptr 0x00 /* absolute value */ +#define DW_EH_PE_pcrel 0x10 /* rel. to addr. of encoded value */ + +/* + * The following are not documented by LSB v1.3, yet they are used by + * GCC, presumably they aren't documented by LSB since they aren't + * used on Linux: + */ +#define DW_EH_PE_funcrel 0x40 /* start-of-procedure-relative */ +#define DW_EH_PE_aligned 0x50 /* aligned pointer */ + +/* Flags intentionaly not handled, since they're not needed: + * #define DW_EH_PE_indirect 0x80 + * #define DW_EH_PE_uleb128 0x01 + * #define DW_EH_PE_udata2 0x02 + * #define DW_EH_PE_sleb128 0x09 + * #define DW_EH_PE_sdata2 0x0a + * #define DW_EH_PE_textrel 0x20 + * #define DW_EH_PE_datarel 0x30 + */ + +struct unwind_info { + struct perf_sample *sample; + struct machine *machine; + struct thread *thread; + u64 sample_uregs; +}; + +#define dw_read(ptr, type, end) ({ \ + type *__p = (type *) ptr; \ + type __v; \ + if ((__p + 1) > (type *) end) \ + return -EINVAL; \ + __v = *__p++; \ + ptr = (typeof(ptr)) __p; \ + __v; \ + }) + +static int __dw_read_encoded_value(u8 **p, u8 *end, u64 *val, + u8 encoding) +{ + u8 *cur = *p; + *val = 0; + + switch (encoding) { + case DW_EH_PE_omit: + *val = 0; + goto out; + case DW_EH_PE_ptr: + *val = dw_read(cur, unsigned long, end); + goto out; + default: + break; + } + + switch (encoding & DW_EH_PE_APPL_MASK) { + case DW_EH_PE_absptr: + break; + case DW_EH_PE_pcrel: + *val = (unsigned long) cur; + break; + default: + return -EINVAL; + } + + if ((encoding & 0x07) == 0x00) + encoding |= DW_EH_PE_udata4; + + switch (encoding & DW_EH_PE_FORMAT_MASK) { + case DW_EH_PE_sdata4: + *val += dw_read(cur, s32, end); + break; + case DW_EH_PE_udata4: + *val += dw_read(cur, u32, end); + break; + case DW_EH_PE_sdata8: + *val += dw_read(cur, s64, end); + break; + case DW_EH_PE_udata8: + *val += dw_read(cur, u64, end); + break; + default: + return -EINVAL; + } + + out: + *p = cur; + return 0; +} + +#define dw_read_encoded_value(ptr, end, enc) ({ \ + u64 __v; \ + if (__dw_read_encoded_value(&ptr, end, &__v, enc)) { \ + return -EINVAL; \ + } \ + __v; \ + }) + +static Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep, + GElf_Shdr *shp, const char *name) +{ + Elf_Scn *sec = NULL; + + 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)) + break; + } + + return sec; +} + +static u64 elf_section_offset(int fd, const char *name) +{ + Elf *elf; + GElf_Ehdr ehdr; + GElf_Shdr shdr; + u64 offset = 0; + + elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL); + if (elf == NULL) + return 0; + + do { + if (gelf_getehdr(elf, &ehdr) == NULL) + break; + + if (!elf_section_by_name(elf, &ehdr, &shdr, name)) + break; + + offset = shdr.sh_offset; + } while (0); + + elf_end(elf); + return offset; +} + +struct table_entry { + u32 start_ip_offset; + u32 fde_offset; +}; + +struct eh_frame_hdr { + unsigned char version; + unsigned char eh_frame_ptr_enc; + unsigned char fde_count_enc; + unsigned char table_enc; + + /* + * The rest of the header is variable-length and consists of the + * following members: + * + * encoded_t eh_frame_ptr; + * encoded_t fde_count; + */ + + /* A single encoded pointer should not be more than 8 bytes. */ + u64 enc[2]; + + /* + * struct { + * encoded_t start_ip; + * encoded_t fde_addr; + * } binary_search_table[fde_count]; + */ + char data[0]; +} __packed; + +static int unwind_spec_ehframe(struct dso *dso, struct machine *machine, + u64 offset, u64 *table_data, u64 *segbase, + u64 *fde_count) +{ + struct eh_frame_hdr hdr; + u8 *enc = (u8 *) &hdr.enc; + u8 *end = (u8 *) &hdr.data; + ssize_t r; + + r = dso__data_read_offset(dso, machine, offset, + (u8 *) &hdr, sizeof(hdr)); + if (r != sizeof(hdr)) + return -EINVAL; + + /* We dont need eh_frame_ptr, just skip it. */ + dw_read_encoded_value(enc, end, hdr.eh_frame_ptr_enc); + + *fde_count = dw_read_encoded_value(enc, end, hdr.fde_count_enc); + *segbase = offset; + *table_data = (enc - (u8 *) &hdr) + offset; + return 0; +} + +static int read_unwind_spec(struct dso *dso, struct machine *machine, + u64 *table_data, u64 *segbase, u64 *fde_count) +{ + int ret = -EINVAL, fd; + u64 offset; + + fd = dso__data_fd(dso, machine); + if (fd < 0) + return -EINVAL; + + offset = elf_section_offset(fd, ".eh_frame_hdr"); + close(fd); + + if (offset) + ret = unwind_spec_ehframe(dso, machine, offset, + table_data, segbase, + fde_count); + + /* TODO .debug_frame check if eh_frame_hdr fails */ + return ret; +} + +static struct map *find_map(unw_word_t ip, struct unwind_info *ui) +{ + struct addr_location al; + + thread__find_addr_map(ui->thread, ui->machine, PERF_RECORD_MISC_USER, + MAP__FUNCTION, ip, &al); + return al.map; +} + +static int +find_proc_info(unw_addr_space_t as, unw_word_t ip, unw_proc_info_t *pi, + int need_unwind_info, void *arg) +{ + struct unwind_info *ui = arg; + struct map *map; + unw_dyn_info_t di; + u64 table_data, segbase, fde_count; + + map = find_map(ip, ui); + if (!map || !map->dso) + return -EINVAL; + + pr_debug("unwind: find_proc_info dso %s\n", map->dso->name); + + if (read_unwind_spec(map->dso, ui->machine, + &table_data, &segbase, &fde_count)) + return -EINVAL; + + memset(&di, 0, sizeof(di)); + di.format = UNW_INFO_FORMAT_REMOTE_TABLE; + di.start_ip = map->start; + di.end_ip = map->end; + di.u.rti.segbase = map->start + segbase; + di.u.rti.table_data = map->start + table_data; + di.u.rti.table_len = fde_count * sizeof(struct table_entry) + / sizeof(unw_word_t); + return dwarf_search_unwind_table(as, ip, &di, pi, + need_unwind_info, arg); +} + +static int access_fpreg(unw_addr_space_t __maybe_unused as, + unw_regnum_t __maybe_unused num, + unw_fpreg_t __maybe_unused *val, + int __maybe_unused __write, + void __maybe_unused *arg) +{ + pr_err("unwind: access_fpreg unsupported\n"); + return -UNW_EINVAL; +} + +static int get_dyn_info_list_addr(unw_addr_space_t __maybe_unused as, + unw_word_t __maybe_unused *dil_addr, + void __maybe_unused *arg) +{ + return -UNW_ENOINFO; +} + +static int resume(unw_addr_space_t __maybe_unused as, + unw_cursor_t __maybe_unused *cu, + void __maybe_unused *arg) +{ + pr_err("unwind: resume unsupported\n"); + return -UNW_EINVAL; +} + +static int +get_proc_name(unw_addr_space_t __maybe_unused as, + unw_word_t __maybe_unused addr, + char __maybe_unused *bufp, size_t __maybe_unused buf_len, + unw_word_t __maybe_unused *offp, void __maybe_unused *arg) +{ + pr_err("unwind: get_proc_name unsupported\n"); + return -UNW_EINVAL; +} + +static int access_dso_mem(struct unwind_info *ui, unw_word_t addr, + unw_word_t *data) +{ + struct addr_location al; + ssize_t size; + + thread__find_addr_map(ui->thread, ui->machine, PERF_RECORD_MISC_USER, + MAP__FUNCTION, addr, &al); + if (!al.map) { + pr_debug("unwind: no map for %lx\n", (unsigned long)addr); + return -1; + } + + if (!al.map->dso) + return -1; + + size = dso__data_read_addr(al.map->dso, al.map, ui->machine, + addr, (u8 *) data, sizeof(*data)); + + return !(size == sizeof(*data)); +} + +static int reg_value(unw_word_t *valp, struct regs_dump *regs, int id, + u64 sample_regs) +{ + int i, idx = 0; + + if (!(sample_regs & (1 << id))) + return -EINVAL; + + for (i = 0; i < id; i++) { + if (sample_regs & (1 << i)) + idx++; + } + + *valp = regs->regs[idx]; + return 0; +} + +static int access_mem(unw_addr_space_t __maybe_unused as, + unw_word_t addr, unw_word_t *valp, + int __write, void *arg) +{ + struct unwind_info *ui = arg; + struct stack_dump *stack = &ui->sample->user_stack; + unw_word_t start, end; + int offset; + int ret; + + /* Don't support write, probably not needed. */ + if (__write || !stack || !ui->sample->user_regs.regs) { + *valp = 0; + return 0; + } + + ret = reg_value(&start, &ui->sample->user_regs, PERF_REG_SP, + ui->sample_uregs); + if (ret) + return ret; + + end = start + stack->size; + + /* Check overflow. */ + if (addr + sizeof(unw_word_t) < addr) + return -EINVAL; + + if (addr < start || addr + sizeof(unw_word_t) >= end) { + ret = access_dso_mem(ui, addr, valp); + if (ret) { + pr_debug("unwind: access_mem %p not inside range %p-%p\n", + (void *)addr, (void *)start, (void *)end); + *valp = 0; + return ret; + } + return 0; + } + + offset = addr - start; + *valp = *(unw_word_t *)&stack->data[offset]; + pr_debug("unwind: access_mem addr %p, val %lx, offset %d\n", + (void *)addr, (unsigned long)*valp, offset); + return 0; +} + +static int access_reg(unw_addr_space_t __maybe_unused as, + unw_regnum_t regnum, unw_word_t *valp, + int __write, void *arg) +{ + struct unwind_info *ui = arg; + int id, ret; + + /* Don't support write, I suspect we don't need it. */ + if (__write) { + pr_err("unwind: access_reg w %d\n", regnum); + return 0; + } + + if (!ui->sample->user_regs.regs) { + *valp = 0; + return 0; + } + + id = unwind__arch_reg_id(regnum); + if (id < 0) + return -EINVAL; + + ret = reg_value(valp, &ui->sample->user_regs, id, ui->sample_uregs); + if (ret) { + pr_err("unwind: can't read reg %d\n", regnum); + return ret; + } + + pr_debug("unwind: reg %d, val %lx\n", regnum, (unsigned long)*valp); + return 0; +} + +static void put_unwind_info(unw_addr_space_t __maybe_unused as, + unw_proc_info_t *pi __maybe_unused, + void *arg __maybe_unused) +{ + pr_debug("unwind: put_unwind_info called\n"); +} + +static int entry(u64 ip, struct thread *thread, struct machine *machine, + unwind_entry_cb_t cb, void *arg) +{ + struct unwind_entry e; + struct addr_location al; + + thread__find_addr_location(thread, machine, + PERF_RECORD_MISC_USER, + MAP__FUNCTION, ip, &al, NULL); + + e.ip = ip; + e.map = al.map; + e.sym = al.sym; + + pr_debug("unwind: %s:ip = 0x%" PRIx64 " (0x%" PRIx64 ")\n", + al.sym ? al.sym->name : "''", + ip, + al.map ? al.map->map_ip(al.map, ip) : (u64) 0); + + return cb(&e, arg); +} + +static void display_error(int err) +{ + switch (err) { + case UNW_EINVAL: + pr_err("unwind: Only supports local.\n"); + break; + case UNW_EUNSPEC: + pr_err("unwind: Unspecified error.\n"); + break; + case UNW_EBADREG: + pr_err("unwind: Register unavailable.\n"); + break; + default: + break; + } +} + +static unw_accessors_t accessors = { + .find_proc_info = find_proc_info, + .put_unwind_info = put_unwind_info, + .get_dyn_info_list_addr = get_dyn_info_list_addr, + .access_mem = access_mem, + .access_reg = access_reg, + .access_fpreg = access_fpreg, + .resume = resume, + .get_proc_name = get_proc_name, +}; + +static int get_entries(struct unwind_info *ui, unwind_entry_cb_t cb, + void *arg) +{ + unw_addr_space_t addr_space; + unw_cursor_t c; + int ret; + + addr_space = unw_create_addr_space(&accessors, 0); + if (!addr_space) { + pr_err("unwind: Can't create unwind address space.\n"); + return -ENOMEM; + } + + ret = unw_init_remote(&c, addr_space, ui); + if (ret) + display_error(ret); + + while (!ret && (unw_step(&c) > 0)) { + unw_word_t ip; + + unw_get_reg(&c, UNW_REG_IP, &ip); + ret = entry(ip, ui->thread, ui->machine, cb, arg); + } + + unw_destroy_addr_space(addr_space); + return ret; +} + +int unwind__get_entries(unwind_entry_cb_t cb, void *arg, + struct machine *machine, struct thread *thread, + u64 sample_uregs, struct perf_sample *data) +{ + unw_word_t ip; + struct unwind_info ui = { + .sample = data, + .sample_uregs = sample_uregs, + .thread = thread, + .machine = machine, + }; + int ret; + + if (!data->user_regs.regs) + return -EINVAL; + + ret = reg_value(&ip, &data->user_regs, PERF_REG_IP, sample_uregs); + if (ret) + return ret; + + ret = entry(ip, thread, machine, cb, arg); + if (ret) + return -ENOMEM; + + return get_entries(&ui, cb, arg); +} diff --git a/tools/perf/util/unwind.h b/tools/perf/util/unwind.h new file mode 100644 index 00000000..cb6bc503 --- /dev/null +++ b/tools/perf/util/unwind.h @@ -0,0 +1,35 @@ +#ifndef __UNWIND_H +#define __UNWIND_H + +#include "types.h" +#include "event.h" +#include "symbol.h" + +struct unwind_entry { + struct map *map; + struct symbol *sym; + u64 ip; +}; + +typedef int (*unwind_entry_cb_t)(struct unwind_entry *entry, void *arg); + +#ifdef LIBUNWIND_SUPPORT +int unwind__get_entries(unwind_entry_cb_t cb, void *arg, + struct machine *machine, + struct thread *thread, + u64 sample_uregs, + struct perf_sample *data); +int unwind__arch_reg_id(int regnum); +#else +static inline int +unwind__get_entries(unwind_entry_cb_t cb __maybe_unused, + void *arg __maybe_unused, + struct machine *machine __maybe_unused, + struct thread *thread __maybe_unused, + u64 sample_uregs __maybe_unused, + struct perf_sample *data __maybe_unused) +{ + return 0; +} +#endif /* LIBUNWIND_SUPPORT */ +#endif /* __UNWIND_H */ diff --git a/tools/perf/util/usage.c b/tools/perf/util/usage.c new file mode 100644 index 00000000..4007aca8 --- /dev/null +++ b/tools/perf/util/usage.c @@ -0,0 +1,84 @@ +/* + * usage.c + * + * Various reporting routines. + * Originally copied from GIT source. + * + * Copyright (C) Linus Torvalds, 2005 + */ +#include "util.h" +#include "debug.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.c b/tools/perf/util/util.c new file mode 100644 index 00000000..805d1f52 --- /dev/null +++ b/tools/perf/util/util.c @@ -0,0 +1,244 @@ +#include "../perf.h" +#include "util.h" +#include <sys/mman.h> +#ifdef BACKTRACE_SUPPORT +#include <execinfo.h> +#endif +#include <stdio.h> +#include <stdlib.h> + +/* + * XXX We need to find a better place for these things... + */ +unsigned int page_size; + +bool test_attr__enabled; + +bool perf_host = true; +bool perf_guest = false; + +void event_attr_init(struct perf_event_attr *attr) +{ + if (!perf_host) + attr->exclude_host = 1; + if (!perf_guest) + attr->exclude_guest = 1; + /* to capture ABI version */ + attr->size = sizeof(*attr); +} + +int mkdir_p(char *path, mode_t mode) +{ + struct stat st; + int err; + char *d = path; + + if (*d != '/') + return -1; + + if (stat(path, &st) == 0) + return 0; + + while (*++d == '/'); + + while ((d = strchr(d, '/'))) { + *d = '\0'; + err = stat(path, &st) && mkdir(path, mode); + *d++ = '/'; + if (err) + return -1; + while (*d == '/') + ++d; + } + return (stat(path, &st) && mkdir(path, mode)) ? -1 : 0; +} + +static int slow_copyfile(const char *from, const char *to) +{ + int err = 0; + char *line = NULL; + size_t n; + FILE *from_fp = fopen(from, "r"), *to_fp; + + if (from_fp == NULL) + goto out; + + to_fp = fopen(to, "w"); + if (to_fp == NULL) + goto out_fclose_from; + + while (getline(&line, &n, from_fp) > 0) + if (fputs(line, to_fp) == EOF) + goto out_fclose_to; + err = 0; +out_fclose_to: + fclose(to_fp); + free(line); +out_fclose_from: + fclose(from_fp); +out: + return err; +} + +int copyfile(const char *from, const char *to) +{ + int fromfd, tofd; + struct stat st; + void *addr; + int err = -1; + + if (stat(from, &st)) + goto out; + + if (st.st_size == 0) /* /proc? do it slowly... */ + return slow_copyfile(from, to); + + fromfd = open(from, O_RDONLY); + if (fromfd < 0) + goto out; + + tofd = creat(to, 0755); + if (tofd < 0) + goto out_close_from; + + addr = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fromfd, 0); + if (addr == MAP_FAILED) + goto out_close_to; + + if (write(tofd, addr, st.st_size) == st.st_size) + err = 0; + + munmap(addr, st.st_size); +out_close_to: + close(tofd); + if (err) + unlink(to); +out_close_from: + close(fromfd); +out: + return err; +} + +unsigned long convert_unit(unsigned long value, char *unit) +{ + *unit = ' '; + + if (value > 1000) { + value /= 1000; + *unit = 'K'; + } + + if (value > 1000) { + value /= 1000; + *unit = 'M'; + } + + if (value > 1000) { + value /= 1000; + *unit = 'G'; + } + + return value; +} + +int readn(int fd, void *buf, size_t n) +{ + void *buf_start = buf; + + while (n) { + int ret = read(fd, buf, n); + + if (ret <= 0) + return ret; + + n -= ret; + buf += ret; + } + + return buf - buf_start; +} + +size_t hex_width(u64 v) +{ + size_t n = 1; + + while ((v >>= 4)) + ++n; + + return n; +} + +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; +} + +/* Obtain a backtrace and print it to stdout. */ +#ifdef BACKTRACE_SUPPORT +void dump_stack(void) +{ + void *array[16]; + size_t size = backtrace(array, ARRAY_SIZE(array)); + char **strings = backtrace_symbols(array, size); + size_t i; + + printf("Obtained %zd stack frames.\n", size); + + for (i = 0; i < size; i++) + printf("%s\n", strings[i]); + + free(strings); +} +#else +void dump_stack(void) {} +#endif + +void get_term_dimensions(struct winsize *ws) +{ + char *s = getenv("LINES"); + + if (s != NULL) { + ws->ws_row = atoi(s); + s = getenv("COLUMNS"); + if (s != NULL) { + ws->ws_col = atoi(s); + if (ws->ws_row && ws->ws_col) + return; + } + } +#ifdef TIOCGWINSZ + if (ioctl(1, TIOCGWINSZ, ws) == 0 && + ws->ws_row && ws->ws_col) + return; +#endif + ws->ws_row = 25; + ws->ws_col = 80; +} diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h new file mode 100644 index 00000000..92f4e91a --- /dev/null +++ b/tools/perf/util/util.h @@ -0,0 +1,280 @@ +#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) + +#define _ALL_SOURCE 1 +#define _BSD_SOURCE 1 +#define HAS_BOOL + +#include <unistd.h> +#include <stdio.h> +#include <sys/stat.h> +#include <sys/statfs.h> +#include <fcntl.h> +#include <stdbool.h> +#include <stddef.h> +#include <stdlib.h> +#include <stdarg.h> +#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> +#include <sys/wait.h> +#include <sys/poll.h> +#include <sys/socket.h> +#include <sys/ioctl.h> +#include <inttypes.h> +#include <linux/magic.h> +#include "types.h" +#include <sys/ttydefaults.h> + +extern const char *graph_line; +extern const char *graph_dotted_line; +extern char buildid_dir[]; + +/* On most systems <limits.h> would have given us this, but + * not on some systems (e.g. GNU/Hurd). + */ +#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))); + +#include "../../../include/linux/stringify.h" + +#define DIE_IF(cnd) \ + do { if (cnd) \ + die(" at (" __FILE__ ":" __stringify(__LINE__) "): " \ + __stringify(cnd) "\n"); \ + } while (0) + + +extern void set_die_routine(void (*routine)(const char *err, va_list params) NORETURN); + +extern int prefixcmp(const char *str, const char *prefix); +extern void set_buildid_dir(void); +extern void disable_buildid_cache(void); + +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; +} + +#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 *xrealloc(void *ptr, size_t size) __attribute__((weak)); + + +static inline void *zalloc(size_t size) +{ + return calloc(1, size); +} + +static inline int has_extension(const char *filename, const char *ext) +{ + size_t len = strlen(filename); + size_t extlen = strlen(ext); + + return len > extlen && !memcmp(filename + len - extlen, ext, extlen); +} + +/* Sane ctype - no locale, and works with signed chars */ +#undef isascii +#undef isspace +#undef isdigit +#undef isxdigit +#undef isalpha +#undef isprint +#undef isalnum +#undef islower +#undef isupper +#undef tolower +#undef toupper + +#ifndef NSEC_PER_MSEC +#define NSEC_PER_MSEC 1000000L +#endif + +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 GIT_PRINT_EXTRA 0x20 +#define GIT_PRINT 0x3E +#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 isxdigit(x) \ + (sane_istest(toupper(x), GIT_ALPHA | GIT_DIGIT) && toupper(x) < 'G') +#define isalpha(x) sane_istest(x,GIT_ALPHA) +#define isalnum(x) sane_istest(x,GIT_ALPHA | GIT_DIGIT) +#define isprint(x) sane_istest(x,GIT_PRINT) +#define islower(x) (sane_istest(x,GIT_ALPHA) && sane_istest(x,0x20)) +#define isupper(x) (sane_istest(x,GIT_ALPHA) && !sane_istest(x,0x20)) +#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; +} + +int mkdir_p(char *path, mode_t mode); +int copyfile(const char *from, const char *to); + +s64 perf_atoll(const char *str); +char **argv_split(const char *str, int *argcp); +void argv_free(char **argv); +bool strglobmatch(const char *str, const char *pat); +bool strlazymatch(const char *str, const char *pat); +int strtailcmp(const char *s1, const char *s2); +char *strxfrchar(char *s, char from, char to); +unsigned long convert_unit(unsigned long value, char *unit); +int readn(int fd, void *buf, size_t size); + +struct perf_event_attr; + +void event_attr_init(struct perf_event_attr *attr); + +#define _STR(x) #x +#define STR(x) _STR(x) + +/* + * Determine whether some value is a power of two, where zero is + * *not* considered a power of two. + */ + +#ifndef _LINUX_LOG2_H /* is_power_of_2 is defined here too, and Bionic drags in that header */ +static inline __attribute__((const)) +bool is_power_of_2(unsigned long n) +{ + return (n != 0 && ((n & (n - 1)) == 0)); +} +#endif + +size_t hex_width(u64 v); +int hex2u64(const char *ptr, u64 *val); + +char *ltrim(char *s); +char *rtrim(char *s); + +void dump_stack(void); + +extern unsigned int page_size; + +struct winsize; +void get_term_dimensions(struct winsize *ws); + +#endif diff --git a/tools/perf/util/values.c b/tools/perf/util/values.c new file mode 100644 index 00000000..697c8b4e --- /dev/null +++ b/tools/perf/util/values.c @@ -0,0 +1,232 @@ +#include <stdlib.h> + +#include "util.h" +#include "values.h" + +void perf_read_values_init(struct perf_read_values *values) +{ + values->threads_max = 16; + values->pid = malloc(values->threads_max * sizeof(*values->pid)); + values->tid = malloc(values->threads_max * sizeof(*values->tid)); + values->value = malloc(values->threads_max * sizeof(*values->value)); + if (!values->pid || !values->tid || !values->value) + die("failed to allocate read_values threads arrays"); + values->threads = 0; + + values->counters_max = 16; + values->counterrawid = malloc(values->counters_max + * sizeof(*values->counterrawid)); + values->countername = malloc(values->counters_max + * sizeof(*values->countername)); + if (!values->counterrawid || !values->countername) + die("failed to allocate read_values counters arrays"); + values->counters = 0; +} + +void perf_read_values_destroy(struct perf_read_values *values) +{ + int i; + + if (!values->threads_max || !values->counters_max) + return; + + for (i = 0; i < values->threads; i++) + free(values->value[i]); + free(values->value); + free(values->pid); + free(values->tid); + free(values->counterrawid); + for (i = 0; i < values->counters; i++) + free(values->countername[i]); + free(values->countername); +} + +static void perf_read_values__enlarge_threads(struct perf_read_values *values) +{ + values->threads_max *= 2; + values->pid = realloc(values->pid, + values->threads_max * sizeof(*values->pid)); + values->tid = realloc(values->tid, + values->threads_max * sizeof(*values->tid)); + values->value = realloc(values->value, + values->threads_max * sizeof(*values->value)); + if (!values->pid || !values->tid || !values->value) + die("failed to enlarge read_values threads arrays"); +} + +static int perf_read_values__findnew_thread(struct perf_read_values *values, + u32 pid, u32 tid) +{ + int i; + + for (i = 0; i < values->threads; i++) + if (values->pid[i] == pid && values->tid[i] == tid) + return i; + + if (values->threads == values->threads_max) + perf_read_values__enlarge_threads(values); + + i = values->threads++; + values->pid[i] = pid; + values->tid[i] = tid; + values->value[i] = malloc(values->counters_max * sizeof(**values->value)); + if (!values->value[i]) + die("failed to allocate read_values counters array"); + + return i; +} + +static void perf_read_values__enlarge_counters(struct perf_read_values *values) +{ + int i; + + values->counters_max *= 2; + values->counterrawid = realloc(values->counterrawid, + values->counters_max * sizeof(*values->counterrawid)); + values->countername = realloc(values->countername, + values->counters_max * sizeof(*values->countername)); + if (!values->counterrawid || !values->countername) + die("failed to enlarge read_values counters arrays"); + + for (i = 0; i < values->threads; i++) { + values->value[i] = realloc(values->value[i], + values->counters_max * sizeof(**values->value)); + if (!values->value[i]) + die("failed to enlarge read_values counters arrays"); + } +} + +static int perf_read_values__findnew_counter(struct perf_read_values *values, + u64 rawid, const char *name) +{ + int i; + + for (i = 0; i < values->counters; i++) + if (values->counterrawid[i] == rawid) + return i; + + if (values->counters == values->counters_max) + perf_read_values__enlarge_counters(values); + + i = values->counters++; + values->counterrawid[i] = rawid; + values->countername[i] = strdup(name); + + return i; +} + +void perf_read_values_add_value(struct perf_read_values *values, + u32 pid, u32 tid, + u64 rawid, const char *name, u64 value) +{ + int tindex, cindex; + + tindex = perf_read_values__findnew_thread(values, pid, tid); + cindex = perf_read_values__findnew_counter(values, rawid, name); + + values->value[tindex][cindex] = value; +} + +static void perf_read_values__display_pretty(FILE *fp, + struct perf_read_values *values) +{ + int i, j; + int pidwidth, tidwidth; + int *counterwidth; + + counterwidth = malloc(values->counters * sizeof(*counterwidth)); + if (!counterwidth) + die("failed to allocate counterwidth array"); + tidwidth = 3; + pidwidth = 3; + for (j = 0; j < values->counters; j++) + counterwidth[j] = strlen(values->countername[j]); + for (i = 0; i < values->threads; i++) { + int width; + + width = snprintf(NULL, 0, "%d", values->pid[i]); + if (width > pidwidth) + pidwidth = width; + width = snprintf(NULL, 0, "%d", values->tid[i]); + if (width > tidwidth) + tidwidth = width; + for (j = 0; j < values->counters; j++) { + width = snprintf(NULL, 0, "%" PRIu64, values->value[i][j]); + if (width > counterwidth[j]) + counterwidth[j] = width; + } + } + + fprintf(fp, "# %*s %*s", pidwidth, "PID", tidwidth, "TID"); + for (j = 0; j < values->counters; j++) + fprintf(fp, " %*s", counterwidth[j], values->countername[j]); + fprintf(fp, "\n"); + + for (i = 0; i < values->threads; i++) { + fprintf(fp, " %*d %*d", pidwidth, values->pid[i], + tidwidth, values->tid[i]); + for (j = 0; j < values->counters; j++) + fprintf(fp, " %*" PRIu64, + counterwidth[j], values->value[i][j]); + fprintf(fp, "\n"); + } + free(counterwidth); +} + +static void perf_read_values__display_raw(FILE *fp, + struct perf_read_values *values) +{ + int width, pidwidth, tidwidth, namewidth, rawwidth, countwidth; + int i, j; + + tidwidth = 3; /* TID */ + pidwidth = 3; /* PID */ + namewidth = 4; /* "Name" */ + rawwidth = 3; /* "Raw" */ + countwidth = 5; /* "Count" */ + + for (i = 0; i < values->threads; i++) { + width = snprintf(NULL, 0, "%d", values->pid[i]); + if (width > pidwidth) + pidwidth = width; + width = snprintf(NULL, 0, "%d", values->tid[i]); + if (width > tidwidth) + tidwidth = width; + } + for (j = 0; j < values->counters; j++) { + width = strlen(values->countername[j]); + if (width > namewidth) + namewidth = width; + width = snprintf(NULL, 0, "%" PRIx64, values->counterrawid[j]); + if (width > rawwidth) + rawwidth = width; + } + for (i = 0; i < values->threads; i++) { + for (j = 0; j < values->counters; j++) { + width = snprintf(NULL, 0, "%" PRIu64, values->value[i][j]); + if (width > countwidth) + countwidth = width; + } + } + + fprintf(fp, "# %*s %*s %*s %*s %*s\n", + pidwidth, "PID", tidwidth, "TID", + namewidth, "Name", rawwidth, "Raw", + countwidth, "Count"); + for (i = 0; i < values->threads; i++) + for (j = 0; j < values->counters; j++) + fprintf(fp, " %*d %*d %*s %*" PRIx64 " %*" PRIu64, + pidwidth, values->pid[i], + tidwidth, values->tid[i], + namewidth, values->countername[j], + rawwidth, values->counterrawid[j], + countwidth, values->value[i][j]); +} + +void perf_read_values_display(FILE *fp, struct perf_read_values *values, int raw) +{ + if (raw) + perf_read_values__display_raw(fp, values); + else + perf_read_values__display_pretty(fp, values); +} diff --git a/tools/perf/util/values.h b/tools/perf/util/values.h new file mode 100644 index 00000000..2fa967e1 --- /dev/null +++ b/tools/perf/util/values.h @@ -0,0 +1,27 @@ +#ifndef __PERF_VALUES_H +#define __PERF_VALUES_H + +#include "types.h" + +struct perf_read_values { + int threads; + int threads_max; + u32 *pid, *tid; + int counters; + int counters_max; + u64 *counterrawid; + char **countername; + u64 **value; +}; + +void perf_read_values_init(struct perf_read_values *values); +void perf_read_values_destroy(struct perf_read_values *values); + +void perf_read_values_add_value(struct perf_read_values *values, + u32 pid, u32 tid, + u64 rawid, const char *name, u64 value); + +void perf_read_values_display(FILE *fp, struct perf_read_values *values, + int raw); + +#endif /* __PERF_VALUES_H */ diff --git a/tools/perf/util/vdso.c b/tools/perf/util/vdso.c new file mode 100644 index 00000000..e60951fc --- /dev/null +++ b/tools/perf/util/vdso.c @@ -0,0 +1,111 @@ + +#include <unistd.h> +#include <stdio.h> +#include <string.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <stdlib.h> +#include <linux/kernel.h> + +#include "vdso.h" +#include "util.h" +#include "symbol.h" +#include "linux/string.h" + +static bool vdso_found; +static char vdso_file[] = "/tmp/perf-vdso.so-XXXXXX"; + +static int find_vdso_map(void **start, void **end) +{ + FILE *maps; + char line[128]; + int found = 0; + + maps = fopen("/proc/self/maps", "r"); + if (!maps) { + pr_err("vdso: cannot open maps\n"); + return -1; + } + + while (!found && fgets(line, sizeof(line), maps)) { + int m = -1; + + /* We care only about private r-x mappings. */ + if (2 != sscanf(line, "%p-%p r-xp %*x %*x:%*x %*u %n", + start, end, &m)) + continue; + if (m < 0) + continue; + + if (!strncmp(&line[m], VDSO__MAP_NAME, + sizeof(VDSO__MAP_NAME) - 1)) + found = 1; + } + + fclose(maps); + return !found; +} + +static char *get_file(void) +{ + char *vdso = NULL; + char *buf = NULL; + void *start, *end; + size_t size; + int fd; + + if (vdso_found) + return vdso_file; + + if (find_vdso_map(&start, &end)) + return NULL; + + size = end - start; + + buf = memdup(start, size); + if (!buf) + return NULL; + + fd = mkstemp(vdso_file); + if (fd < 0) + goto out; + + if (size == (size_t) write(fd, buf, size)) + vdso = vdso_file; + + close(fd); + + out: + free(buf); + + vdso_found = (vdso != NULL); + return vdso; +} + +void vdso__exit(void) +{ + if (vdso_found) + unlink(vdso_file); +} + +struct dso *vdso__dso_findnew(struct list_head *head) +{ + struct dso *dso = dsos__find(head, VDSO__MAP_NAME); + + if (!dso) { + char *file; + + file = get_file(); + if (!file) + return NULL; + + dso = dso__new(VDSO__MAP_NAME); + if (dso != NULL) { + dsos__add(head, dso); + dso__set_long_name(dso, file); + } + } + + return dso; +} diff --git a/tools/perf/util/vdso.h b/tools/perf/util/vdso.h new file mode 100644 index 00000000..0f76e7ca --- /dev/null +++ b/tools/perf/util/vdso.h @@ -0,0 +1,18 @@ +#ifndef __PERF_VDSO__ +#define __PERF_VDSO__ + +#include <linux/types.h> +#include <string.h> +#include <stdbool.h> + +#define VDSO__MAP_NAME "[vdso]" + +static inline bool is_vdso_map(const char *filename) +{ + return !strcmp(filename, VDSO__MAP_NAME); +} + +struct dso *vdso__dso_findnew(struct list_head *head); +void vdso__exit(void); + +#endif /* __PERF_VDSO__ */ diff --git a/tools/perf/util/wrapper.c b/tools/perf/util/wrapper.c new file mode 100644 index 00000000..19f15b65 --- /dev/null +++ b/tools/perf/util/wrapper.c @@ -0,0 +1,41 @@ +/* + * 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 __maybe_unused, + int flag __maybe_unused) +{ +} + +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 *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; +} diff --git a/tools/perf/util/xyarray.c b/tools/perf/util/xyarray.c new file mode 100644 index 00000000..22afbf6c --- /dev/null +++ b/tools/perf/util/xyarray.c @@ -0,0 +1,20 @@ +#include "xyarray.h" +#include "util.h" + +struct xyarray *xyarray__new(int xlen, int ylen, size_t entry_size) +{ + size_t row_size = ylen * entry_size; + struct xyarray *xy = zalloc(sizeof(*xy) + xlen * row_size); + + if (xy != NULL) { + xy->entry_size = entry_size; + xy->row_size = row_size; + } + + return xy; +} + +void xyarray__delete(struct xyarray *xy) +{ + free(xy); +} diff --git a/tools/perf/util/xyarray.h b/tools/perf/util/xyarray.h new file mode 100644 index 00000000..c488a072 --- /dev/null +++ b/tools/perf/util/xyarray.h @@ -0,0 +1,20 @@ +#ifndef _PERF_XYARRAY_H_ +#define _PERF_XYARRAY_H_ 1 + +#include <sys/types.h> + +struct xyarray { + size_t row_size; + size_t entry_size; + char contents[]; +}; + +struct xyarray *xyarray__new(int xlen, int ylen, size_t entry_size); +void xyarray__delete(struct xyarray *xy); + +static inline void *xyarray__entry(struct xyarray *xy, int x, int y) +{ + return &xy->contents[x * xy->row_size + y * xy->entry_size]; +} + +#endif /* _PERF_XYARRAY_H_ */ |