[PATCH 00/34] perf/core improvements and fixes

[PATCH 00/34] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>

Hi Ingo,

	I ended up deciding sending the whole series, repeating the first
dozen, with the "perf evsel: Skip ignored symbols while printing callchain"
one fixed up wrt Adrian's report.

	Please consider applying,

Best Regards,

- Arnaldo

The following changes since commit 65661f96d3b32f4b28fef26d21be81d7e173b965:

  perf/x86: Add RAPL hrtimer support (2013-11-27 15:31:23 +0100)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux tags/perf-core-for-mingo

for you to fetch changes up to aa880acd0853437933cefcb7d745b03fde3e6a3d:

  perf tools unwinding: Use the per-feature check flags (2013-11-27 16:50:28 -0300)

----------------------------------------------------------------
perf/core improvements and fixes:

. Make per-cpu mmaps the default in 'perf record', from Adrian Hunter.

. Default -t/--thread 'perf record' option to no inheritance,
  from Adrian Hunter.

. Make 'perf top -g' refer to callchains, for consistency with other tools,
  from David Ahern.

. Skip ignored symbols while printing callchain, from David Ahern.

. Print callchains and symbols if they exist in 'perf script',
  from David Ahern.

. Remove thread summary coloring in 'perf trace', from Pekka Enberg.

. zsh completion support, from Ramkumar Ramachandra.

. 'perf timechart' improvements, including backtrace support,
  from Stanislav Fomichev.

. Fix using kcore files stored in the buildid cache when doing report/annotate
  in non-live sessions, from Adrian Hunter

. Minor 'timechart' cleanups.

. Fix libunwind build suport on ARM, from Jean Pihet.

. Fix tags/TAGS targets rebuilding, from Jiri Olsa.

. Add options to show comm, fork, exit and mmap PERF_RECORD_ events in
  'perf script', from Namhyung Kim.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

----------------------------------------------------------------
Adrian Hunter (5):
      perf record: Make per-cpu mmaps the default.
      perf tools: Allow '--inherit' as the negation of '--no-inherit'
      perf tools: Add option macro OPT_BOOLEAN_SET
      perf record: Default -t option to no inheritance
      perf symbols: Fix not finding kcore in buildid cache

Arnaldo Carvalho de Melo (2):
      perf timechart: Remove some needless struct forward declarations
      perf timechart: Remove misplaced __maybe_unused

David Ahern (6):
      perf top: Make -g refer to callchains
      perf evsel: Skip ignored symbols while printing callchain
      perf symbols: Move idle syms check from top to generic function
      perf thread: Move comm_list check into function
      perf tools: Export setup_list
      perf script: Print callchains and symbols if they exist

Jean Pihet (2):
      perf tools: Add per-feature check flags
      perf tools unwinding: Use the per-feature check flags

Jiri Olsa (1):
      perf tools: Fix tags/TAGS targets rebuilding

Namhyung Kim (3):
      perf script: Move evname print code to process_event()
      perf script: Print comm, fork and exit events also
      perf script: Print mmap[2] events also

Pekka Enberg (1):
      perf trace: Remove thread summary coloring

Ramkumar Ramachandra (5):
      perf completion: Introduce a layer of indirection
      perf completion: Factor out compgen stuff
      perf completion: Factor out call to __ltrim_colon_completions
      perf completion: Introduce zsh support
      perf completion: Rename file to reflect zsh support

Stanislav Fomichev (8):
      perf timechart: Always try to print at least 15 tasks
      perf timechart: Add option to limit number of tasks
      perf timechart: Use proc_num to implement --power-only
      perf timechart: Add support for displaying only tasks related data
      perf timechart: Group figures and add title with details
      perf timechart: Add support for -P and -T in timechart recording
      perf timechart: Add backtrace support
      perf timechart: dynamically determine event fields offset

Steven Rostedt (1):
      tools lib traceevent: Use helper trace-seq in print functions like kernel does

 tools/lib/traceevent/event-parse.c                 |   7 +-
 tools/perf/Documentation/perf-record.txt           |  12 +-
 tools/perf/Documentation/perf-script.txt           |   6 +
 tools/perf/Documentation/perf-timechart.txt        |  26 +-
 tools/perf/Documentation/perf-top.txt              |   5 +-
 tools/perf/Makefile                                |   7 +-
 tools/perf/Makefile.perf                           |   4 +-
 tools/perf/builtin-record.c                        |  13 +-
 tools/perf/builtin-script.c                        | 209 ++++++++++-
 tools/perf/builtin-timechart.c                     | 418 ++++++++++++++-------
 tools/perf/builtin-top.c                           |  29 +-
 tools/perf/builtin-trace.c                         |  13 +-
 tools/perf/config/Makefile                         |  38 +-
 tools/perf/config/feature-checks/Makefile          |   6 +-
 tools/perf/{bash_completion => perf-completion.sh} | 100 ++++-
 tools/perf/perf.h                                  |   1 +
 tools/perf/tests/attr/test-record-no-inherit       |   2 +-
 tools/perf/util/event.c                            |   3 +-
 tools/perf/util/evlist.c                           |   6 +-
 tools/perf/util/evsel.c                            |   5 +-
 tools/perf/util/parse-options.c                    |  21 ++
 tools/perf/util/parse-options.h                    |   8 +
 tools/perf/util/session.c                          |  10 +-
 tools/perf/util/svghelper.c                        |  77 +++-
 tools/perf/util/svghelper.h                        |  11 +-
 tools/perf/util/symbol.c                           |  41 +-
 tools/perf/util/symbol.h                           |   4 +
 tools/perf/util/target.c                           |  11 +-
 tools/perf/util/target.h                           |   4 +-
 tools/perf/util/thread.h                           |  12 +
 30 files changed, 855 insertions(+), 254 deletions(-)
 rename tools/perf/{bash_completion => perf-completion.sh} (65%)

[PATCH 01/34] tools lib traceevent: Use helper trace-seq in print functions like kernel does

[Arnaldo Carvalho de Melo]

From: Steven Rostedt <rostedt@goodmis.org>

Jiri Olsa reported that his plugin for scsi was chopping off part of the
output. Investigating this, I found that Jiri used the same functions as
what is in the kernel, which adds the following:

	trace_seq_putc(p, 0);

This adds a '\0' to the output string. The reason this works in the
kernel is that the "p" that is passed to the function helper is a
temporary trace_seq. But in the libtraceevent library, it's the pointer
to the trace_seq used to output. By adding the '\0', it truncates the
line and nothing added after that will be printed.

We can solve this in two ways. One is to have the helper functions for
the library not add the unnecessary '\0'. The other is to change the
library to also use a helper trace_seq structure that gets copied to the
main trace_seq just like the kernel does.

The latter allows the helper functions in the plugins to be the same as
the kernel, which is the better solution.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Reported-by: Jiri Olsa <jolsa@redhat.com>
Tested-by: Jiri Olsa <jolsa@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/20131119182937.401668e3@gandalf.local.home
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/lib/traceevent/event-parse.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/tools/lib/traceevent/event-parse.c b/tools/lib/traceevent/event-parse.c
index 217c82ee3665..900fca01bdd3 100644
--- a/tools/lib/traceevent/event-parse.c
+++ b/tools/lib/traceevent/event-parse.c
@@ -4099,6 +4099,7 @@ static void pretty_print(struct trace_seq *s, void *data, int size, struct event
 	unsigned long long val;
 	struct func_map *func;
 	const char *saveptr;
+	struct trace_seq p;
 	char *bprint_fmt = NULL;
 	char format[32];
 	int show_func;
@@ -4306,8 +4307,12 @@ static void pretty_print(struct trace_seq *s, void *data, int size, struct event
 				format[len] = 0;
 				if (!len_as_arg)
 					len_arg = -1;
-				print_str_arg(s, data, size, event,
+				/* Use helper trace_seq */
+				trace_seq_init(&p);
+				print_str_arg(&p, data, size, event,
 					      format, len_arg, arg);
+				trace_seq_terminate(&p);
+				trace_seq_puts(s, p.buffer);
 				arg = arg->next;
 				break;
 			default:
-- 
1.8.1.4

[PATCH 02/34] perf trace: Remove thread summary coloring

[Arnaldo Carvalho de Melo]

From: Pekka Enberg <penberg@kernel.org>

Thread summary line coloring looks ugly.  It doesn't add much value so
remove coloring completely.

Signed-off-by: Pekka Enberg <penberg@kernel.org>
Acked-by: David Ahern <dsahern@gmail.com>
Link: http://lkml.kernel.org/r/1384447410-1771-1-git-send-email-penberg@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 8be17fc462ba..e9f345e2551a 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -2158,7 +2158,6 @@ static int trace__fprintf_one_thread(struct thread *thread, void *priv)
 	size_t printed = data->printed;
 	struct trace *trace = data->trace;
 	struct thread_trace *ttrace = thread->priv;
-	const char *color;
 	double ratio;
 
 	if (ttrace == NULL)
@@ -2166,17 +2165,9 @@ static int trace__fprintf_one_thread(struct thread *thread, void *priv)
 
 	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, " %s (%d), ", thread__comm_str(thread), thread->tid);
+	printed += fprintf(fp, " %s (%d), ", thread__comm_str(thread), thread->tid);
 	printed += fprintf(fp, "%lu events, ", ttrace->nr_events);
-	printed += color_fprintf(fp, color, "%.1f%%", ratio);
+	printed += fprintf(fp, "%.1f%%", ratio);
 	printed += fprintf(fp, ", %.3f msec\n", ttrace->runtime_ms);
 	printed += thread__dump_stats(ttrace, trace, fp);
 
-- 
1.8.1.4

[PATCH 03/34] perf top: Make -g refer to callchains

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

In most commands -g is used for callchains. Make perf-top follow suit.
Move group to just --group with no short cut making it similar to
perf-record.

Signed-off-by: David Ahern <dsahern@gmail.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1384487490-6865-1-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-top.txt | 5 ++---
 tools/perf/builtin-top.c              | 4 ++--
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt
index 7de01dd79688..cdd8d4946dba 100644
--- a/tools/perf/Documentation/perf-top.txt
+++ b/tools/perf/Documentation/perf-top.txt
@@ -50,7 +50,6 @@ Default is to monitor all CPUS.
 --count-filter=<count>::
 	Only display functions with more events than this.
 
--g::
 --group::
         Put the counters into a counter group.
 
@@ -143,12 +142,12 @@ Default is to monitor all CPUS.
 --asm-raw::
 	Show raw instruction encoding of assembly instructions.
 
--G::
+-g::
 	Enables call-graph (stack chain/backtrace) recording.
 
 --call-graph::
 	Setup and enable call-graph (stack chain/backtrace) recording,
-	implies -G.
+	implies -g.
 
 --max-stack::
 	Set the stack depth limit when parsing the callchain, anything
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 71e6402729a8..531522d3d97b 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -1084,7 +1084,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 			    "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,
+	OPT_BOOLEAN(0, "group", &opts->group,
 			    "put the counters into a counter group"),
 	OPT_BOOLEAN('i', "no-inherit", &opts->no_inherit,
 		    "child tasks do not inherit counters"),
@@ -1105,7 +1105,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 		   " abort, in_tx, transaction"),
 	OPT_BOOLEAN('n', "show-nr-samples", &symbol_conf.show_nr_samples,
 		    "Show a column with the number of samples"),
-	OPT_CALLBACK_NOOPT('G', NULL, &top.record_opts,
+	OPT_CALLBACK_NOOPT('g', NULL, &top.record_opts,
 			   NULL, "enables call-graph recording",
 			   &callchain_opt),
 	OPT_CALLBACK(0, "call-graph", &top.record_opts,
-- 
1.8.1.4

[PATCH 04/34] perf completion: Introduce a layer of indirection

[Arnaldo Carvalho de Melo]

From: Ramkumar Ramachandra <artagnon@gmail.com>

Define the variables cur, words, cword, and prev outside the main
completion function so that we have a chance to override it when we
introduce zsh support.

Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Link: http://lkml.kernel.org/r/1384704807-15779-2-git-send-email-artagnon@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/bash_completion | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/tools/perf/bash_completion b/tools/perf/bash_completion
index 62e157db2e2b..3efdc84f112c 100644
--- a/tools/perf/bash_completion
+++ b/tools/perf/bash_completion
@@ -89,15 +89,12 @@ __ltrim_colon_completions()
 	fi
 }
 
-type perf &>/dev/null &&
-_perf()
+__perf_main ()
 {
-	local cur words cword prev cmd
-
-	COMPREPLY=()
-	_get_comp_words_by_ref -n =: cur words cword prev
+	local cmd
 
 	cmd=${words[0]}
+	COMPREPLY=()
 
 	# List perf subcommands or long options
 	if [ $cword -eq 1 ]; then
@@ -120,6 +117,14 @@ _perf()
 		opts=$($cmd $subcmd --list-opts)
 		COMPREPLY=( $( compgen -W '$opts' -- "$cur" ) )
 	fi
+}
+
+type perf &>/dev/null &&
+_perf()
+{
+	local cur words cword prev
+	_get_comp_words_by_ref -n =: cur words cword prev
+	__perf_main
 } &&
 
 complete -o bashdefault -o default -o nospace -F _perf perf 2>/dev/null \
-- 
1.8.1.4

[PATCH 05/34] perf completion: Factor out compgen stuff

[Arnaldo Carvalho de Melo]

From: Ramkumar Ramachandra <artagnon@gmail.com>

compgen is a bash-builtin; factor out the invocations into a separate
function to give us a chance to override it with a zsh equivalent in
future patches.

Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Link: http://lkml.kernel.org/r/1384704807-15779-3-git-send-email-artagnon@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/bash_completion | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/tools/perf/bash_completion b/tools/perf/bash_completion
index 3efdc84f112c..824312681601 100644
--- a/tools/perf/bash_completion
+++ b/tools/perf/bash_completion
@@ -89,6 +89,11 @@ __ltrim_colon_completions()
 	fi
 }
 
+__perfcomp ()
+{
+	COMPREPLY=( $( compgen -W "$1" -- "$2" ) )
+}
+
 __perf_main ()
 {
 	local cmd
@@ -99,23 +104,23 @@ __perf_main ()
 	# List perf subcommands or long options
 	if [ $cword -eq 1 ]; then
 		if [[ $cur == --* ]]; then
-			COMPREPLY=( $( compgen -W '--help --version \
+			__perfcomp '--help --version \
 			--exec-path --html-path --paginate --no-pager \
-			--perf-dir --work-tree --debugfs-dir' -- "$cur" ) )
+			--perf-dir --work-tree --debugfs-dir' -- "$cur"
 		else
 			cmds=$($cmd --list-cmds)
-			COMPREPLY=( $( compgen -W '$cmds' -- "$cur" ) )
+			__perfcomp "$cmds" "$cur"
 		fi
 	# List possible events for -e option
 	elif [[ $prev == "-e" && "${words[1]}" == @(record|stat|top) ]]; then
 		evts=$($cmd list --raw-dump)
-		COMPREPLY=( $( compgen -W '$evts' -- "$cur" ) )
+		__perfcomp "$evts" "$cur"
 		__ltrim_colon_completions $cur
 	# List long option names
 	elif [[ $cur == --* ]];  then
 		subcmd=${words[1]}
 		opts=$($cmd $subcmd --list-opts)
-		COMPREPLY=( $( compgen -W '$opts' -- "$cur" ) )
+		__perfcomp "$opts" "$cur"
 	fi
 }
 
-- 
1.8.1.4

[PATCH 06/34] perf completion: Factor out call to __ltrim_colon_completions

[Arnaldo Carvalho de Melo]

From: Ramkumar Ramachandra <artagnon@gmail.com>

In our sole callsite, __ltrim_colon_completions is called after
__perfcomp, to modify the COMPREPLY set by the invocation.

This is problematic, because in the zsh equivalent (using compset/
compadd), we'll have to generate completions in one-shot.

So factor out this entire callsite into a special override'able
__perfcomp_colon function; we will override it when introducing zsh
support.

Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Link: http://lkml.kernel.org/r/1384704807-15779-4-git-send-email-artagnon@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/bash_completion | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/tools/perf/bash_completion b/tools/perf/bash_completion
index 824312681601..573599b42be2 100644
--- a/tools/perf/bash_completion
+++ b/tools/perf/bash_completion
@@ -94,6 +94,12 @@ __perfcomp ()
 	COMPREPLY=( $( compgen -W "$1" -- "$2" ) )
 }
 
+__perfcomp_colon ()
+{
+	__perfcomp "$1" "$2"
+	__ltrim_colon_completions $cur
+}
+
 __perf_main ()
 {
 	local cmd
@@ -114,8 +120,7 @@ __perf_main ()
 	# List possible events for -e option
 	elif [[ $prev == "-e" && "${words[1]}" == @(record|stat|top) ]]; then
 		evts=$($cmd list --raw-dump)
-		__perfcomp "$evts" "$cur"
-		__ltrim_colon_completions $cur
+		__perfcomp_colon "$evts" "$cur"
 	# List long option names
 	elif [[ $cur == --* ]];  then
 		subcmd=${words[1]}
-- 
1.8.1.4

[PATCH 07/34] perf completion: Introduce zsh support

[Arnaldo Carvalho de Melo]

From: Ramkumar Ramachandra <artagnon@gmail.com>

__perfcomp(), __perfcomp_colon(), and _perf() have to be overridden.
Inspired by the way the git.git completion system is structured.

Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1384704807-15779-5-git-send-email-artagnon@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/bash_completion | 63 +++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 62 insertions(+), 1 deletion(-)

diff --git a/tools/perf/bash_completion b/tools/perf/bash_completion
index 573599b42be2..49494882d9bb 100644
--- a/tools/perf/bash_completion
+++ b/tools/perf/bash_completion
@@ -1,4 +1,4 @@
-# perf completion
+# perf bash and zsh completion
 
 # Taken from git.git's completion script.
 __my_reassemble_comp_words_by_ref()
@@ -129,6 +129,67 @@ __perf_main ()
 	fi
 }
 
+if [[ -n ${ZSH_VERSION-} ]]; then
+	autoload -U +X compinit && compinit
+
+	__perfcomp ()
+	{
+		emulate -L zsh
+
+		local c IFS=$' \t\n'
+		local -a array
+
+		for c in ${=1}; do
+			case $c in
+			--*=*|*.) ;;
+			*) c="$c " ;;
+			esac
+			array[${#array[@]}+1]="$c"
+		done
+
+		compset -P '*[=:]'
+		compadd -Q -S '' -a -- array && _ret=0
+	}
+
+	__perfcomp_colon ()
+	{
+		emulate -L zsh
+
+		local cur_="${2-$cur}"
+		local c IFS=$' \t\n'
+		local -a array
+
+		if [[ "$cur_" == *:* ]]; then
+			local colon_word=${cur_%"${cur_##*:}"}
+		fi
+
+		for c in ${=1}; do
+			case $c in
+			--*=*|*.) ;;
+			*) c="$c " ;;
+			esac
+			array[$#array+1]=${c#"$colon_word"}
+		done
+
+		compset -P '*[=:]'
+		compadd -Q -S '' -a -- array && _ret=0
+	}
+
+	_perf ()
+	{
+		local _ret=1 cur cword prev
+		cur=${words[CURRENT]}
+		prev=${words[CURRENT-1]}
+		let cword=CURRENT-1
+		emulate ksh -c __perf_main
+		let _ret && _default && _ret=0
+		return _ret
+	}
+
+	compdef _perf perf
+	return
+fi
+
 type perf &>/dev/null &&
 _perf()
 {
-- 
1.8.1.4

[PATCH 08/34] perf completion: Rename file to reflect zsh support

[Arnaldo Carvalho de Melo]

From: Ramkumar Ramachandra <artagnon@gmail.com>

Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Link: http://lkml.kernel.org/r/1384704807-15779-6-git-send-email-artagnon@gmail.com
[ Fix 'make install' target ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Makefile.perf      |   4 +-
 tools/perf/bash_completion    | 202 ------------------------------------------
 tools/perf/perf-completion.sh | 202 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 204 insertions(+), 204 deletions(-)
 delete mode 100644 tools/perf/bash_completion
 create mode 100644 tools/perf/perf-completion.sh

diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index 7fc8f179cae7..e416ccc7d831 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -840,9 +840,9 @@ ifndef NO_LIBPYTHON
 		$(INSTALL) scripts/python/*.py -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python'; \
 		$(INSTALL) scripts/python/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/bin'
 endif
-	$(call QUIET_INSTALL, bash_completion-script) \
+	$(call QUIET_INSTALL, perf_completion-script) \
 		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(sysconfdir_SQ)/bash_completion.d'; \
-		$(INSTALL) bash_completion '$(DESTDIR_SQ)$(sysconfdir_SQ)/bash_completion.d/perf'
+		$(INSTALL) perf-completion.sh '$(DESTDIR_SQ)$(sysconfdir_SQ)/bash_completion.d/perf'
 	$(call QUIET_INSTALL, tests) \
 		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests'; \
 		$(INSTALL) tests/attr.py '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests'; \
diff --git a/tools/perf/bash_completion b/tools/perf/bash_completion
deleted file mode 100644
index 49494882d9bb..000000000000
--- a/tools/perf/bash_completion
+++ /dev/null
@@ -1,202 +0,0 @@
-# perf bash and zsh completion
-
-# Taken from git.git's completion script.
-__my_reassemble_comp_words_by_ref()
-{
-	local exclude i j first
-	# Which word separators to exclude?
-	exclude="${1//[^$COMP_WORDBREAKS]}"
-	cword_=$COMP_CWORD
-	if [ -z "$exclude" ]; then
-		words_=("${COMP_WORDS[@]}")
-		return
-	fi
-	# List of word completion separators has shrunk;
-	# re-assemble words to complete.
-	for ((i=0, j=0; i < ${#COMP_WORDS[@]}; i++, j++)); do
-		# Append each nonempty word consisting of just
-		# word separator characters to the current word.
-		first=t
-		while
-			[ $i -gt 0 ] &&
-			[ -n "${COMP_WORDS[$i]}" ] &&
-			# word consists of excluded word separators
-			[ "${COMP_WORDS[$i]//[^$exclude]}" = "${COMP_WORDS[$i]}" ]
-		do
-			# Attach to the previous token,
-			# unless the previous token is the command name.
-			if [ $j -ge 2 ] && [ -n "$first" ]; then
-				((j--))
-			fi
-			first=
-			words_[$j]=${words_[j]}${COMP_WORDS[i]}
-			if [ $i = $COMP_CWORD ]; then
-				cword_=$j
-			fi
-			if (($i < ${#COMP_WORDS[@]} - 1)); then
-				((i++))
-			else
-				# Done.
-				return
-			fi
-		done
-		words_[$j]=${words_[j]}${COMP_WORDS[i]}
-		if [ $i = $COMP_CWORD ]; then
-			cword_=$j
-		fi
-	done
-}
-
-type _get_comp_words_by_ref &>/dev/null ||
-_get_comp_words_by_ref()
-{
-	local exclude cur_ words_ cword_
-	if [ "$1" = "-n" ]; then
-		exclude=$2
-		shift 2
-	fi
-	__my_reassemble_comp_words_by_ref "$exclude"
-	cur_=${words_[cword_]}
-	while [ $# -gt 0 ]; do
-		case "$1" in
-		cur)
-			cur=$cur_
-			;;
-		prev)
-			prev=${words_[$cword_-1]}
-			;;
-		words)
-			words=("${words_[@]}")
-			;;
-		cword)
-			cword=$cword_
-			;;
-		esac
-		shift
-	done
-}
-
-type __ltrim_colon_completions &>/dev/null ||
-__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
-}
-
-__perfcomp ()
-{
-	COMPREPLY=( $( compgen -W "$1" -- "$2" ) )
-}
-
-__perfcomp_colon ()
-{
-	__perfcomp "$1" "$2"
-	__ltrim_colon_completions $cur
-}
-
-__perf_main ()
-{
-	local cmd
-
-	cmd=${words[0]}
-	COMPREPLY=()
-
-	# List perf subcommands or long options
-	if [ $cword -eq 1 ]; then
-		if [[ $cur == --* ]]; then
-			__perfcomp '--help --version \
-			--exec-path --html-path --paginate --no-pager \
-			--perf-dir --work-tree --debugfs-dir' -- "$cur"
-		else
-			cmds=$($cmd --list-cmds)
-			__perfcomp "$cmds" "$cur"
-		fi
-	# List possible events for -e option
-	elif [[ $prev == "-e" && "${words[1]}" == @(record|stat|top) ]]; then
-		evts=$($cmd list --raw-dump)
-		__perfcomp_colon "$evts" "$cur"
-	# List long option names
-	elif [[ $cur == --* ]];  then
-		subcmd=${words[1]}
-		opts=$($cmd $subcmd --list-opts)
-		__perfcomp "$opts" "$cur"
-	fi
-}
-
-if [[ -n ${ZSH_VERSION-} ]]; then
-	autoload -U +X compinit && compinit
-
-	__perfcomp ()
-	{
-		emulate -L zsh
-
-		local c IFS=$' \t\n'
-		local -a array
-
-		for c in ${=1}; do
-			case $c in
-			--*=*|*.) ;;
-			*) c="$c " ;;
-			esac
-			array[${#array[@]}+1]="$c"
-		done
-
-		compset -P '*[=:]'
-		compadd -Q -S '' -a -- array && _ret=0
-	}
-
-	__perfcomp_colon ()
-	{
-		emulate -L zsh
-
-		local cur_="${2-$cur}"
-		local c IFS=$' \t\n'
-		local -a array
-
-		if [[ "$cur_" == *:* ]]; then
-			local colon_word=${cur_%"${cur_##*:}"}
-		fi
-
-		for c in ${=1}; do
-			case $c in
-			--*=*|*.) ;;
-			*) c="$c " ;;
-			esac
-			array[$#array+1]=${c#"$colon_word"}
-		done
-
-		compset -P '*[=:]'
-		compadd -Q -S '' -a -- array && _ret=0
-	}
-
-	_perf ()
-	{
-		local _ret=1 cur cword prev
-		cur=${words[CURRENT]}
-		prev=${words[CURRENT-1]}
-		let cword=CURRENT-1
-		emulate ksh -c __perf_main
-		let _ret && _default && _ret=0
-		return _ret
-	}
-
-	compdef _perf perf
-	return
-fi
-
-type perf &>/dev/null &&
-_perf()
-{
-	local cur words cword prev
-	_get_comp_words_by_ref -n =: cur words cword prev
-	__perf_main
-} &&
-
-complete -o bashdefault -o default -o nospace -F _perf perf 2>/dev/null \
-	|| complete -o default -o nospace -F _perf perf
diff --git a/tools/perf/perf-completion.sh b/tools/perf/perf-completion.sh
new file mode 100644
index 000000000000..49494882d9bb
--- /dev/null
+++ b/tools/perf/perf-completion.sh
@@ -0,0 +1,202 @@
+# perf bash and zsh completion
+
+# Taken from git.git's completion script.
+__my_reassemble_comp_words_by_ref()
+{
+	local exclude i j first
+	# Which word separators to exclude?
+	exclude="${1//[^$COMP_WORDBREAKS]}"
+	cword_=$COMP_CWORD
+	if [ -z "$exclude" ]; then
+		words_=("${COMP_WORDS[@]}")
+		return
+	fi
+	# List of word completion separators has shrunk;
+	# re-assemble words to complete.
+	for ((i=0, j=0; i < ${#COMP_WORDS[@]}; i++, j++)); do
+		# Append each nonempty word consisting of just
+		# word separator characters to the current word.
+		first=t
+		while
+			[ $i -gt 0 ] &&
+			[ -n "${COMP_WORDS[$i]}" ] &&
+			# word consists of excluded word separators
+			[ "${COMP_WORDS[$i]//[^$exclude]}" = "${COMP_WORDS[$i]}" ]
+		do
+			# Attach to the previous token,
+			# unless the previous token is the command name.
+			if [ $j -ge 2 ] && [ -n "$first" ]; then
+				((j--))
+			fi
+			first=
+			words_[$j]=${words_[j]}${COMP_WORDS[i]}
+			if [ $i = $COMP_CWORD ]; then
+				cword_=$j
+			fi
+			if (($i < ${#COMP_WORDS[@]} - 1)); then
+				((i++))
+			else
+				# Done.
+				return
+			fi
+		done
+		words_[$j]=${words_[j]}${COMP_WORDS[i]}
+		if [ $i = $COMP_CWORD ]; then
+			cword_=$j
+		fi
+	done
+}
+
+type _get_comp_words_by_ref &>/dev/null ||
+_get_comp_words_by_ref()
+{
+	local exclude cur_ words_ cword_
+	if [ "$1" = "-n" ]; then
+		exclude=$2
+		shift 2
+	fi
+	__my_reassemble_comp_words_by_ref "$exclude"
+	cur_=${words_[cword_]}
+	while [ $# -gt 0 ]; do
+		case "$1" in
+		cur)
+			cur=$cur_
+			;;
+		prev)
+			prev=${words_[$cword_-1]}
+			;;
+		words)
+			words=("${words_[@]}")
+			;;
+		cword)
+			cword=$cword_
+			;;
+		esac
+		shift
+	done
+}
+
+type __ltrim_colon_completions &>/dev/null ||
+__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
+}
+
+__perfcomp ()
+{
+	COMPREPLY=( $( compgen -W "$1" -- "$2" ) )
+}
+
+__perfcomp_colon ()
+{
+	__perfcomp "$1" "$2"
+	__ltrim_colon_completions $cur
+}
+
+__perf_main ()
+{
+	local cmd
+
+	cmd=${words[0]}
+	COMPREPLY=()
+
+	# List perf subcommands or long options
+	if [ $cword -eq 1 ]; then
+		if [[ $cur == --* ]]; then
+			__perfcomp '--help --version \
+			--exec-path --html-path --paginate --no-pager \
+			--perf-dir --work-tree --debugfs-dir' -- "$cur"
+		else
+			cmds=$($cmd --list-cmds)
+			__perfcomp "$cmds" "$cur"
+		fi
+	# List possible events for -e option
+	elif [[ $prev == "-e" && "${words[1]}" == @(record|stat|top) ]]; then
+		evts=$($cmd list --raw-dump)
+		__perfcomp_colon "$evts" "$cur"
+	# List long option names
+	elif [[ $cur == --* ]];  then
+		subcmd=${words[1]}
+		opts=$($cmd $subcmd --list-opts)
+		__perfcomp "$opts" "$cur"
+	fi
+}
+
+if [[ -n ${ZSH_VERSION-} ]]; then
+	autoload -U +X compinit && compinit
+
+	__perfcomp ()
+	{
+		emulate -L zsh
+
+		local c IFS=$' \t\n'
+		local -a array
+
+		for c in ${=1}; do
+			case $c in
+			--*=*|*.) ;;
+			*) c="$c " ;;
+			esac
+			array[${#array[@]}+1]="$c"
+		done
+
+		compset -P '*[=:]'
+		compadd -Q -S '' -a -- array && _ret=0
+	}
+
+	__perfcomp_colon ()
+	{
+		emulate -L zsh
+
+		local cur_="${2-$cur}"
+		local c IFS=$' \t\n'
+		local -a array
+
+		if [[ "$cur_" == *:* ]]; then
+			local colon_word=${cur_%"${cur_##*:}"}
+		fi
+
+		for c in ${=1}; do
+			case $c in
+			--*=*|*.) ;;
+			*) c="$c " ;;
+			esac
+			array[$#array+1]=${c#"$colon_word"}
+		done
+
+		compset -P '*[=:]'
+		compadd -Q -S '' -a -- array && _ret=0
+	}
+
+	_perf ()
+	{
+		local _ret=1 cur cword prev
+		cur=${words[CURRENT]}
+		prev=${words[CURRENT-1]}
+		let cword=CURRENT-1
+		emulate ksh -c __perf_main
+		let _ret && _default && _ret=0
+		return _ret
+	}
+
+	compdef _perf perf
+	return
+fi
+
+type perf &>/dev/null &&
+_perf()
+{
+	local cur words cword prev
+	_get_comp_words_by_ref -n =: cur words cword prev
+	__perf_main
+} &&
+
+complete -o bashdefault -o default -o nospace -F _perf perf 2>/dev/null \
+	|| complete -o default -o nospace -F _perf perf
-- 
1.8.1.4

[PATCH 09/34] perf script: Move evname print code to process_event()

[Arnaldo Carvalho de Melo]

From: Namhyung Kim <namhyung@kernel.org>

The print_sample_start() will be reused by other printing routine for
internal events like COMM, FORK and EXIT from next patch.  And because
they're not tied to a specific event, move the evname print code to its
caller.

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Reviewed-by: David Ahern <dsahern@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1384752894-10974-1-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-script.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index baf17989a216..b392770766dd 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -288,7 +288,6 @@ static void print_sample_start(struct perf_sample *sample,
 			       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;
@@ -323,11 +322,6 @@ static void print_sample_start(struct perf_sample *sample,
 		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)
@@ -434,6 +428,11 @@ static void process_event(union perf_event *event, struct perf_sample *sample,
 
 	print_sample_start(sample, thread, evsel);
 
+	if (PRINT_FIELD(EVNAME)) {
+		const char *evname = perf_evsel__name(evsel);
+		printf("%s: ", evname ? evname : "[unknown]");
+	}
+
 	if (is_bts_event(attr)) {
 		print_sample_bts(event, sample, evsel, machine, thread);
 		return;
-- 
1.8.1.4

[PATCH 10/34] perf record: Make per-cpu mmaps the default.

[Arnaldo Carvalho de Melo]

From: Adrian Hunter <adrian.hunter@intel.com>

This affects the -p, -t and -u options that previously defaulted to
per-thread mmaps.

Consequently add an option to select per-thread mmaps to support the old
behaviour.

Note that per-thread can be used with a workload-only (i.e. none of -p,
-t, -u, -a or -C is selected) to get a per-thread mmap with no
inheritance.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/5286271D.3020808@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-record.txt     | 10 +++++-----
 tools/perf/builtin-record.c                  |  5 +++--
 tools/perf/tests/attr/test-record-no-inherit |  2 +-
 tools/perf/util/evlist.c                     |  6 ++++--
 tools/perf/util/evsel.c                      |  5 +++--
 tools/perf/util/target.c                     | 11 ++++++++++-
 tools/perf/util/target.h                     |  4 +++-
 7 files changed, 29 insertions(+), 14 deletions(-)

diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index 43b42c4f4a91..6ac867e7667f 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -201,11 +201,11 @@ abort events and some memory events in precise mode on modern Intel CPUs.
 --transaction::
 Record transaction flags for transaction related events.
 
---force-per-cpu::
-Force the use of per-cpu mmaps.  By default, when tasks are specified (i.e. -p,
--t or -u options) per-thread mmaps are created.  This option overrides that and
-forces per-cpu mmaps.  A side-effect of that is that inheritance is
-automatically enabled.  Add the -i option also to disable inheritance.
+--per-thread::
+Use per-thread mmaps.  By default per-cpu mmaps are created.  This option
+overrides that and uses per-thread mmaps.  A side-effect of that is that
+inheritance is automatically disabled.  --per-thread is ignored with a warning
+if combined with -a or -C options.
 
 SEE ALSO
 --------
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 7c8020a32784..f5b18b8fe8c9 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -800,6 +800,7 @@ static struct perf_record record = {
 		.freq		     = 4000,
 		.target		     = {
 			.uses_mmap   = true,
+			.default_per_cpu = true,
 		},
 	},
 };
@@ -888,8 +889,8 @@ const struct option record_options[] = {
 		    "sample by weight (on special events only)"),
 	OPT_BOOLEAN(0, "transaction", &record.opts.sample_transaction,
 		    "sample transaction flags (special events only)"),
-	OPT_BOOLEAN(0, "force-per-cpu", &record.opts.target.force_per_cpu,
-		    "force the use of per-cpu mmaps"),
+	OPT_BOOLEAN(0, "per-thread", &record.opts.target.per_thread,
+		    "use per-thread mmaps"),
 	OPT_END()
 };
 
diff --git a/tools/perf/tests/attr/test-record-no-inherit b/tools/perf/tests/attr/test-record-no-inherit
index 9079a25cd643..44edcb2edcd5 100644
--- a/tools/perf/tests/attr/test-record-no-inherit
+++ b/tools/perf/tests/attr/test-record-no-inherit
@@ -3,5 +3,5 @@ command = record
 args    = -i kill >/dev/null 2>&1
 
 [event:base-record]
-sample_type=259
+sample_type=263
 inherit=0
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index bbc746aa5716..76fa76431329 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -819,8 +819,10 @@ int perf_evlist__create_maps(struct perf_evlist *evlist, struct target *target)
 	if (evlist->threads == NULL)
 		return -1;
 
-	if (target->force_per_cpu)
-		evlist->cpus = cpu_map__new(target->cpu_list);
+	if (target->default_per_cpu)
+		evlist->cpus = target->per_thread ?
+					cpu_map__dummy_new() :
+					cpu_map__new(target->cpu_list);
 	else if (target__has_task(target))
 		evlist->cpus = cpu_map__dummy_new();
 	else if (!target__has_cpu(target) && !target->uses_mmap)
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index dad64926170f..b5fe7f9b2e15 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -574,6 +574,7 @@ void perf_evsel__config(struct perf_evsel *evsel,
 	struct perf_evsel *leader = evsel->leader;
 	struct perf_event_attr *attr = &evsel->attr;
 	int track = !evsel->idx; /* only the first counter needs these */
+	bool per_cpu = opts->target.default_per_cpu && !opts->target.per_thread;
 
 	attr->sample_id_all = perf_missing_features.sample_id_all ? 0 : 1;
 	attr->inherit	    = !opts->no_inherit;
@@ -647,7 +648,7 @@ void perf_evsel__config(struct perf_evsel *evsel,
 		}
 	}
 
-	if (target__has_cpu(&opts->target) || opts->target.force_per_cpu)
+	if (target__has_cpu(&opts->target))
 		perf_evsel__set_sample_bit(evsel, CPU);
 
 	if (opts->period)
@@ -655,7 +656,7 @@ void perf_evsel__config(struct perf_evsel *evsel,
 
 	if (!perf_missing_features.sample_id_all &&
 	    (opts->sample_time || !opts->no_inherit ||
-	     target__has_cpu(&opts->target) || opts->target.force_per_cpu))
+	     target__has_cpu(&opts->target) || per_cpu))
 		perf_evsel__set_sample_bit(evsel, TIME);
 
 	if (opts->raw_samples) {
diff --git a/tools/perf/util/target.c b/tools/perf/util/target.c
index 3c778a07b7cc..e74c5963dc7a 100644
--- a/tools/perf/util/target.c
+++ b/tools/perf/util/target.c
@@ -55,6 +55,13 @@ enum target_errno target__validate(struct target *target)
 			ret = TARGET_ERRNO__UID_OVERRIDE_SYSTEM;
 	}
 
+	/* THREAD and SYSTEM/CPU are mutually exclusive */
+	if (target->per_thread && (target->system_wide || target->cpu_list)) {
+		target->per_thread = false;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__SYSTEM_OVERRIDE_THREAD;
+	}
+
 	return ret;
 }
 
@@ -100,6 +107,7 @@ static const char *target__error_str[] = {
 	"UID switch overriding CPU",
 	"PID/TID switch overriding SYSTEM",
 	"UID switch overriding SYSTEM",
+	"SYSTEM/CPU switch overriding PER-THREAD",
 	"Invalid User: %s",
 	"Problems obtaining information for user %s",
 };
@@ -131,7 +139,8 @@ int target__strerror(struct target *target, int errnum,
 	msg = target__error_str[idx];
 
 	switch (errnum) {
-	case TARGET_ERRNO__PID_OVERRIDE_CPU ... TARGET_ERRNO__UID_OVERRIDE_SYSTEM:
+	case TARGET_ERRNO__PID_OVERRIDE_CPU ...
+	     TARGET_ERRNO__SYSTEM_OVERRIDE_THREAD:
 		snprintf(buf, buflen, "%s", msg);
 		break;
 
diff --git a/tools/perf/util/target.h b/tools/perf/util/target.h
index 2d0c50690892..31dd2e9a27d0 100644
--- a/tools/perf/util/target.h
+++ b/tools/perf/util/target.h
@@ -12,7 +12,8 @@ struct target {
 	uid_t	     uid;
 	bool	     system_wide;
 	bool	     uses_mmap;
-	bool	     force_per_cpu;
+	bool	     default_per_cpu;
+	bool	     per_thread;
 };
 
 enum target_errno {
@@ -33,6 +34,7 @@ enum target_errno {
 	TARGET_ERRNO__UID_OVERRIDE_CPU,
 	TARGET_ERRNO__PID_OVERRIDE_SYSTEM,
 	TARGET_ERRNO__UID_OVERRIDE_SYSTEM,
+	TARGET_ERRNO__SYSTEM_OVERRIDE_THREAD,
 
 	/* for target__parse_uid() */
 	TARGET_ERRNO__INVALID_UID,
-- 
1.8.1.4

[PATCH 11/34] perf tools: Allow '--inherit' as the negation of '--no-inherit'

[Arnaldo Carvalho de Melo]

From: Adrian Hunter <adrian.hunter@intel.com>

Long options can be negated by prefixing them with 'no-'.  However
options that already start with 'no-', such as '--no-inherit' result in
ugly double 'no's.

Avoid that by accepting that the removal of 'no-' also negates the long
option.

Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1384768557-23331-3-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/parse-options.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/tools/perf/util/parse-options.c b/tools/perf/util/parse-options.c
index 31f404a032a9..b6b39ff14fbb 100644
--- a/tools/perf/util/parse-options.c
+++ b/tools/perf/util/parse-options.c
@@ -224,6 +224,24 @@ static int parse_long_opt(struct parse_opt_ctx_t *p, const char *arg,
 			return 0;
 		}
 		if (!rest) {
+			if (!prefixcmp(options->long_name, "no-")) {
+				/*
+				 * The long name itself starts with "no-", so
+				 * accept the option without "no-" so that users
+				 * do not have to enter "no-no-" to get the
+				 * negation.
+				 */
+				rest = skip_prefix(arg, options->long_name + 3);
+				if (rest) {
+					flags |= OPT_UNSET;
+					goto match;
+				}
+				/* Abbreviated case */
+				if (!prefixcmp(options->long_name + 3, arg)) {
+					flags |= OPT_UNSET;
+					goto is_abbreviated;
+				}
+			}
 			/* abbreviated? */
 			if (!strncmp(options->long_name, arg, arg_end - arg)) {
 is_abbreviated:
@@ -259,6 +277,7 @@ is_abbreviated:
 			if (!rest)
 				continue;
 		}
+match:
 		if (*rest) {
 			if (*rest != '=')
 				continue;
-- 
1.8.1.4

[PATCH 12/34] perf tools: Add option macro OPT_BOOLEAN_SET

[Arnaldo Carvalho de Melo]

From: Adrian Hunter <adrian.hunter@intel.com>

OPT_BOOLEAN_SET records whether a boolean option was set by the user.

That information can be used to change the default value for the option
after the options have been parsed.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1384768557-23331-4-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/parse-options.c | 2 ++
 tools/perf/util/parse-options.h | 8 ++++++++
 2 files changed, 10 insertions(+)

diff --git a/tools/perf/util/parse-options.c b/tools/perf/util/parse-options.c
index b6b39ff14fbb..d22e3f8017dc 100644
--- a/tools/perf/util/parse-options.c
+++ b/tools/perf/util/parse-options.c
@@ -78,6 +78,8 @@ static int get_value(struct parse_opt_ctx_t *p,
 
 	case OPTION_BOOLEAN:
 		*(bool *)opt->value = unset ? false : true;
+		if (opt->set)
+			*(bool *)opt->set = true;
 		return 0;
 
 	case OPTION_INCR:
diff --git a/tools/perf/util/parse-options.h b/tools/perf/util/parse-options.h
index b0241e28eaf7..cbf0149cf221 100644
--- a/tools/perf/util/parse-options.h
+++ b/tools/perf/util/parse-options.h
@@ -82,6 +82,9 @@ typedef int parse_opt_cb(const struct option *, const char *arg, int unset);
  *   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.
+ *
+ * `set`::
+ *   whether an option was set by the user
  */
 struct option {
 	enum parse_opt_type type;
@@ -94,6 +97,7 @@ struct option {
 	int flags;
 	parse_opt_cb *callback;
 	intptr_t defval;
+	bool *set;
 };
 
 #define check_vtype(v, type) ( BUILD_BUG_ON_ZERO(!__builtin_types_compatible_p(typeof(v), type)) + v )
@@ -103,6 +107,10 @@ struct option {
 #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_BOOLEAN_SET(s, l, v, os, h) \
+	{ .type = OPTION_BOOLEAN, .short_name = (s), .long_name = (l), \
+	.value = check_vtype(v, bool *), .help = (h), \
+	.set = check_vtype(os, bool *)}
 #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) }
-- 
1.8.1.4

[PATCH 13/34] perf record: Default -t option to no inheritance

[Arnaldo Carvalho de Melo]

From: Adrian Hunter <adrian.hunter@intel.com>

The change to per-cpu mmaps causes the -p, -t and -u options now to have
inheritance enabled by default.  Change that back to no inheritance but
for the -t option only.

Requested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1384768557-23331-5-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-record.txt | 2 ++
 tools/perf/builtin-record.c              | 8 ++++++--
 tools/perf/perf.h                        | 1 +
 3 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index 6ac867e7667f..c407897f0435 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -57,6 +57,8 @@ OPTIONS
 -t::
 --tid=::
         Record events on existing thread ID (comma separated list).
+        This option also disables inheritance by default.  Enable it by adding
+        --inherit.
 
 -u::
 --uid=::
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index f5b18b8fe8c9..65615a8bc25e 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -843,8 +843,9 @@ const struct option record_options[] = {
 	OPT_U64('c', "count", &record.opts.user_interval, "event period to sample"),
 	OPT_STRING('o', "output", &record.file.path, "file",
 		    "output file name"),
-	OPT_BOOLEAN('i', "no-inherit", &record.opts.no_inherit,
-		    "child tasks do not inherit counters"),
+	OPT_BOOLEAN_SET('i', "no-inherit", &record.opts.no_inherit,
+			&record.opts.no_inherit_set,
+			"child tasks do not inherit counters"),
 	OPT_UINTEGER('F', "freq", &record.opts.user_freq, "profile at this frequency"),
 	OPT_CALLBACK('m', "mmap-pages", &record.opts.mmap_pages, "pages",
 		     "number of mmap data pages",
@@ -939,6 +940,9 @@ int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
 		goto out_symbol_exit;
 	}
 
+	if (rec->opts.target.tid && !rec->opts.no_inherit_set)
+		rec->opts.no_inherit = true;
+
 	err = target__validate(&rec->opts.target);
 	if (err) {
 		target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index b079304bd53d..b23fed527514 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -254,6 +254,7 @@ struct perf_record_opts {
 	bool	     inherit_stat;
 	bool	     no_delay;
 	bool	     no_inherit;
+	bool	     no_inherit_set;
 	bool	     no_samples;
 	bool	     raw_samples;
 	bool	     sample_address;
-- 
1.8.1.4

[PATCH 14/34] perf timechart: Always try to print at least 15 tasks

[Arnaldo Carvalho de Melo]

From: Stanislav Fomichev <stfomichev@yandex-team.ru>

Always try to print at least 15 tasks no matter how long they run.

Signed-off-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1383323151-19810-2-git-send-email-stfomichev@yandex-team.ru
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-timechart.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index 41c9bde2fb67..bb21e57ff9bb 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -945,15 +945,17 @@ static void write_svg_file(const char *filename)
 {
 	u64 i;
 	int count;
+	int thresh = TIME_THRESH;
 
 	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);
+	/* We'd like to show at least proc_num tasks;
+	 * be less picky if we have fewer */
+	do {
+		count = determine_display_tasks(thresh);
+		thresh /= 10;
+	} while (!process_filter && thresh && count < 15);
 
 	open_svg(filename, numcpus, count, first_time, last_time);
 
-- 
1.8.1.4

[PATCH 15/34] perf timechart: Add option to limit number of tasks

[Arnaldo Carvalho de Melo]

From: Stanislav Fomichev <stfomichev@yandex-team.ru>

Add -n option to specify min. number of tasks to print.

Signed-off-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1383323151-19810-3-git-send-email-stfomichev@yandex-team.ru
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-timechart.txt | 4 ++++
 tools/perf/builtin-timechart.c              | 5 ++++-
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/tools/perf/Documentation/perf-timechart.txt b/tools/perf/Documentation/perf-timechart.txt
index 3ff8bd4f0b4d..4373144a43b4 100644
--- a/tools/perf/Documentation/perf-timechart.txt
+++ b/tools/perf/Documentation/perf-timechart.txt
@@ -54,6 +54,10 @@ $ perf timechart
 
   Written 10.2 seconds of trace to output.svg.
 
+-n::
+--proc-num::
+        Print task info for at least given number of tasks.
+
 SEE ALSO
 --------
 linkperf:perf-record[1]
diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index bb21e57ff9bb..c352be418f98 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -41,6 +41,7 @@
 #define SUPPORT_OLD_POWER_EVENTS 1
 #define PWR_EVENT_EXIT -1
 
+static int proc_num = 15;
 
 static unsigned int	numcpus;
 static u64		min_freq;	/* Lowest CPU frequency seen */
@@ -955,7 +956,7 @@ static void write_svg_file(const char *filename)
 	do {
 		count = determine_display_tasks(thresh);
 		thresh /= 10;
-	} while (!process_filter && thresh && count < 15);
+	} while (!process_filter && thresh && count < proc_num);
 
 	open_svg(filename, numcpus, count, first_time, last_time);
 
@@ -1102,6 +1103,8 @@ int cmd_timechart(int argc, const char **argv,
 		       parse_process),
 	OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
 		    "Look for files with symbols relative to this directory"),
+	OPT_INTEGER('n', "proc-num", &proc_num,
+		    "min. number of tasks to print"),
 	OPT_END()
 	};
 	const char * const timechart_usage[] = {
-- 
1.8.1.4

[PATCH 16/34] perf timechart: Use proc_num to implement --power-only

[Arnaldo Carvalho de Melo]

From: Stanislav Fomichev <stfomichev@yandex-team.ru>

Don't use special flag to indicate power-only mode, just set proc_num to
0.

Signed-off-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1383323151-19810-4-git-send-email-stfomichev@yandex-team.ru
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-timechart.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index c352be418f98..6410c9ead9e2 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -912,7 +912,7 @@ static int determine_display_tasks(u64 threshold)
 		/* 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)
+		if (p->total_time >= threshold)
 			p->display = 1;
 
 		c = p->all;
@@ -923,7 +923,7 @@ static int determine_display_tasks(u64 threshold)
 			if (c->start_time == 1)
 				c->start_time = first_time;
 
-			if (c->total_time >= threshold && !power_only) {
+			if (c->total_time >= threshold) {
 				c->display = 1;
 				count++;
 			}
@@ -950,6 +950,8 @@ static void write_svg_file(const char *filename)
 
 	numcpus++;
 
+	if (power_only)
+		proc_num = 0;
 
 	/* We'd like to show at least proc_num tasks;
 	 * be less picky if we have fewer */
@@ -967,9 +969,11 @@ static void write_svg_file(const char *filename)
 		svg_cpu_box(i, max_freq, turbo_frequency);
 
 	draw_cpu_usage();
-	draw_process_bars();
+	if (proc_num)
+		draw_process_bars();
 	draw_c_p_states();
-	draw_wakeups();
+	if (proc_num)
+		draw_wakeups();
 
 	svg_close();
 }
-- 
1.8.1.4

[PATCH 17/34] perf timechart: Add support for displaying only tasks related data

[Arnaldo Carvalho de Melo]

From: Stanislav Fomichev <stfomichev@yandex-team.ru>

In order to make SVG smaller and faster to browse add possibility to
switch off power related information with -T switch.

Signed-off-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1383323151-19810-5-git-send-email-stfomichev@yandex-team.ru
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-timechart.txt |  3 +++
 tools/perf/builtin-timechart.c              | 11 ++++++++++-
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/tools/perf/Documentation/perf-timechart.txt b/tools/perf/Documentation/perf-timechart.txt
index 4373144a43b4..d8441894d9ec 100644
--- a/tools/perf/Documentation/perf-timechart.txt
+++ b/tools/perf/Documentation/perf-timechart.txt
@@ -35,6 +35,9 @@ OPTIONS
 -P::
 --power-only::
         Only output the CPU power section of the diagram
+-T::
+--tasks-only::
+        Don't output processor state transitions
 -p::
 --process::
         Select the processes to display, by name or PID
diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index 6410c9ead9e2..b3f175a30d94 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -51,6 +51,7 @@ static u64		turbo_frequency;
 static u64		first_time, last_time;
 
 static bool		power_only;
+static bool		tasks_only;
 
 
 struct per_pid;
@@ -971,7 +972,8 @@ static void write_svg_file(const char *filename)
 	draw_cpu_usage();
 	if (proc_num)
 		draw_process_bars();
-	draw_c_p_states();
+	if (!tasks_only)
+		draw_c_p_states();
 	if (proc_num)
 		draw_wakeups();
 
@@ -1102,6 +1104,8 @@ int cmd_timechart(int argc, const char **argv,
 	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_BOOLEAN('T', "tasks-only", &tasks_only,
+		    "output processes data only"),
 	OPT_CALLBACK('p', "process", NULL, "process",
 		      "process selector. Pass a pid or process name.",
 		       parse_process),
@@ -1119,6 +1123,11 @@ int cmd_timechart(int argc, const char **argv,
 	argc = parse_options(argc, argv, options, timechart_usage,
 			PARSE_OPT_STOP_AT_NON_OPTION);
 
+	if (power_only && tasks_only) {
+		pr_err("-P and -T options cannot be used at the same time.\n");
+		return -1;
+	}
+
 	symbol__init();
 
 	if (argc && !strncmp(argv[0], "rec", 3))
-- 
1.8.1.4

[PATCH 18/34] perf timechart: Group figures and add title with details

[Arnaldo Carvalho de Melo]

From: Stanislav Fomichev <stfomichev@yandex-team.ru>

Add titles to figures so we can run SVG interactively in Firefox and
check event details in the tooltips.

This also aids exploring SVG with Inkscape because when user clicks on
one part of logical figure, all parts are selected.

It's also possible to read titles with Inkscape in the object details.

Signed-off-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1383323151-19810-6-git-send-email-stfomichev@yandex-team.ru
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-timechart.c |  6 ++---
 tools/perf/util/svghelper.c    | 56 ++++++++++++++++++++++++++++++++++++++++--
 tools/perf/util/svghelper.h    |  5 ++--
 3 files changed, 60 insertions(+), 7 deletions(-)

diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index b3f175a30d94..6a848b8c4c16 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -798,11 +798,11 @@ static void draw_process_bars(void)
 			sample = c->samples;
 			while (sample) {
 				if (sample->type == TYPE_RUNNING)
-					svg_sample(Y, sample->cpu, sample->start_time, sample->end_time);
+					svg_running(Y, sample->cpu, sample->start_time, sample->end_time);
 				if (sample->type == TYPE_BLOCKED)
-					svg_box(Y, sample->start_time, sample->end_time, "blocked");
+					svg_blocked(Y, sample->cpu, sample->start_time, sample->end_time);
 				if (sample->type == TYPE_WAITING)
-					svg_waiting(Y, sample->start_time, sample->end_time);
+					svg_waiting(Y, sample->cpu, sample->start_time, sample->end_time);
 				sample = sample->next;
 			}
 
diff --git a/tools/perf/util/svghelper.c b/tools/perf/util/svghelper.c
index 96c866045d60..9a5b41392e01 100644
--- a/tools/perf/util/svghelper.c
+++ b/tools/perf/util/svghelper.c
@@ -95,6 +95,7 @@ void open_svg(const char *filename, int cpus, int rows, u64 start, u64 end)
 
 	total_height = (1 + rows + cpu2slot(cpus)) * SLOT_MULT;
 	fprintf(svgfile, "<?xml version=\"1.0\" standalone=\"no\"?> \n");
+	fprintf(svgfile, "<!DOCTYPE svg SYSTEM \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\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");
@@ -128,12 +129,29 @@ void svg_box(int Yslot, u64 start, u64 end, const char *type)
 		time2pixels(start), time2pixels(end)-time2pixels(start), Yslot * SLOT_MULT, SLOT_HEIGHT, type);
 }
 
-void svg_sample(int Yslot, int cpu, u64 start, u64 end)
+static char *time_to_string(u64 duration);
+void svg_blocked(int Yslot, int cpu, u64 start, u64 end)
+{
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<g>\n");
+	fprintf(svgfile, "<title>#%d blocked %s</title>\n", cpu,
+		time_to_string(end - start));
+	svg_box(Yslot, start, end, "blocked");
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_running(int Yslot, int cpu, u64 start, u64 end)
 {
 	double text_size;
 	if (!svgfile)
 		return;
 
+	fprintf(svgfile, "<g>\n");
+
+	fprintf(svgfile, "<title>#%d running %s</title>\n",
+		cpu, time_to_string(end - start));
 	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);
 
@@ -148,6 +166,7 @@ void svg_sample(int Yslot, int cpu, u64 start, u64 end)
 		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);
 
+	fprintf(svgfile, "</g>\n");
 }
 
 static char *time_to_string(u64 duration)
@@ -168,7 +187,7 @@ static char *time_to_string(u64 duration)
 	return text;
 }
 
-void svg_waiting(int Yslot, u64 start, u64 end)
+void svg_waiting(int Yslot, int cpu, u64 start, u64 end)
 {
 	char *text;
 	const char *style;
@@ -192,6 +211,7 @@ void svg_waiting(int Yslot, u64 start, u64 end)
 	font_size = round_text_size(font_size);
 
 	fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\">\n", time2pixels(start), Yslot * SLOT_MULT);
+	fprintf(svgfile, "<title>#%d waiting %s</title>\n", cpu, time_to_string(end - start));
 	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)
@@ -242,6 +262,8 @@ void svg_cpu_box(int cpu, u64 __max_freq, u64 __turbo_freq)
 	max_freq = __max_freq;
 	turbo_frequency = __turbo_freq;
 
+	fprintf(svgfile, "<g>\n");
+
 	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),
@@ -253,6 +275,8 @@ void svg_cpu_box(int cpu, u64 __max_freq, u64 __turbo_freq)
 
 	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());
+
+	fprintf(svgfile, "</g>\n");
 }
 
 void svg_process(int cpu, u64 start, u64 end, const char *type, const char *name)
@@ -264,6 +288,7 @@ void svg_process(int cpu, u64 start, u64 end, const char *type, const char *name
 
 
 	fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\">\n", time2pixels(start), cpu2y(cpu));
+	fprintf(svgfile, "<title>%s %s</title>\n", name, time_to_string(end - start));
 	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);
@@ -288,6 +313,8 @@ void svg_cstate(int cpu, u64 start, u64 end, int type)
 		return;
 
 
+	fprintf(svgfile, "<g>\n");
+
 	if (type > 6)
 		type = 6;
 	sprintf(style, "c%i", type);
@@ -306,6 +333,8 @@ void svg_cstate(int cpu, u64 start, u64 end, int type)
 	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);
+
+	fprintf(svgfile, "</g>\n");
 }
 
 static char *HzToHuman(unsigned long hz)
@@ -339,6 +368,8 @@ void svg_pstate(int cpu, u64 start, u64 end, u64 freq)
 	if (!svgfile)
 		return;
 
+	fprintf(svgfile, "<g>\n");
+
 	if (max_freq)
 		height = freq * 1.0 / max_freq * (SLOT_HEIGHT + SLOT_MULT);
 	height = 1 + cpu2y(cpu) + SLOT_MULT + SLOT_HEIGHT - height;
@@ -347,6 +378,7 @@ void svg_pstate(int cpu, u64 start, u64 end, u64 freq)
 	fprintf(svgfile, "<text x=\"%4.8f\" y=\"%4.8f\" font-size=\"0.25pt\">%s</text>\n",
 		time2pixels(start), height+0.9, HzToHuman(freq));
 
+	fprintf(svgfile, "</g>\n");
 }
 
 
@@ -358,6 +390,12 @@ void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc
 		return;
 
 
+	fprintf(svgfile, "<g>\n");
+
+	fprintf(svgfile, "<title>%s wakes up %s</title>\n",
+		desc1 ? desc1 : "?",
+		desc2 ? desc2 : "?");
+
 	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",
@@ -395,6 +433,8 @@ void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc
 	if (row1)
 		fprintf(svgfile, "<circle  cx=\"%4.8f\" cy=\"%4.2f\" r = \"0.01\"  style=\"fill:rgb(32,255,32)\"/>\n",
 			time2pixels(start), height);
+
+	fprintf(svgfile, "</g>\n");
 }
 
 void svg_wakeline(u64 start, int row1, int row2)
@@ -405,6 +445,8 @@ void svg_wakeline(u64 start, int row1, int row2)
 		return;
 
 
+	fprintf(svgfile, "<g>\n");
+
 	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);
@@ -417,6 +459,8 @@ void svg_wakeline(u64 start, int row1, int row2)
 		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);
+
+	fprintf(svgfile, "</g>\n");
 }
 
 void svg_interrupt(u64 start, int row)
@@ -424,10 +468,16 @@ void svg_interrupt(u64 start, int row)
 	if (!svgfile)
 		return;
 
+	fprintf(svgfile, "<g>\n");
+
+	fprintf(svgfile, "<title>Wakeup from interrupt</title>\n");
+
 	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);
+
+	fprintf(svgfile, "</g>\n");
 }
 
 void svg_text(int Yslot, u64 start, const char *text)
@@ -455,6 +505,7 @@ void svg_legenda(void)
 	if (!svgfile)
 		return;
 
+	fprintf(svgfile, "<g>\n");
 	svg_legenda_box(0,	"Running", "sample");
 	svg_legenda_box(100,	"Idle","c1");
 	svg_legenda_box(200,	"Deeper Idle", "c3");
@@ -462,6 +513,7 @@ void svg_legenda(void)
 	svg_legenda_box(550,	"Sleeping", "process2");
 	svg_legenda_box(650,	"Waiting for cpu", "waiting");
 	svg_legenda_box(800,	"Blocked on IO", "blocked");
+	fprintf(svgfile, "</g>\n");
 }
 
 void svg_time_grid(void)
diff --git a/tools/perf/util/svghelper.h b/tools/perf/util/svghelper.h
index e0781989cc31..aa533459ab76 100644
--- a/tools/perf/util/svghelper.h
+++ b/tools/perf/util/svghelper.h
@@ -5,8 +5,9 @@
 
 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_blocked(int Yslot, int cpu, u64 start, u64 end);
+extern void svg_running(int Yslot, int cpu, u64 start, u64 end);
+extern void svg_waiting(int Yslot, int cpu, u64 start, u64 end);
 extern void svg_cpu_box(int cpu, u64 max_frequency, u64 turbo_frequency);
 
 
-- 
1.8.1.4

[PATCH 19/34] perf timechart: Add support for -P and -T in timechart recording

[Arnaldo Carvalho de Melo]

From: Stanislav Fomichev <stfomichev@yandex-team.ru>

If we don't want either power or task events we may use -T or -P with
the `perf timechart record` command to filter out events while recording
to keep perf.data small.

Signed-off-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1383323151-19810-7-git-send-email-stfomichev@yandex-team.ru
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-timechart.txt |  16 +++--
 tools/perf/builtin-timechart.c              | 101 +++++++++++++++++++++-------
 2 files changed, 87 insertions(+), 30 deletions(-)

diff --git a/tools/perf/Documentation/perf-timechart.txt b/tools/perf/Documentation/perf-timechart.txt
index d8441894d9ec..8359cfadf0ac 100644
--- a/tools/perf/Documentation/perf-timechart.txt
+++ b/tools/perf/Documentation/perf-timechart.txt
@@ -8,8 +8,7 @@ perf-timechart - Tool to visualize total system behavior during a workload
 SYNOPSIS
 --------
 [verse]
-'perf timechart' record <command>
-'perf timechart' [<options>]
+'perf timechart' [<timechart options>] {record} [<record options>]
 
 DESCRIPTION
 -----------
@@ -21,8 +20,8 @@ There are two variants of perf timechart:
   'perf timechart' to turn a trace into a Scalable Vector Graphics file,
   that can be viewed with popular SVG viewers such as 'Inkscape'.
 
-OPTIONS
--------
+TIMECHART OPTIONS
+-----------------
 -o::
 --output=::
         Select the output file (default: output.svg)
@@ -61,6 +60,15 @@ $ perf timechart
 --proc-num::
         Print task info for at least given number of tasks.
 
+RECORD OPTIONS
+--------------
+-P::
+--power-only::
+        Record only power-related events
+-T::
+--tasks-only::
+        Record only tasks-related events
+
 SEE ALSO
 --------
 linkperf:perf-record[1]
diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index 6a848b8c4c16..1c60ed3f5b97 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -1040,50 +1040,81 @@ out_delete:
 
 static int __cmd_record(int argc, const char **argv)
 {
-#ifdef SUPPORT_OLD_POWER_EVENTS
-	const char * const record_old_args[] = {
+	unsigned int rec_argc, i, j;
+	const char **rec_argv;
+	const char **p;
+	unsigned int record_elems;
+
+	const char * const common_args[] = {
 		"record", "-a", "-R", "-c", "1",
+	};
+	unsigned int common_args_nr = ARRAY_SIZE(common_args);
+
+	const char * const power_args[] = {
+		"-e", "power:cpu_frequency",
+		"-e", "power:cpu_idle",
+	};
+	unsigned int power_args_nr = ARRAY_SIZE(power_args);
+
+	const char * const old_power_args[] = {
+#ifdef SUPPORT_OLD_POWER_EVENTS
 		"-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", "-c", "1",
-		"-e", "power:cpu_frequency",
-		"-e", "power:cpu_idle",
+	};
+	unsigned int old_power_args_nr = ARRAY_SIZE(old_power_args);
+
+	const char * const tasks_args[] = {
 		"-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);
+	unsigned int tasks_args_nr = ARRAY_SIZE(tasks_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);
+		power_args_nr = 0;
+	} else {
+		old_power_args_nr = 0;
 	}
 #endif
 
-	rec_argc = record_elems + argc - 1;
+	if (power_only)
+		tasks_args_nr = 0;
+
+	if (tasks_only) {
+		power_args_nr = 0;
+		old_power_args_nr = 0;
+	}
+
+	record_elems = common_args_nr + tasks_args_nr +
+		power_args_nr + old_power_args_nr;
+
+	rec_argc = record_elems + argc;
 	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]);
+	p = rec_argv;
+	for (i = 0; i < common_args_nr; i++)
+		*p++ = strdup(common_args[i]);
+
+	for (i = 0; i < tasks_args_nr; i++)
+		*p++ = strdup(tasks_args[i]);
+
+	for (i = 0; i < power_args_nr; i++)
+		*p++ = strdup(power_args[i]);
 
-	for (j = 1; j < (unsigned int)argc; j++, i++)
-		rec_argv[i] = argv[j];
+	for (i = 0; i < old_power_args_nr; i++)
+		*p++ = strdup(old_power_args[i]);
 
-	return cmd_record(i, rec_argv, NULL);
+	for (j = 1; j < (unsigned int)argc; j++)
+		*p++ = argv[j];
+
+	return cmd_record(rec_argc, rec_argv, NULL);
 }
 
 static int
@@ -1099,7 +1130,7 @@ int cmd_timechart(int argc, const char **argv,
 		  const char *prefix __maybe_unused)
 {
 	const char *output_name = "output.svg";
-	const struct option options[] = {
+	const struct option timechart_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"),
@@ -1120,7 +1151,17 @@ int cmd_timechart(int argc, const char **argv,
 		NULL
 	};
 
-	argc = parse_options(argc, argv, options, timechart_usage,
+	const struct option record_options[] = {
+	OPT_BOOLEAN('P', "power-only", &power_only, "output power data only"),
+	OPT_BOOLEAN('T', "tasks-only", &tasks_only,
+		    "output processes data only"),
+	OPT_END()
+	};
+	const char * const record_usage[] = {
+		"perf timechart record [<options>]",
+		NULL
+	};
+	argc = parse_options(argc, argv, timechart_options, timechart_usage,
 			PARSE_OPT_STOP_AT_NON_OPTION);
 
 	if (power_only && tasks_only) {
@@ -1130,10 +1171,18 @@ int cmd_timechart(int argc, const char **argv,
 
 	symbol__init();
 
-	if (argc && !strncmp(argv[0], "rec", 3))
+	if (argc && !strncmp(argv[0], "rec", 3)) {
+		argc = parse_options(argc, argv, record_options, record_usage,
+				     PARSE_OPT_STOP_AT_NON_OPTION);
+
+		if (power_only && tasks_only) {
+			pr_err("-P and -T options cannot be used at the same time.\n");
+			return -1;
+		}
+
 		return __cmd_record(argc, argv);
-	else if (argc)
-		usage_with_options(timechart_usage, options);
+	} else if (argc)
+		usage_with_options(timechart_usage, timechart_options);
 
 	setup_pager();
 
-- 
1.8.1.4

[PATCH 20/34] perf timechart: Add backtrace support

[Arnaldo Carvalho de Melo]

From: Stanislav Fomichev <stfomichev@yandex-team.ru>

Add -g flag to `perf timechart record` which saves callchain info in the
perf.data.

When generating SVG, add backtrace information to the figure details, so
now it's possible to see which code path woke up the task and why some
task went to sleep.

Signed-off-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1383323151-19810-8-git-send-email-stfomichev@yandex-team.ru
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-timechart.txt |   3 +
 tools/perf/builtin-timechart.c              | 169 ++++++++++++++++++++++++----
 tools/perf/util/svghelper.c                 |  27 ++++-
 tools/perf/util/svghelper.h                 |  12 +-
 4 files changed, 175 insertions(+), 36 deletions(-)

diff --git a/tools/perf/Documentation/perf-timechart.txt b/tools/perf/Documentation/perf-timechart.txt
index 8359cfadf0ac..271dd4ed5b05 100644
--- a/tools/perf/Documentation/perf-timechart.txt
+++ b/tools/perf/Documentation/perf-timechart.txt
@@ -68,6 +68,9 @@ RECORD OPTIONS
 -T::
 --tasks-only::
         Record only tasks-related events
+-g::
+--callchain::
+        Do call-graph (stack chain/backtrace) recording
 
 SEE ALSO
 --------
diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index 1c60ed3f5b97..491662bdfe0b 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -52,6 +52,7 @@ static u64		first_time, last_time;
 
 static bool		power_only;
 static bool		tasks_only;
+static bool		with_backtrace;
 
 
 struct per_pid;
@@ -126,6 +127,7 @@ struct cpu_sample {
 	u64 end_time;
 	int type;
 	int cpu;
+	const char *backtrace;
 };
 
 static struct per_pid *all_data;
@@ -147,6 +149,7 @@ struct wake_event {
 	int waker;
 	int wakee;
 	u64 time;
+	const char *backtrace;
 };
 
 static struct power_event    *power_events;
@@ -231,7 +234,8 @@ static void pid_exit(int pid, u64 timestamp)
 }
 
 static void
-pid_put_sample(int pid, int type, unsigned int cpu, u64 start, u64 end)
+pid_put_sample(int pid, int type, unsigned int cpu, u64 start, u64 end,
+	       const char *backtrace)
 {
 	struct per_pid *p;
 	struct per_pidcomm *c;
@@ -254,6 +258,7 @@ pid_put_sample(int pid, int type, unsigned int cpu, u64 start, u64 end)
 	sample->type = type;
 	sample->next = c->samples;
 	sample->cpu = cpu;
+	sample->backtrace = backtrace;
 	c->samples = sample;
 
 	if (sample->type == TYPE_RUNNING && end > start && start > 0) {
@@ -405,7 +410,8 @@ static void p_state_change(int cpu, u64 timestamp, u64 new_freq)
 }
 
 static void
-sched_wakeup(int cpu, u64 timestamp, int pid, struct trace_entry *te)
+sched_wakeup(int cpu, u64 timestamp, int pid, struct trace_entry *te,
+	     const char *backtrace)
 {
 	struct per_pid *p;
 	struct wakeup_entry *wake = (void *)te;
@@ -416,6 +422,7 @@ sched_wakeup(int cpu, u64 timestamp, int pid, struct trace_entry *te)
 
 	we->time = timestamp;
 	we->waker = pid;
+	we->backtrace = backtrace;
 
 	if ((te->flags & TRACE_FLAG_HARDIRQ) || (te->flags & TRACE_FLAG_SOFTIRQ))
 		we->waker = -1;
@@ -430,13 +437,15 @@ sched_wakeup(int cpu, u64 timestamp, int pid, struct trace_entry *te)
 		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);
+		pid_put_sample(p->pid, p->current->state, cpu,
+			       p->current->state_since, timestamp, NULL);
 		p->current->state_since = timestamp;
 		p->current->state = TYPE_WAITING;
 	}
 }
 
-static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te)
+static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te,
+			 const char *backtrace)
 {
 	struct per_pid *p = NULL, *prev_p;
 	struct sched_switch *sw = (void *)te;
@@ -447,10 +456,14 @@ static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te)
 	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);
+		pid_put_sample(sw->prev_pid, TYPE_RUNNING, cpu,
+			       prev_p->current->state_since, timestamp,
+			       backtrace);
 	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);
+			pid_put_sample(sw->next_pid, p->current->state, cpu,
+				       p->current->state_since, timestamp,
+				       backtrace);
 
 		p->current->state_since = timestamp;
 		p->current->state = TYPE_RUNNING;
@@ -466,8 +479,87 @@ static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te)
 	}
 }
 
+static const char *cat_backtrace(union perf_event *event,
+				 struct perf_sample *sample,
+				 struct machine *machine)
+{
+	struct addr_location al;
+	unsigned int i;
+	char *p = NULL;
+	size_t p_len;
+	u8 cpumode = PERF_RECORD_MISC_USER;
+	struct addr_location tal;
+	struct ip_callchain *chain = sample->callchain;
+	FILE *f = open_memstream(&p, &p_len);
+
+	if (!f) {
+		perror("open_memstream error");
+		return NULL;
+	}
+
+	if (!chain)
+		goto exit;
+
+	if (perf_event__preprocess_sample(event, machine, &al, sample) < 0) {
+		fprintf(stderr, "problem processing %d event, skipping it.\n",
+			event->header.type);
+		goto exit;
+	}
+
+	for (i = 0; i < chain->nr; i++) {
+		u64 ip;
+
+		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.
+				 */
+				free(p);
+				p = NULL;
+				goto exit;
+			}
+			continue;
+		}
+
+		tal.filtered = false;
+		thread__find_addr_location(al.thread, machine, cpumode,
+					   MAP__FUNCTION, ip, &tal);
+
+		if (tal.sym)
+			fprintf(f, "..... %016" PRIx64 " %s\n", ip,
+				tal.sym->name);
+		else
+			fprintf(f, "..... %016" PRIx64 "\n", ip);
+	}
+
+exit:
+	fclose(f);
+
+	return p;
+}
+
 typedef int (*tracepoint_handler)(struct perf_evsel *evsel,
-				  struct perf_sample *sample);
+				  struct perf_sample *sample,
+				  const char *backtrace);
 
 static int process_sample_event(struct perf_tool *tool __maybe_unused,
 				union perf_event *event __maybe_unused,
@@ -487,7 +579,7 @@ static int process_sample_event(struct perf_tool *tool __maybe_unused,
 
 	if (evsel->handler != NULL) {
 		tracepoint_handler f = evsel->handler;
-		return f(evsel, sample);
+		return f(evsel, sample, cat_backtrace(event, sample, machine));
 	}
 
 	return 0;
@@ -495,7 +587,8 @@ static int process_sample_event(struct perf_tool *tool __maybe_unused,
 
 static int
 process_sample_cpu_idle(struct perf_evsel *evsel __maybe_unused,
-			struct perf_sample *sample)
+			struct perf_sample *sample,
+			const char *backtrace __maybe_unused)
 {
 	struct power_processor_entry *ppe = sample->raw_data;
 
@@ -508,7 +601,8 @@ process_sample_cpu_idle(struct perf_evsel *evsel __maybe_unused,
 
 static int
 process_sample_cpu_frequency(struct perf_evsel *evsel __maybe_unused,
-			     struct perf_sample *sample)
+			     struct perf_sample *sample,
+			     const char *backtrace __maybe_unused)
 {
 	struct power_processor_entry *ppe = sample->raw_data;
 
@@ -518,28 +612,31 @@ process_sample_cpu_frequency(struct perf_evsel *evsel __maybe_unused,
 
 static int
 process_sample_sched_wakeup(struct perf_evsel *evsel __maybe_unused,
-			    struct perf_sample *sample)
+			    struct perf_sample *sample,
+			    const char *backtrace)
 {
 	struct trace_entry *te = sample->raw_data;
 
-	sched_wakeup(sample->cpu, sample->time, sample->pid, te);
+	sched_wakeup(sample->cpu, sample->time, sample->pid, te, backtrace);
 	return 0;
 }
 
 static int
 process_sample_sched_switch(struct perf_evsel *evsel __maybe_unused,
-			    struct perf_sample *sample)
+			    struct perf_sample *sample,
+			    const char *backtrace)
 {
 	struct trace_entry *te = sample->raw_data;
 
-	sched_switch(sample->cpu, sample->time, te);
+	sched_switch(sample->cpu, sample->time, te, backtrace);
 	return 0;
 }
 
 #ifdef SUPPORT_OLD_POWER_EVENTS
 static int
 process_sample_power_start(struct perf_evsel *evsel __maybe_unused,
-			   struct perf_sample *sample)
+			   struct perf_sample *sample,
+			   const char *backtrace __maybe_unused)
 {
 	struct power_entry_old *peo = sample->raw_data;
 
@@ -549,7 +646,8 @@ process_sample_power_start(struct perf_evsel *evsel __maybe_unused,
 
 static int
 process_sample_power_end(struct perf_evsel *evsel __maybe_unused,
-			 struct perf_sample *sample)
+			 struct perf_sample *sample,
+			 const char *backtrace __maybe_unused)
 {
 	c_state_end(sample->cpu, sample->time);
 	return 0;
@@ -557,7 +655,8 @@ process_sample_power_end(struct perf_evsel *evsel __maybe_unused,
 
 static int
 process_sample_power_frequency(struct perf_evsel *evsel __maybe_unused,
-			       struct perf_sample *sample)
+			       struct perf_sample *sample,
+			       const char *backtrace __maybe_unused)
 {
 	struct power_entry_old *peo = sample->raw_data;
 
@@ -741,11 +840,12 @@ static void draw_wakeups(void)
 		}
 
 		if (we->waker == -1)
-			svg_interrupt(we->time, to);
+			svg_interrupt(we->time, to, we->backtrace);
 		else if (from && to && abs(from - to) == 1)
-			svg_wakeline(we->time, from, to);
+			svg_wakeline(we->time, from, to, we->backtrace);
 		else
-			svg_partial_wakeline(we->time, from, task_from, to, task_to);
+			svg_partial_wakeline(we->time, from, task_from, to,
+					     task_to, we->backtrace);
 		we = we->next;
 
 		free(task_from);
@@ -798,11 +898,20 @@ static void draw_process_bars(void)
 			sample = c->samples;
 			while (sample) {
 				if (sample->type == TYPE_RUNNING)
-					svg_running(Y, sample->cpu, sample->start_time, sample->end_time);
+					svg_running(Y, sample->cpu,
+						    sample->start_time,
+						    sample->end_time,
+						    sample->backtrace);
 				if (sample->type == TYPE_BLOCKED)
-					svg_blocked(Y, sample->cpu, sample->start_time, sample->end_time);
+					svg_blocked(Y, sample->cpu,
+						    sample->start_time,
+						    sample->end_time,
+						    sample->backtrace);
 				if (sample->type == TYPE_WAITING)
-					svg_waiting(Y, sample->cpu, sample->start_time, sample->end_time);
+					svg_waiting(Y, sample->cpu,
+						    sample->start_time,
+						    sample->end_time,
+						    sample->backtrace);
 				sample = sample->next;
 			}
 
@@ -1050,6 +1159,11 @@ static int __cmd_record(int argc, const char **argv)
 	};
 	unsigned int common_args_nr = ARRAY_SIZE(common_args);
 
+	const char * const backtrace_args[] = {
+		"-g",
+	};
+	unsigned int backtrace_args_no = ARRAY_SIZE(backtrace_args);
+
 	const char * const power_args[] = {
 		"-e", "power:cpu_frequency",
 		"-e", "power:cpu_idle",
@@ -1089,8 +1203,11 @@ static int __cmd_record(int argc, const char **argv)
 		old_power_args_nr = 0;
 	}
 
+	if (!with_backtrace)
+		backtrace_args_no = 0;
+
 	record_elems = common_args_nr + tasks_args_nr +
-		power_args_nr + old_power_args_nr;
+		power_args_nr + old_power_args_nr + backtrace_args_no;
 
 	rec_argc = record_elems + argc;
 	rec_argv = calloc(rec_argc + 1, sizeof(char *));
@@ -1102,6 +1219,9 @@ static int __cmd_record(int argc, const char **argv)
 	for (i = 0; i < common_args_nr; i++)
 		*p++ = strdup(common_args[i]);
 
+	for (i = 0; i < backtrace_args_no; i++)
+		*p++ = strdup(backtrace_args[i]);
+
 	for (i = 0; i < tasks_args_nr; i++)
 		*p++ = strdup(tasks_args[i]);
 
@@ -1155,6 +1275,7 @@ int cmd_timechart(int argc, const char **argv,
 	OPT_BOOLEAN('P', "power-only", &power_only, "output power data only"),
 	OPT_BOOLEAN('T', "tasks-only", &tasks_only,
 		    "output processes data only"),
+	OPT_BOOLEAN('g', "callchain", &with_backtrace, "record callchain"),
 	OPT_END()
 	};
 	const char * const record_usage[] = {
diff --git a/tools/perf/util/svghelper.c b/tools/perf/util/svghelper.c
index 9a5b41392e01..8b79d3ad1246 100644
--- a/tools/perf/util/svghelper.c
+++ b/tools/perf/util/svghelper.c
@@ -130,7 +130,7 @@ void svg_box(int Yslot, u64 start, u64 end, const char *type)
 }
 
 static char *time_to_string(u64 duration);
-void svg_blocked(int Yslot, int cpu, u64 start, u64 end)
+void svg_blocked(int Yslot, int cpu, u64 start, u64 end, const char *backtrace)
 {
 	if (!svgfile)
 		return;
@@ -138,11 +138,13 @@ void svg_blocked(int Yslot, int cpu, u64 start, u64 end)
 	fprintf(svgfile, "<g>\n");
 	fprintf(svgfile, "<title>#%d blocked %s</title>\n", cpu,
 		time_to_string(end - start));
+	if (backtrace)
+		fprintf(svgfile, "<desc>Blocked on:\n%s</desc>\n", backtrace);
 	svg_box(Yslot, start, end, "blocked");
 	fprintf(svgfile, "</g>\n");
 }
 
-void svg_running(int Yslot, int cpu, u64 start, u64 end)
+void svg_running(int Yslot, int cpu, u64 start, u64 end, const char *backtrace)
 {
 	double text_size;
 	if (!svgfile)
@@ -152,6 +154,8 @@ void svg_running(int Yslot, int cpu, u64 start, u64 end)
 
 	fprintf(svgfile, "<title>#%d running %s</title>\n",
 		cpu, time_to_string(end - start));
+	if (backtrace)
+		fprintf(svgfile, "<desc>Switched because:\n%s</desc>\n", backtrace);
 	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);
 
@@ -187,7 +191,7 @@ static char *time_to_string(u64 duration)
 	return text;
 }
 
-void svg_waiting(int Yslot, int cpu, u64 start, u64 end)
+void svg_waiting(int Yslot, int cpu, u64 start, u64 end, const char *backtrace)
 {
 	char *text;
 	const char *style;
@@ -212,6 +216,8 @@ void svg_waiting(int Yslot, int cpu, u64 start, u64 end)
 
 	fprintf(svgfile, "<g transform=\"translate(%4.8f,%4.8f)\">\n", time2pixels(start), Yslot * SLOT_MULT);
 	fprintf(svgfile, "<title>#%d waiting %s</title>\n", cpu, time_to_string(end - start));
+	if (backtrace)
+		fprintf(svgfile, "<desc>Waiting on:\n%s</desc>\n", backtrace);
 	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)
@@ -382,7 +388,7 @@ void svg_pstate(int cpu, u64 start, u64 end, u64 freq)
 }
 
 
-void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2)
+void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2, const char *backtrace)
 {
 	double height;
 
@@ -396,6 +402,9 @@ void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc
 		desc1 ? desc1 : "?",
 		desc2 ? desc2 : "?");
 
+	if (backtrace)
+		fprintf(svgfile, "<desc>%s</desc>\n", backtrace);
+
 	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",
@@ -437,7 +446,7 @@ void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc
 	fprintf(svgfile, "</g>\n");
 }
 
-void svg_wakeline(u64 start, int row1, int row2)
+void svg_wakeline(u64 start, int row1, int row2, const char *backtrace)
 {
 	double height;
 
@@ -447,6 +456,9 @@ void svg_wakeline(u64 start, int row1, int row2)
 
 	fprintf(svgfile, "<g>\n");
 
+	if (backtrace)
+		fprintf(svgfile, "<desc>%s</desc>\n", backtrace);
+
 	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);
@@ -463,7 +475,7 @@ void svg_wakeline(u64 start, int row1, int row2)
 	fprintf(svgfile, "</g>\n");
 }
 
-void svg_interrupt(u64 start, int row)
+void svg_interrupt(u64 start, int row, const char *backtrace)
 {
 	if (!svgfile)
 		return;
@@ -472,6 +484,9 @@ void svg_interrupt(u64 start, int row)
 
 	fprintf(svgfile, "<title>Wakeup from interrupt</title>\n");
 
+	if (backtrace)
+		fprintf(svgfile, "<desc>%s</desc>\n", backtrace);
+
 	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",
diff --git a/tools/perf/util/svghelper.h b/tools/perf/util/svghelper.h
index aa533459ab76..fad79ceeac1a 100644
--- a/tools/perf/util/svghelper.h
+++ b/tools/perf/util/svghelper.h
@@ -5,9 +5,9 @@
 
 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_blocked(int Yslot, int cpu, u64 start, u64 end);
-extern void svg_running(int Yslot, int cpu, u64 start, u64 end);
-extern void svg_waiting(int Yslot, int cpu, u64 start, u64 end);
+extern void svg_blocked(int Yslot, int cpu, u64 start, u64 end, const char *backtrace);
+extern void svg_running(int Yslot, int cpu, u64 start, u64 end, const char *backtrace);
+extern void svg_waiting(int Yslot, int cpu, u64 start, u64 end, const char *backtrace);
 extern void svg_cpu_box(int cpu, u64 max_frequency, u64 turbo_frequency);
 
 
@@ -18,9 +18,9 @@ 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_wakeline(u64 start, int row1, int row2, const char *backtrace);
+extern void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2, const char *backtrace);
+extern void svg_interrupt(u64 start, int row, const char *backtrace);
 extern void svg_text(int Yslot, u64 start, const char *text);
 extern void svg_close(void);
 
-- 
1.8.1.4

[PATCH 21/34] perf evsel: Skip ignored symbols while printing callchain

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

Allows a command to have a symbol_filter controlled by the user to skip
certain functions in a backtrace. One example is to allow the user to
reduce repeating patterns like:

    do_select  core_sys_select  sys_select

to just sys_select when dumping callchains, consuming less real estate
on the screen while still conveying the essential message - the process
is in a select call.

This option is leveraged by the upcoming timehist command.

Signed-off-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1384806771-2945-2-git-send-email-dsahern@gmail.com
[ Checked if al.sym is NULL before touching al.sym->ignored, as noted by Adrian Hunter ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/session.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index f36d24a02445..b0b15e213df5 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -1522,6 +1522,9 @@ void perf_evsel__print_ip(struct perf_evsel *evsel, union perf_event *event,
 			if (!node)
 				break;
 
+			if (node->sym && node->sym->ignore)
+				goto next;
+
 			if (print_ip)
 				printf("%c%16" PRIx64, s, node->ip);
 
@@ -1544,12 +1547,15 @@ void perf_evsel__print_ip(struct perf_evsel *evsel, union perf_event *event,
 			if (!print_oneline)
 				printf("\n");
 
-			callchain_cursor_advance(&callchain_cursor);
-
 			stack_depth--;
+next:
+			callchain_cursor_advance(&callchain_cursor);
 		}
 
 	} else {
+		if (al.sym && al.sym->ignore)
+			return;
+
 		if (print_ip)
 			printf("%16" PRIx64, sample->ip);
 
-- 
1.8.1.4

[PATCH 22/34] perf symbols: Move idle syms check from top to generic function

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

Allows list of idle symbols to be leveraged by other commands, such as
the upcoming timehist command.

Signed-off-by: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1384806771-2945-3-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-top.c | 25 ++-----------------------
 tools/perf/util/symbol.c | 30 ++++++++++++++++++++++++++++++
 tools/perf/util/symbol.h |  1 +
 3 files changed, 33 insertions(+), 23 deletions(-)

diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 531522d3d97b..03d37a76c612 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -634,26 +634,9 @@ repeat:
 	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
@@ -671,12 +654,8 @@ static int symbol_filter(struct map *map __maybe_unused, struct symbol *sym)
 	    strstr(name, "_text_end"))
 		return 1;
 
-	for (i = 0; skip_symbols[i]; i++) {
-		if (!strcmp(skip_symbols[i], name)) {
-			sym->ignore = true;
-			break;
-		}
-	}
+	if (symbol__is_idle(sym))
+		sym->ignore = true;
 
 	return 0;
 }
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index c0c36965fff0..f55c18da1e40 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -573,6 +573,36 @@ static u8 kallsyms2elf_type(char type)
 	return isupper(type) ? STB_GLOBAL : STB_LOCAL;
 }
 
+bool symbol__is_idle(struct symbol *sym)
+{
+	const char * const idle_symbols[] = {
+		"cpu_idle",
+		"intel_idle",
+		"default_idle",
+		"native_safe_halt",
+		"enter_idle",
+		"exit_idle",
+		"mwait_idle",
+		"mwait_idle_with_hints",
+		"poll_idle",
+		"ppc64_runlatch_off",
+		"pseries_dedicated_idle_sleep",
+		NULL
+	};
+
+	int i;
+
+	if (!sym)
+		return false;
+
+	for (i = 0; idle_symbols[i]; i++) {
+		if (!strcmp(idle_symbols[i], sym->name))
+			return true;
+	}
+
+	return false;
+}
+
 static int map__process_kallsym_symbol(void *arg, const char *name,
 				       char type, u64 start)
 {
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index 07de8fea2f48..ad13c5d50b91 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -240,6 +240,7 @@ 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);
+bool symbol__is_idle(struct symbol *sym);
 
 int dso__load_sym(struct dso *dso, struct map *map, struct symsrc *syms_ss,
 		  struct symsrc *runtime_ss, symbol_filter_t filter,
-- 
1.8.1.4

[PATCH 23/34] perf thread: Move comm_list check into function

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

Signed-off-by: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1384806771-2945-5-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/event.c  |  3 +--
 tools/perf/util/thread.h | 12 ++++++++++++
 2 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index bb788c109fe6..c77814bf01e1 100644
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -732,8 +732,7 @@ int perf_event__preprocess_sample(const union perf_event *event,
 	if (thread == NULL)
 		return -1;
 
-	if (symbol_conf.comm_list &&
-	    !strlist__has_entry(symbol_conf.comm_list, thread__comm_str(thread)))
+	if (thread__is_filtered(thread))
 		goto out_filtered;
 
 	dump_printf(" ... thread: %s:%d\n", thread__comm_str(thread), thread->tid);
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
index 897c1b2a750a..5b856bf942e1 100644
--- a/tools/perf/util/thread.h
+++ b/tools/perf/util/thread.h
@@ -6,6 +6,7 @@
 #include <unistd.h>
 #include <sys/types.h>
 #include "symbol.h"
+#include <strlist.h>
 
 struct thread {
 	union {
@@ -66,4 +67,15 @@ static inline void thread__set_priv(struct thread *thread, void *p)
 {
 	thread->priv = p;
 }
+
+static inline bool thread__is_filtered(struct thread *thread)
+{
+	if (symbol_conf.comm_list &&
+	    !strlist__has_entry(symbol_conf.comm_list, thread__comm_str(thread))) {
+		return true;
+	}
+
+	return false;
+}
+
 #endif	/* __PERF_THREAD_H */
-- 
1.8.1.4

[PATCH 24/34] perf tools: Export setup_list

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

Used in upcoming patches (perf sched timehist command).

Signed-off-by: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1384806771-2945-6-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/symbol.c | 2 +-
 tools/perf/util/symbol.h | 3 +++
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index f55c18da1e40..ce9ce101adf4 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -1749,7 +1749,7 @@ out_fail:
 	return -1;
 }
 
-static int setup_list(struct strlist **list, const char *list_str,
+int setup_list(struct strlist **list, const char *list_str,
 		      const char *list_name)
 {
 	if (list_str == NULL)
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index ad13c5d50b91..f1031a1358a1 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -274,4 +274,7 @@ void kcore_extract__delete(struct kcore_extract *kce);
 int kcore_copy(const char *from_dir, const char *to_dir);
 int compare_proc_modules(const char *from, const char *to);
 
+int setup_list(struct strlist **list, const char *list_str,
+	       const char *list_name);
+
 #endif /* __PERF_SYMBOL */
-- 
1.8.1.4

[PATCH 25/34] perf script: Print callchains and symbols if they exist

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

The intent of perf-script is to dump the events and information in the
file. H/W, S/W and raw events all dump callchains if they are present;
might as well make that the default for tracepoints too.

v2: Only add options for sym, dso and ip if callchains are present

Signed-off-by: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/1384920457-5986-1-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-script.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index b392770766dd..9f3ba4404a14 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -280,6 +280,30 @@ static int perf_session__check_output_opt(struct perf_session *session)
 		set_print_ip_opts(&evsel->attr);
 	}
 
+	/*
+	 * set default for tracepoints to print symbols only
+	 * if callchains are present
+	 */
+	if (symbol_conf.use_callchain &&
+	    !output[PERF_TYPE_TRACEPOINT].user_set) {
+		struct perf_event_attr *attr;
+
+		j = PERF_TYPE_TRACEPOINT;
+		evsel = perf_session__find_first_evtype(session, j);
+		if (evsel == NULL)
+			goto out;
+
+		attr = &evsel->attr;
+
+		if (attr->sample_type & PERF_SAMPLE_CALLCHAIN) {
+			output[j].fields |= PERF_OUTPUT_IP;
+			output[j].fields |= PERF_OUTPUT_SYM;
+			output[j].fields |= PERF_OUTPUT_DSO;
+			set_print_ip_opts(attr);
+		}
+	}
+
+out:
 	return 0;
 }
 
-- 
1.8.1.4

[PATCH 26/34] perf script: Print comm, fork and exit events also

[Arnaldo Carvalho de Melo]

From: Namhyung Kim <namhyung@kernel.org>

If --show-task-events option is given, also print internal COMM, FORK
and EXIT events.  It would be helpful for debugging.

  $ perf script --show-task-events
  ...
         swapper     0 [009] 3350640.335261: sched:sched_switch: prev_comm=swapper/9
           sleep  9486 [009] 3350640.335509: PERF_RECORD_COMM: sleep:9486
           sleep  9486 [009] 3350640.335806: sched:sched_stat_runtime: comm=sleep pid=9486
         firefox  2635 [003] 3350641.275896: PERF_RECORD_FORK(2635:9487):(2635:2635)
         firefox  2635 [003] 3350641.275896: sched:sched_process_fork: comm=firefox pid=2635
           sleep  9486 [009] 3350641.336009: PERF_RECORD_EXIT(9486:9486):(9486:9486)

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Suggested-by: Frederic Weisbecker <fweisbec@gmail.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1385455873-25865-1-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-script.txt |   3 +
 tools/perf/builtin-script.c              | 105 +++++++++++++++++++++++++++++++
 2 files changed, 108 insertions(+)

diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt
index e9cbfcddfa3f..67af9b77ea78 100644
--- a/tools/perf/Documentation/perf-script.txt
+++ b/tools/perf/Documentation/perf-script.txt
@@ -203,6 +203,9 @@ OPTIONS
 --show-kernel-path::
 	Try to resolve the path of [kernel.kallsyms]
 
+--show-task-events
+	Display task related events (e.g. FORK, COMM, EXIT).
+
 SEE ALSO
 --------
 linkperf:perf-record[1], linkperf:perf-script-perl[1],
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index 9f3ba4404a14..e2b9aff6506e 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -572,6 +572,7 @@ static int process_sample_event(struct perf_tool *tool __maybe_unused,
 struct perf_script {
 	struct perf_tool	tool;
 	struct perf_session	*session;
+	bool			show_task_events;
 };
 
 static int process_attr(struct perf_tool *tool, union perf_event *event,
@@ -602,6 +603,101 @@ static int process_attr(struct perf_tool *tool, union perf_event *event,
 	return perf_evsel__check_attr(evsel, scr->session);
 }
 
+static int process_comm_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+	int ret = -1;
+
+	thread = machine__findnew_thread(machine, event->comm.pid, event->comm.tid);
+	if (thread == NULL) {
+		pr_debug("problem processing COMM event, skipping it.\n");
+		return -1;
+	}
+
+	if (perf_event__process_comm(tool, event, sample, machine) < 0)
+		goto out;
+
+	if (!evsel->attr.sample_id_all) {
+		sample->cpu = 0;
+		sample->time = 0;
+		sample->tid = event->comm.tid;
+		sample->pid = event->comm.pid;
+	}
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+	ret = 0;
+
+out:
+	return ret;
+}
+
+static int process_fork_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+
+	if (perf_event__process_fork(tool, event, sample, machine) < 0)
+		return -1;
+
+	thread = machine__findnew_thread(machine, event->fork.pid, event->fork.tid);
+	if (thread == NULL) {
+		pr_debug("problem processing FORK event, skipping it.\n");
+		return -1;
+	}
+
+	if (!evsel->attr.sample_id_all) {
+		sample->cpu = 0;
+		sample->time = event->fork.time;
+		sample->tid = event->fork.tid;
+		sample->pid = event->fork.pid;
+	}
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+
+	return 0;
+}
+static int process_exit_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+
+	thread = machine__findnew_thread(machine, event->fork.pid, event->fork.tid);
+	if (thread == NULL) {
+		pr_debug("problem processing EXIT event, skipping it.\n");
+		return -1;
+	}
+
+	if (!evsel->attr.sample_id_all) {
+		sample->cpu = 0;
+		sample->time = 0;
+		sample->tid = event->comm.tid;
+		sample->pid = event->comm.pid;
+	}
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+
+	if (perf_event__process_exit(tool, event, sample, machine) < 0)
+		return -1;
+
+	return 0;
+}
+
 static void sig_handler(int sig __maybe_unused)
 {
 	session_done = 1;
@@ -613,6 +709,13 @@ static int __cmd_script(struct perf_script *script)
 
 	signal(SIGINT, sig_handler);
 
+	/* override event processing functions */
+	if (script->show_task_events) {
+		script->tool.comm = process_comm_event;
+		script->tool.fork = process_fork_event;
+		script->tool.exit = process_exit_event;
+	}
+
 	ret = perf_session__process_events(script->session, &script->tool);
 
 	if (debug_mode)
@@ -1375,6 +1478,8 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 		    "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_BOOLEAN('\0', "show-task-events", &script.show_task_events,
+		    "Show the fork/comm/exit events"),
 	OPT_END()
 	};
 	const char * const script_usage[] = {
-- 
1.8.1.4

[PATCH 27/34] perf script: Print mmap[2] events also

[Arnaldo Carvalho de Melo]

From: Namhyung Kim <namhyung@kernel.org>

If --show-mmap-events option is given, also print internal MMAP and
MMAP2 events.  It would be helpful for debugging.

  $ perf script --show-mmap-events
  ...
           sleep  9486 [009] 3350640.335531: PERF_RECORD_MMAP 9486/9486: [0x400000(0x6000) @ 0]: x /usr/bin/sleep
           sleep  9486 [009] 3350640.335542: PERF_RECORD_MMAP 9486/9486: [0x3153a00000(0x223000) @ 0]: x /usr/lib64/ld-2.17.so
           sleep  9486 [009] 3350640.335553: PERF_RECORD_MMAP 9486/9486: [0x7fff8b5fe000(0x2000) @ 0x7fff8b5fe000]: x [vdso]
           sleep  9486 [009] 3350640.335643: PERF_RECORD_MMAP 9486/9486: [0x3153e00000(0x3c0000) @ 0]: x /usr/lib64/libc-2.17.so

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Suggested-by: Frederic Weisbecker <fweisbec@gmail.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1385456066-26592-1-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-script.txt |  3 ++
 tools/perf/builtin-script.c              | 69 ++++++++++++++++++++++++++++++++
 2 files changed, 72 insertions(+)

diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt
index 67af9b77ea78..cfdbb1e045b5 100644
--- a/tools/perf/Documentation/perf-script.txt
+++ b/tools/perf/Documentation/perf-script.txt
@@ -206,6 +206,9 @@ OPTIONS
 --show-task-events
 	Display task related events (e.g. FORK, COMM, EXIT).
 
+--show-mmap-events
+	Display mmap related events (e.g. MMAP, MMAP2).
+
 SEE ALSO
 --------
 linkperf:perf-record[1], linkperf:perf-script-perl[1],
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index e2b9aff6506e..952dce979252 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -573,6 +573,7 @@ struct perf_script {
 	struct perf_tool	tool;
 	struct perf_session	*session;
 	bool			show_task_events;
+	bool			show_mmap_events;
 };
 
 static int process_attr(struct perf_tool *tool, union perf_event *event,
@@ -698,6 +699,68 @@ static int process_exit_event(struct perf_tool *tool,
 	return 0;
 }
 
+static int process_mmap_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+
+	if (perf_event__process_mmap(tool, event, sample, machine) < 0)
+		return -1;
+
+	thread = machine__findnew_thread(machine, event->mmap.pid, event->mmap.tid);
+	if (thread == NULL) {
+		pr_debug("problem processing MMAP event, skipping it.\n");
+		return -1;
+	}
+
+	if (!evsel->attr.sample_id_all) {
+		sample->cpu = 0;
+		sample->time = 0;
+		sample->tid = event->mmap.tid;
+		sample->pid = event->mmap.pid;
+	}
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+
+	return 0;
+}
+
+static int process_mmap2_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+
+	if (perf_event__process_mmap2(tool, event, sample, machine) < 0)
+		return -1;
+
+	thread = machine__findnew_thread(machine, event->mmap2.pid, event->mmap2.tid);
+	if (thread == NULL) {
+		pr_debug("problem processing MMAP2 event, skipping it.\n");
+		return -1;
+	}
+
+	if (!evsel->attr.sample_id_all) {
+		sample->cpu = 0;
+		sample->time = 0;
+		sample->tid = event->mmap2.tid;
+		sample->pid = event->mmap2.pid;
+	}
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+
+	return 0;
+}
+
 static void sig_handler(int sig __maybe_unused)
 {
 	session_done = 1;
@@ -715,6 +778,10 @@ static int __cmd_script(struct perf_script *script)
 		script->tool.fork = process_fork_event;
 		script->tool.exit = process_exit_event;
 	}
+	if (script->show_mmap_events) {
+		script->tool.mmap = process_mmap_event;
+		script->tool.mmap2 = process_mmap2_event;
+	}
 
 	ret = perf_session__process_events(script->session, &script->tool);
 
@@ -1480,6 +1547,8 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 		    "Show the path of [kernel.kallsyms]"),
 	OPT_BOOLEAN('\0', "show-task-events", &script.show_task_events,
 		    "Show the fork/comm/exit events"),
+	OPT_BOOLEAN('\0', "show-mmap-events", &script.show_mmap_events,
+		    "Show the mmap events"),
 	OPT_END()
 	};
 	const char * const script_usage[] = {
-- 
1.8.1.4

[PATCH 28/34] perf symbols: Fix not finding kcore in buildid cache

[Arnaldo Carvalho de Melo]

From: Adrian Hunter <adrian.hunter@intel.com>

The logic was not looking in the buildid cache for kcore if the host
kernel buildid did not match the recorded kernel buildid.

This affects the non-live case i.e. the kernel has changed and we are
looking at a special copy of kcore that we placed in the buildid cache
(using "perf buildid-cache -v -k /proc/kcore") when the data was
recorded.

After this fix kernel symbols get resolved/annotated correctly.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1385471964-4037-1-git-send-email-adrian.hunter@intel.com
[ Added further explanation extracted from conversation between Ingo & Adrian on lkml ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/symbol.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index ce9ce101adf4..360eefecb81d 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -1526,14 +1526,15 @@ static char *dso__find_kallsyms(struct dso *dso, struct map *map)
 
 	build_id__sprintf(dso->build_id, sizeof(dso->build_id), sbuild_id);
 
+	scnprintf(path, sizeof(path), "%s/[kernel.kcore]/%s", buildid_dir,
+		  sbuild_id);
+
 	/* Use /proc/kallsyms if possible */
 	if (is_host) {
 		DIR *d;
 		int fd;
 
 		/* If no cached kcore go with /proc/kallsyms */
-		scnprintf(path, sizeof(path), "%s/[kernel.kcore]/%s",
-			  buildid_dir, sbuild_id);
 		d = opendir(path);
 		if (!d)
 			goto proc_kallsyms;
@@ -1558,6 +1559,10 @@ static char *dso__find_kallsyms(struct dso *dso, struct map *map)
 		goto proc_kallsyms;
 	}
 
+	/* Find kallsyms in build-id cache with kcore */
+	if (!find_matching_kcore(map, path, sizeof(path)))
+		return strdup(path);
+
 	scnprintf(path, sizeof(path), "%s/[kernel.kallsyms]/%s",
 		  buildid_dir, sbuild_id);
 
-- 
1.8.1.4

[PATCH 29/34] perf timechart: dynamically determine event fields offset

[Arnaldo Carvalho de Melo]

From: Stanislav Fomichev <stfomichev@yandex-team.ru>

Since b000c8065a92 "tracing: Remove the extra 4 bytes of padding in
events" removed padding bytes, perf timechart got out of sync with the
kernel's trace_entry structure.

Convert perf timechart to use dynamic fields offsets (via
perf_evsel__intval) not relying on a hardcoded copy of fields layout
from the kernel.

Signed-off-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Chia-I Wu <olvaffe@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20131127104459.GB3309@stfomichev-desktop
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-timechart.c | 119 +++++++++++++++--------------------------
 1 file changed, 42 insertions(+), 77 deletions(-)

diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index 491662bdfe0b..436cb5f9d751 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -306,50 +306,10 @@ static int process_exit_event(struct perf_tool *tool __maybe_unused,
 	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;
-};
-
-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;
@@ -409,25 +369,23 @@ static void p_state_change(int cpu, u64 timestamp, u64 new_freq)
 			turbo_frequency = max_freq;
 }
 
-static void
-sched_wakeup(int cpu, u64 timestamp, int pid, struct trace_entry *te,
-	     const char *backtrace)
+static void sched_wakeup(int cpu, u64 timestamp, int waker, int wakee,
+			 u8 flags, const char *backtrace)
 {
 	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;
+	we->waker = waker;
 	we->backtrace = backtrace;
 
-	if ((te->flags & TRACE_FLAG_HARDIRQ) || (te->flags & TRACE_FLAG_SOFTIRQ))
+	if ((flags & TRACE_FLAG_HARDIRQ) || (flags & TRACE_FLAG_SOFTIRQ))
 		we->waker = -1;
 
-	we->wakee = wake->pid;
+	we->wakee = wakee;
 	we->next = wake_events;
 	wake_events = we;
 	p = find_create_pid(we->wakee);
@@ -444,24 +402,22 @@ sched_wakeup(int cpu, u64 timestamp, int pid, struct trace_entry *te,
 	}
 }
 
-static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te,
-			 const char *backtrace)
+static void sched_switch(int cpu, u64 timestamp, int prev_pid, int next_pid,
+			 u64 prev_state, const char *backtrace)
 {
 	struct per_pid *p = NULL, *prev_p;
-	struct sched_switch *sw = (void *)te;
-
 
-	prev_p = find_create_pid(sw->prev_pid);
+	prev_p = find_create_pid(prev_pid);
 
-	p = find_create_pid(sw->next_pid);
+	p = find_create_pid(next_pid);
 
 	if (prev_p->current && prev_p->current->state != TYPE_NONE)
-		pid_put_sample(sw->prev_pid, TYPE_RUNNING, cpu,
+		pid_put_sample(prev_pid, TYPE_RUNNING, cpu,
 			       prev_p->current->state_since, timestamp,
 			       backtrace);
 	if (p && p->current) {
 		if (p->current->state != TYPE_NONE)
-			pid_put_sample(sw->next_pid, p->current->state, cpu,
+			pid_put_sample(next_pid, p->current->state, cpu,
 				       p->current->state_since, timestamp,
 				       backtrace);
 
@@ -472,9 +428,9 @@ static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te,
 	if (prev_p->current) {
 		prev_p->current->state = TYPE_NONE;
 		prev_p->current->state_since = timestamp;
-		if (sw->prev_state & 2)
+		if (prev_state & 2)
 			prev_p->current->state = TYPE_BLOCKED;
-		if (sw->prev_state == 0)
+		if (prev_state == 0)
 			prev_p->current->state = TYPE_WAITING;
 	}
 }
@@ -586,61 +542,69 @@ static int process_sample_event(struct perf_tool *tool __maybe_unused,
 }
 
 static int
-process_sample_cpu_idle(struct perf_evsel *evsel __maybe_unused,
+process_sample_cpu_idle(struct perf_evsel *evsel,
 			struct perf_sample *sample,
 			const char *backtrace __maybe_unused)
 {
-	struct power_processor_entry *ppe = sample->raw_data;
+	u32 state = perf_evsel__intval(evsel, sample, "state");
+	u32 cpu_id = perf_evsel__intval(evsel, sample, "cpu_id");
 
-	if (ppe->state == (u32) PWR_EVENT_EXIT)
-		c_state_end(ppe->cpu_id, sample->time);
+	if (state == (u32)PWR_EVENT_EXIT)
+		c_state_end(cpu_id, sample->time);
 	else
-		c_state_start(ppe->cpu_id, sample->time, ppe->state);
+		c_state_start(cpu_id, sample->time, state);
 	return 0;
 }
 
 static int
-process_sample_cpu_frequency(struct perf_evsel *evsel __maybe_unused,
+process_sample_cpu_frequency(struct perf_evsel *evsel,
 			     struct perf_sample *sample,
 			     const char *backtrace __maybe_unused)
 {
-	struct power_processor_entry *ppe = sample->raw_data;
+	u32 state = perf_evsel__intval(evsel, sample, "state");
+	u32 cpu_id = perf_evsel__intval(evsel, sample, "cpu_id");
 
-	p_state_change(ppe->cpu_id, sample->time, ppe->state);
+	p_state_change(cpu_id, sample->time, state);
 	return 0;
 }
 
 static int
-process_sample_sched_wakeup(struct perf_evsel *evsel __maybe_unused,
+process_sample_sched_wakeup(struct perf_evsel *evsel,
 			    struct perf_sample *sample,
 			    const char *backtrace)
 {
-	struct trace_entry *te = sample->raw_data;
+	u8 flags = perf_evsel__intval(evsel, sample, "common_flags");
+	int waker = perf_evsel__intval(evsel, sample, "common_pid");
+	int wakee = perf_evsel__intval(evsel, sample, "pid");
 
-	sched_wakeup(sample->cpu, sample->time, sample->pid, te, backtrace);
+	sched_wakeup(sample->cpu, sample->time, waker, wakee, flags, backtrace);
 	return 0;
 }
 
 static int
-process_sample_sched_switch(struct perf_evsel *evsel __maybe_unused,
+process_sample_sched_switch(struct perf_evsel *evsel,
 			    struct perf_sample *sample,
 			    const char *backtrace)
 {
-	struct trace_entry *te = sample->raw_data;
+	int prev_pid = perf_evsel__intval(evsel, sample, "prev_pid");
+	int next_pid = perf_evsel__intval(evsel, sample, "next_pid");
+	u64 prev_state = perf_evsel__intval(evsel, sample, "prev_state");
 
-	sched_switch(sample->cpu, sample->time, te, backtrace);
+	sched_switch(sample->cpu, sample->time, prev_pid, next_pid, prev_state,
+		     backtrace);
 	return 0;
 }
 
 #ifdef SUPPORT_OLD_POWER_EVENTS
 static int
-process_sample_power_start(struct perf_evsel *evsel __maybe_unused,
+process_sample_power_start(struct perf_evsel *evsel,
 			   struct perf_sample *sample,
 			   const char *backtrace __maybe_unused)
 {
-	struct power_entry_old *peo = sample->raw_data;
+	u64 cpu_id = perf_evsel__intval(evsel, sample, "cpu_id");
+	u64 value = perf_evsel__intval(evsel, sample, "value");
 
-	c_state_start(peo->cpu_id, sample->time, peo->value);
+	c_state_start(cpu_id, sample->time, value);
 	return 0;
 }
 
@@ -654,13 +618,14 @@ process_sample_power_end(struct perf_evsel *evsel __maybe_unused,
 }
 
 static int
-process_sample_power_frequency(struct perf_evsel *evsel __maybe_unused,
+process_sample_power_frequency(struct perf_evsel *evsel,
 			       struct perf_sample *sample,
 			       const char *backtrace __maybe_unused)
 {
-	struct power_entry_old *peo = sample->raw_data;
+	u64 cpu_id = perf_evsel__intval(evsel, sample, "cpu_id");
+	u64 value = perf_evsel__intval(evsel, sample, "value");
 
-	p_state_change(peo->cpu_id, sample->time, peo->value);
+	p_state_change(cpu_id, sample->time, value);
 	return 0;
 }
 #endif /* SUPPORT_OLD_POWER_EVENTS */
-- 
1.8.1.4

[PATCH 30/34] perf timechart: Remove some needless struct forward declarations

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stanislav Fomichev <stfomichev@yandex-team.ru>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-jomi6mjv5zi9vsn4vmih5xps@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-timechart.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index 436cb5f9d751..c246f0290623 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -55,14 +55,8 @@ static bool		tasks_only;
 static bool		with_backtrace;
 
 
-struct per_pid;
 struct per_pidcomm;
-
 struct cpu_sample;
-struct power_event;
-struct wake_event;
-
-struct sample_wrapper;
 
 /*
  * Datastructure layout:
@@ -155,7 +149,6 @@ struct wake_event {
 static struct power_event    *power_events;
 static struct wake_event     *wake_events;
 
-struct process_filter;
 struct process_filter {
 	char			*name;
 	int			pid;
-- 
1.8.1.4

[PATCH 31/34] perf timechart: Remove misplaced __maybe_unused

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

The 'event' parameter _is_ used.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stanislav Fomichev <stfomichev@yandex-team.ru>
Cc: Stephane Eranian <eranian@google.com>
echo Link: http://lkml.kernel.org/n/tip-`ranpwd -l 24`@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-timechart.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
index c246f0290623..680632d7e26a 100644
--- a/tools/perf/builtin-timechart.c
+++ b/tools/perf/builtin-timechart.c
@@ -511,7 +511,7 @@ typedef int (*tracepoint_handler)(struct perf_evsel *evsel,
 				  const char *backtrace);
 
 static int process_sample_event(struct perf_tool *tool __maybe_unused,
-				union perf_event *event __maybe_unused,
+				union perf_event *event,
 				struct perf_sample *sample,
 				struct perf_evsel *evsel,
 				struct machine *machine __maybe_unused)
-- 
1.8.1.4

[PATCH 32/34] perf tools: Fix tags/TAGS targets rebuilding

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@redhat.com>

Once the tags/TAGS file is generated it's never rebuilt until it's
removed by hand.

The reason is that the Makefile does not treat tags/TAGS as targets but
as files and thus won't rebuilt them once they are in place.

Adding PHONY tags/TAGS targets into Makefile.

Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/20131126125412.GJ1267@krava.brq.redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Makefile | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 4835618a5608..eefb9fb0c02f 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -60,8 +60,11 @@ endef
 
 #
 # Needed if no target specified:
+# (Except for tags and TAGS targets. The reason is that the
+# Makefile does not treat tags/TAGS as targets but as files
+# and thus won't rebuilt them once they are in place.)
 #
-all:
+all tags TAGS:
 	$(print_msg)
 	$(make)
 
@@ -77,3 +80,5 @@ clean:
 %:
 	$(print_msg)
 	$(make)
+
+.PHONY: tags TAGS
-- 
1.8.1.4

[PATCH 33/34] perf tools: Add per-feature check flags

[Arnaldo Carvalho de Melo]

From: Jean Pihet <jean.pihet@linaro.org>

Add CFLAGS and LDFLAGS for each feature to be checked. This allows
to pass flags and parameters to the feature checks compilation.
Also simplifies the feature check makefile, to come in a subsequent patch.
Signed-off-by: Jean Pihet <jean.pihet@linaro.org>

Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: linaro-kernel@lists.linaro.org
Cc: patches@linaro.org
Link: http://lkml.kernel.org/r/1385045539-24537-2-git-send-email-jean.pihet@linaro.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/config/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
index f7d11a811c74..c55149538d92 100644
--- a/tools/perf/config/Makefile
+++ b/tools/perf/config/Makefile
@@ -102,7 +102,7 @@ endif
 
 feature_check = $(eval $(feature_check_code))
 define feature_check_code
-  feature-$(1) := $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) CFLAGS="$(EXTRA_CFLAGS)" LDFLAGS="$(LDFLAGS)" LIBUNWIND_LIBS="$(LIBUNWIND_LIBS)" -C config/feature-checks test-$1 >/dev/null 2>/dev/null && echo 1 || echo 0)
+  feature-$(1) := $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) CFLAGS="$(EXTRA_CFLAGS) $(FEATURE_CHECK_CFLAGS-$(1))" LDFLAGS="$(LDFLAGS) $(FEATURE_CHECK_LDFLAGS-$(1))" LIBUNWIND_LIBS="$(LIBUNWIND_LIBS)" -C config/feature-checks test-$1 >/dev/null 2>/dev/null && echo 1 || echo 0)
 endef
 
 feature_set = $(eval $(feature_set_code))
-- 
1.8.1.4

[PATCH 34/34] perf tools unwinding: Use the per-feature check flags

[Arnaldo Carvalho de Melo]

From: Jean Pihet <jean.pihet@linaro.org>

Use the per-feature check flags for the unwinding feature in order to
correctly compile the libunwind and libunwind-debug-frame feature
checks.

Tested on ARMv7 and ARMv8 with 'make DEBUG=1 LIBUNWIND_DIR=/usr/local -C
tools/perf'

Signed-off-by: Jean Pihet <jean.pihet@linaro.org>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: linaro-kernel@lists.linaro.org
Cc: patches@linaro.org
Link: http://lkml.kernel.org/r/1385045539-24537-3-git-send-email-jean.pihet@linaro.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/config/Makefile                | 38 +++++++++++++++++--------------
 tools/perf/config/feature-checks/Makefile |  6 ++---
 2 files changed, 24 insertions(+), 20 deletions(-)

diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
index c55149538d92..80ea6d823908 100644
--- a/tools/perf/config/Makefile
+++ b/tools/perf/config/Makefile
@@ -36,6 +36,26 @@ ifeq ($(ARCH),arm)
   LIBUNWIND_LIBS = -lunwind -lunwind-arm
 endif
 
+ifeq ($(LIBUNWIND_LIBS),)
+  NO_LIBUNWIND := 1
+else
+  #
+  # For linking with debug library, run like:
+  #
+  #   make DEBUG=1 LIBUNWIND_DIR=/opt/libunwind/
+  #
+  ifdef LIBUNWIND_DIR
+    LIBUNWIND_CFLAGS  = -I$(LIBUNWIND_DIR)/include
+    LIBUNWIND_LDFLAGS = -L$(LIBUNWIND_DIR)/lib
+  endif
+  LIBUNWIND_LDFLAGS += $(LIBUNWIND_LIBS)
+
+  FEATURE_CHECK_CFLAGS-libunwind = $(LIBUNWIND_CFLAGS)
+  FEATURE_CHECK_LDFLAGS-libunwind = $(LIBUNWIND_LDFLAGS)
+  FEATURE_CHECK_CFLAGS-libunwind-debug-frame = $(LIBUNWIND_CFLAGS)
+  FEATURE_CHECK_LDFLAGS-libunwind-debug-frame = $(LIBUNWIND_LDFLAGS)
+endif
+
 ifeq ($(NO_PERF_REGS),0)
   CFLAGS += -DHAVE_PERF_REGS_SUPPORT
 endif
@@ -102,7 +122,7 @@ endif
 
 feature_check = $(eval $(feature_check_code))
 define feature_check_code
-  feature-$(1) := $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) CFLAGS="$(EXTRA_CFLAGS) $(FEATURE_CHECK_CFLAGS-$(1))" LDFLAGS="$(LDFLAGS) $(FEATURE_CHECK_LDFLAGS-$(1))" LIBUNWIND_LIBS="$(LIBUNWIND_LIBS)" -C config/feature-checks test-$1 >/dev/null 2>/dev/null && echo 1 || echo 0)
+  feature-$(1) := $(shell $(MAKE) OUTPUT=$(OUTPUT_FEATURES) CFLAGS="$(EXTRA_CFLAGS) $(FEATURE_CHECK_CFLAGS-$(1))" LDFLAGS="$(LDFLAGS) $(FEATURE_CHECK_LDFLAGS-$(1))" -C config/feature-checks test-$1 >/dev/null 2>/dev/null && echo 1 || echo 0)
 endef
 
 feature_set = $(eval $(feature_set_code))
@@ -310,21 +330,7 @@ ifndef NO_LIBELF
   endif # NO_DWARF
 endif # NO_LIBELF
 
-ifeq ($(LIBUNWIND_LIBS),)
-  NO_LIBUNWIND := 1
-endif
-
 ifndef NO_LIBUNWIND
-  #
-  # For linking with debug library, run like:
-  #
-  #   make DEBUG=1 LIBUNWIND_DIR=/opt/libunwind/
-  #
-  ifdef LIBUNWIND_DIR
-    LIBUNWIND_CFLAGS  := -I$(LIBUNWIND_DIR)/include
-    LIBUNWIND_LDFLAGS := -L$(LIBUNWIND_DIR)/lib
-  endif
-
   ifneq ($(feature-libunwind), 1)
     msg := $(warning No libunwind found, disabling post unwind support. Please install libunwind-dev[el] >= 1.1);
     NO_LIBUNWIND := 1
@@ -340,9 +346,7 @@ ifndef NO_LIBUNWIND
       CFLAGS += -DNO_LIBUNWIND_DEBUG_FRAME
     endif
   endif
-endif
 
-ifndef NO_LIBUNWIND
   CFLAGS += -DHAVE_LIBUNWIND_SUPPORT
   EXTLIBS += $(LIBUNWIND_LIBS)
   CFLAGS += $(LIBUNWIND_CFLAGS)
diff --git a/tools/perf/config/feature-checks/Makefile b/tools/perf/config/feature-checks/Makefile
index 87e790017c69..8dffb628dac8 100644
--- a/tools/perf/config/feature-checks/Makefile
+++ b/tools/perf/config/feature-checks/Makefile
@@ -38,7 +38,7 @@ BUILD = $(CC) $(CFLAGS) $(LDFLAGS) -o $(OUTPUT)$@ $@.c
 ###############################
 
 test-all:
-	$(BUILD) -Werror -fstack-protector -fstack-protector-all -O2 -Werror -D_FORTIFY_SOURCE=2 -ldw -lelf -lnuma $(LIBUNWIND_LIBS) -lelf -laudit -I/usr/include/slang -lslang $(shell pkg-config --libs --cflags gtk+-2.0 2>/dev/null) $(FLAGS_PERL_EMBED) $(FLAGS_PYTHON_EMBED) -DPACKAGE='"perf"' -lbfd -ldl
+	$(BUILD) -Werror -fstack-protector -fstack-protector-all -O2 -Werror -D_FORTIFY_SOURCE=2 -ldw -lelf -lnuma -lelf -laudit -I/usr/include/slang -lslang $(shell pkg-config --libs --cflags gtk+-2.0 2>/dev/null) $(FLAGS_PERL_EMBED) $(FLAGS_PYTHON_EMBED) -DPACKAGE='"perf"' -lbfd -ldl
 
 test-hello:
 	$(BUILD)
@@ -74,10 +74,10 @@ test-libnuma:
 	$(BUILD) -lnuma
 
 test-libunwind:
-	$(BUILD) $(LIBUNWIND_LIBS) -lelf
+	$(BUILD) -lelf
 
 test-libunwind-debug-frame:
-	$(BUILD) $(LIBUNWIND_LIBS) -lelf
+	$(BUILD) -lelf
 
 test-libaudit:
 	$(BUILD) -laudit
-- 
1.8.1.4

Re: [PATCH 34/34] perf tools unwinding: Use the per-feature check flags

[Jiri Olsa]

On Wed, Nov 27, 2013 at 05:16:34PM -0300, Arnaldo Carvalho de Melo wrote:
> From: Jean Pihet <jean.pihet@linaro.org>
> 
> Use the per-feature check flags for the unwinding feature in order to
> correctly compile the libunwind and libunwind-debug-frame feature
> checks.
> 
> Tested on ARMv7 and ARMv8 with 'make DEBUG=1 LIBUNWIND_DIR=/usr/local -C
> tools/perf'

this one makes my build fail:

[jolsa@krava perf]$ make
  BUILD:   Doing 'make -j4' parallel build

Auto-detecting system features:
...                     backtrace: [ on  ]
...                         dwarf: [ on  ]
...                fortify-source: [ on  ]
...                         glibc: [ on  ]
...                          gtk2: [ on  ]
...                  gtk2-infobar: [ on  ]
...                      libaudit: [ on  ]
...                        libbfd: [ on  ]
...                        libelf: [ on  ]
...             libelf-getphdrnum: [ on  ]
...                   libelf-mmap: [ on  ]
...                       libnuma: [ on  ]
...                       libperl: [ on  ]
...                     libpython: [ on  ]
...             libpython-version: [ on  ]
...                      libslang: [ on  ]
...                     libunwind: [ OFF ]
...                       on-exit: [ on  ]
...                stackprotector: [ on  ]
...            stackprotector-all: [ on  ]
...                       timerfd: [ on  ]

config/Makefile:335: No libunwind found, disabling post unwind support. Please install libunwind-dev[el] >= 1.1

...

  LINK     perf
/bin/ld: cannot find -lunwind
/bin/ld: cannot find -lunwind-x86_64
collect2: error: ld returned 1 exit status
make[1]: *** [perf] Error 1
make[1]: *** Waiting for unfinished jobs....
make: *** [all] Error 2


I haven't checked this one.. will do tomorrow

jirka

Re: [PATCH 34/34] perf tools unwinding: Use the per-feature check flags

[Jiri Olsa]

On Wed, Nov 27, 2013 at 11:43:23PM +0100, Jiri Olsa wrote:
> On Wed, Nov 27, 2013 at 05:16:34PM -0300, Arnaldo Carvalho de Melo wrote:
> > From: Jean Pihet <jean.pihet@linaro.org>
> > 
> > Use the per-feature check flags for the unwinding feature in order to
> > correctly compile the libunwind and libunwind-debug-frame feature
> > checks.
> > 
> > Tested on ARMv7 and ARMv8 with 'make DEBUG=1 LIBUNWIND_DIR=/usr/local -C
> > tools/perf'
> 
> this one makes my build fail:
> 
> [jolsa@krava perf]$ make
>   BUILD:   Doing 'make -j4' parallel build
> 
> Auto-detecting system features:
> ...                     backtrace: [ on  ]
> ...                         dwarf: [ on  ]
> ...                fortify-source: [ on  ]
> ...                         glibc: [ on  ]
> ...                          gtk2: [ on  ]
> ...                  gtk2-infobar: [ on  ]
> ...                      libaudit: [ on  ]
> ...                        libbfd: [ on  ]
> ...                        libelf: [ on  ]
> ...             libelf-getphdrnum: [ on  ]
> ...                   libelf-mmap: [ on  ]
> ...                       libnuma: [ on  ]
> ...                       libperl: [ on  ]
> ...                     libpython: [ on  ]
> ...             libpython-version: [ on  ]
> ...                      libslang: [ on  ]
> ...                     libunwind: [ OFF ]
> ...                       on-exit: [ on  ]
> ...                stackprotector: [ on  ]
> ...            stackprotector-all: [ on  ]
> ...                       timerfd: [ on  ]
> 
> config/Makefile:335: No libunwind found, disabling post unwind support. Please install libunwind-dev[el] >= 1.1
> 
> ...
> 
>   LINK     perf
> /bin/ld: cannot find -lunwind
> /bin/ld: cannot find -lunwind-x86_64
> collect2: error: ld returned 1 exit status
> make[1]: *** [perf] Error 1
> make[1]: *** Waiting for unfinished jobs....
> make: *** [all] Error 2
> 
> 
> I haven't checked this one.. will do tomorrow


we need to plug libunwind flags/libs only if
the $(feature-libunwind) is enabled..

NO_LIBUNWIND - user's decision not to link with libunwind or
               architecture that does not support it

$(feature-libunwind) - if it's actually installed

attached change fixies that for me, feel free to use/merge it

jirka

---
diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
index 80ea6d8..5f531db 100644
--- a/tools/perf/config/Makefile
+++ b/tools/perf/config/Makefile
@@ -345,12 +345,12 @@ ifndef NO_LIBUNWIND
       # non-ARM has no dwarf_find_debug_frame() function:
       CFLAGS += -DNO_LIBUNWIND_DEBUG_FRAME
     endif
-  endif
 
-  CFLAGS += -DHAVE_LIBUNWIND_SUPPORT
-  EXTLIBS += $(LIBUNWIND_LIBS)
-  CFLAGS += $(LIBUNWIND_CFLAGS)
-  LDFLAGS += $(LIBUNWIND_LDFLAGS)
+    CFLAGS += -DHAVE_LIBUNWIND_SUPPORT
+    CFLAGS += $(LIBUNWIND_CFLAGS)
+    EXTLIBS += $(LIBUNWIND_LIBS)
+    LDFLAGS += $(LIBUNWIND_LDFLAGS)
+  endif # ifneq ($(feature-libunwind), 1)
 endif
 
 ifndef NO_LIBAUDIT

Re: [PATCH 34/34] perf tools unwinding: Use the per-feature check flags

[Arnaldo Carvalho de Melo]

Em Thu, Nov 28, 2013 at 10:58:01AM +0100, Jiri Olsa escreveu:
> On Wed, Nov 27, 2013 at 11:43:23PM +0100, Jiri Olsa wrote:
> > On Wed, Nov 27, 2013 at 05:16:34PM -0300, Arnaldo Carvalho de Melo wrote:
> >   LINK     perf
> > /bin/ld: cannot find -lunwind
> > /bin/ld: cannot find -lunwind-x86_64
> > collect2: error: ld returned 1 exit status
> > make[1]: *** [perf] Error 1
> > make[1]: *** Waiting for unfinished jobs....
> > make: *** [all] Error 2

> > I haven't checked this one.. will do tomorrow
 
> we need to plug libunwind flags/libs only if
> the $(feature-libunwind) is enabled..
 
> NO_LIBUNWIND - user's decision not to link with libunwind or
>                architecture that does not support it
 
> $(feature-libunwind) - if it's actually installed
 
> attached change fixies that for me, feel free to use/merge it

Argh, I used tests/make on one machine where those two patches by Jean
were not applied, then rebased on another, the one I use to submit,
those got included but not tests/make tested, which probably explains
why this got thru :-\

Jean, can you please check that this works for you on ARM too?

- Arnaldo

Re: [PATCH 34/34] perf tools unwinding: Use the per-feature check flags

[Arnaldo Carvalho de Melo]

Em Thu, Nov 28, 2013 at 09:56:19AM -0300, Arnaldo Carvalho de Melo escreveu:
> Em Thu, Nov 28, 2013 at 10:58:01AM +0100, Jiri Olsa escreveu:
> > On Wed, Nov 27, 2013 at 11:43:23PM +0100, Jiri Olsa wrote:
> > > On Wed, Nov 27, 2013 at 05:16:34PM -0300, Arnaldo Carvalho de Melo wrote:
> > >   LINK     perf
> > > /bin/ld: cannot find -lunwind
> > > /bin/ld: cannot find -lunwind-x86_64
> > > collect2: error: ld returned 1 exit status
> > > make[1]: *** [perf] Error 1
> > > make[1]: *** Waiting for unfinished jobs....
> > > make: *** [all] Error 2
> 
> > > I haven't checked this one.. will do tomorrow
>  
> > we need to plug libunwind flags/libs only if
> > the $(feature-libunwind) is enabled..
>  
> > NO_LIBUNWIND - user's decision not to link with libunwind or
> >                architecture that does not support it
>  
> > $(feature-libunwind) - if it's actually installed
>  
> > attached change fixies that for me, feel free to use/merge it
> 
> Argh, I used tests/make on one machine where those two patches by Jean
> were not applied, then rebased on another, the one I use to submit,
> those got included but not tests/make tested, which probably explains
> why this got thru :-\
> 
> Jean, can you please check that this works for you on ARM too?

I just noticed that this patch breaks the feature detection mechanism,
after it is applied it is back performing all tests at every make call,
this needs rethinking, so I'm dropping both.

Ingo, please disregard, yet again, my latest pull request, sigh.

Jiri, this could be something for tests/make, till then I'll try
to check this manually.

- Arnaldo

Re: [PATCH 34/34] perf tools unwinding: Use the per-feature check flags

[Jean Pihet]

On 28 November 2013 14:46, Arnaldo Carvalho de Melo <acme@infradead.org> wrote:
> Em Thu, Nov 28, 2013 at 09:56:19AM -0300, Arnaldo Carvalho de Melo escreveu:
>> Em Thu, Nov 28, 2013 at 10:58:01AM +0100, Jiri Olsa escreveu:
>> > On Wed, Nov 27, 2013 at 11:43:23PM +0100, Jiri Olsa wrote:
>> > > On Wed, Nov 27, 2013 at 05:16:34PM -0300, Arnaldo Carvalho de Melo wrote:
>> > >   LINK     perf
>> > > /bin/ld: cannot find -lunwind
>> > > /bin/ld: cannot find -lunwind-x86_64
>> > > collect2: error: ld returned 1 exit status
>> > > make[1]: *** [perf] Error 1
>> > > make[1]: *** Waiting for unfinished jobs....
>> > > make: *** [all] Error 2
>>
>> > > I haven't checked this one.. will do tomorrow
>>
>> > we need to plug libunwind flags/libs only if
>> > the $(feature-libunwind) is enabled..
>>
>> > NO_LIBUNWIND - user's decision not to link with libunwind or
>> >                architecture that does not support it
>>
>> > $(feature-libunwind) - if it's actually installed
>>
>> > attached change fixies that for me, feel free to use/merge it
>>
>> Argh, I used tests/make on one machine where those two patches by Jean
>> were not applied, then rebased on another, the one I use to submit,
>> those got included but not tests/make tested, which probably explains
>> why this got thru :-\
>>
>> Jean, can you please check that this works for you on ARM too?
>
> I just noticed that this patch breaks the feature detection mechanism,
> after it is applied it is back performing all tests at every make call,
> this needs rethinking, so I'm dropping both.
Oh I am sorry about that. I tested on ARM with and without the
LIBUNWIND_DIR option set.
Let me rethink/rework this and come back to you with a proper fix.

>
> Ingo, please disregard, yet again, my latest pull request, sigh.
>
> Jiri, this could be something for tests/make, till then I'll try
> to check this manually.

One question though: are you OK with the principle of having
per-feature check flags? This brings two things to the feature
detection and build:
1. the ability to specify specific flags for the feature check, which
is not possible on the current code,
2. a simplification in the Makefiles.

Jean

>
> - Arnaldo

Re: [PATCH 34/34] perf tools unwinding: Use the per-feature check flags

[Jiri Olsa]

On Thu, Nov 28, 2013 at 09:02:23PM +0100, Jean Pihet wrote:
> On 28 November 2013 14:46, Arnaldo Carvalho de Melo <acme@infradead.org> wrote:
> > Em Thu, Nov 28, 2013 at 09:56:19AM -0300, Arnaldo Carvalho de Melo escreveu:
> >> Em Thu, Nov 28, 2013 at 10:58:01AM +0100, Jiri Olsa escreveu:
> >> > On Wed, Nov 27, 2013 at 11:43:23PM +0100, Jiri Olsa wrote:
> >> > > On Wed, Nov 27, 2013 at 05:16:34PM -0300, Arnaldo Carvalho de Melo wrote:
> >> > >   LINK     perf
> >> > > /bin/ld: cannot find -lunwind
> >> > > /bin/ld: cannot find -lunwind-x86_64
> >> > > collect2: error: ld returned 1 exit status
> >> > > make[1]: *** [perf] Error 1
> >> > > make[1]: *** Waiting for unfinished jobs....
> >> > > make: *** [all] Error 2
> >>
> >> > > I haven't checked this one.. will do tomorrow
> >>
> >> > we need to plug libunwind flags/libs only if
> >> > the $(feature-libunwind) is enabled..
> >>
> >> > NO_LIBUNWIND - user's decision not to link with libunwind or
> >> >                architecture that does not support it
> >>
> >> > $(feature-libunwind) - if it's actually installed
> >>
> >> > attached change fixies that for me, feel free to use/merge it
> >>
> >> Argh, I used tests/make on one machine where those two patches by Jean
> >> were not applied, then rebased on another, the one I use to submit,
> >> those got included but not tests/make tested, which probably explains
> >> why this got thru :-\
> >>
> >> Jean, can you please check that this works for you on ARM too?
> >
> > I just noticed that this patch breaks the feature detection mechanism,
> > after it is applied it is back performing all tests at every make call,
> > this needs rethinking, so I'm dropping both.
> Oh I am sorry about that. I tested on ARM with and without the
> LIBUNWIND_DIR option set.
> Let me rethink/rework this and come back to you with a proper fix.
> 
> >
> > Ingo, please disregard, yet again, my latest pull request, sigh.
> >
> > Jiri, this could be something for tests/make, till then I'll try
> > to check this manually.
> 
> One question though: are you OK with the principle of having
> per-feature check flags? This brings two things to the feature
> detection and build:
> 1. the ability to specify specific flags for the feature check, which
> is not possible on the current code,
> 2. a simplification in the Makefiles.

looks good to me

jirka

Re: [PATCH 34/34] perf tools unwinding: Use the per-feature check flags

[Jean Pihet]

I just resent the patches as v2.

I tested on x86_64, ARMv7 and ARMv8 with and without LIBUNWIND_DIR=
set in 'make -C tools/perf'. Can you check on your build setup?

Thanks,
Jean

On 29 November 2013 16:45, Jiri Olsa <jolsa@redhat.com> wrote:
> On Thu, Nov 28, 2013 at 09:02:23PM +0100, Jean Pihet wrote:
>> On 28 November 2013 14:46, Arnaldo Carvalho de Melo <acme@infradead.org> wrote:
>> > Em Thu, Nov 28, 2013 at 09:56:19AM -0300, Arnaldo Carvalho de Melo escreveu:
>> >> Em Thu, Nov 28, 2013 at 10:58:01AM +0100, Jiri Olsa escreveu:
>> >> > On Wed, Nov 27, 2013 at 11:43:23PM +0100, Jiri Olsa wrote:
>> >> > > On Wed, Nov 27, 2013 at 05:16:34PM -0300, Arnaldo Carvalho de Melo wrote:
>> >> > >   LINK     perf
>> >> > > /bin/ld: cannot find -lunwind
>> >> > > /bin/ld: cannot find -lunwind-x86_64
>> >> > > collect2: error: ld returned 1 exit status
>> >> > > make[1]: *** [perf] Error 1
>> >> > > make[1]: *** Waiting for unfinished jobs....
>> >> > > make: *** [all] Error 2
>> >>
>> >> > > I haven't checked this one.. will do tomorrow
>> >>
>> >> > we need to plug libunwind flags/libs only if
>> >> > the $(feature-libunwind) is enabled..
>> >>
>> >> > NO_LIBUNWIND - user's decision not to link with libunwind or
>> >> >                architecture that does not support it
>> >>
>> >> > $(feature-libunwind) - if it's actually installed
>> >>
>> >> > attached change fixies that for me, feel free to use/merge it
>> >>
>> >> Argh, I used tests/make on one machine where those two patches by Jean
>> >> were not applied, then rebased on another, the one I use to submit,
>> >> those got included but not tests/make tested, which probably explains
>> >> why this got thru :-\
>> >>
>> >> Jean, can you please check that this works for you on ARM too?
>> >
>> > I just noticed that this patch breaks the feature detection mechanism,
>> > after it is applied it is back performing all tests at every make call,
>> > this needs rethinking, so I'm dropping both.
>> Oh I am sorry about that. I tested on ARM with and without the
>> LIBUNWIND_DIR option set.
>> Let me rethink/rework this and come back to you with a proper fix.
>>
>> >
>> > Ingo, please disregard, yet again, my latest pull request, sigh.
>> >
>> > Jiri, this could be something for tests/make, till then I'll try
>> > to check this manually.
>>
>> One question though: are you OK with the principle of having
>> per-feature check flags? This brings two things to the feature
>> detection and build:
>> 1. the ability to specify specific flags for the feature check, which
>> is not possible on the current code,
>> 2. a simplification in the Makefiles.
>
> looks good to me
>
> jirka

Re: [PATCH 34/34] perf tools unwinding: Use the per-feature check flags

[Jean Pihet]

Hi,

On 3 December 2013 17:15, Jean Pihet <jean.pihet@linaro.org> wrote:
> I just resent the patches as v2.
>
> I tested on x86_64, ARMv7 and ARMv8 with and without LIBUNWIND_DIR=
> set in 'make -C tools/perf'. Can you check on your build setup?
I just found that the linker options may not work on some distros (in
my case Ubuntu on ARMv7). The fix is here below.
v3 is on its way.

Can you please check?

diff --git a/tools/perf/config/feature-checks/Makefile
b/tools/perf/config/feature-checks/Makefile
index 87e7900..0bf3af7 100644
--- a/tools/perf/config/feature-checks/Makefile
+++ b/tools/perf/config/feature-checks/Makefile
@@ -33,7 +33,7 @@ CC := $(CC) -MD

 all: $(FILES)

-BUILD = $(CC) $(CFLAGS) $(LDFLAGS) -o $(OUTPUT)$@ $@.c
+BUILD = $(CC) $(CFLAGS) -o $(OUTPUT)$@ $@.c $(LDFLAGS)

 ###############################



Thanks,
Jean

>
> On 29 November 2013 16:45, Jiri Olsa <jolsa@redhat.com> wrote:
>> On Thu, Nov 28, 2013 at 09:02:23PM +0100, Jean Pihet wrote:
>>> On 28 November 2013 14:46, Arnaldo Carvalho de Melo <acme@infradead.org> wrote:
>>> > Em Thu, Nov 28, 2013 at 09:56:19AM -0300, Arnaldo Carvalho de Melo escreveu:
>>> >> Em Thu, Nov 28, 2013 at 10:58:01AM +0100, Jiri Olsa escreveu:
>>> >> > On Wed, Nov 27, 2013 at 11:43:23PM +0100, Jiri Olsa wrote:
>>> >> > > On Wed, Nov 27, 2013 at 05:16:34PM -0300, Arnaldo Carvalho de Melo wrote:
>>> >> > >   LINK     perf
>>> >> > > /bin/ld: cannot find -lunwind
>>> >> > > /bin/ld: cannot find -lunwind-x86_64
>>> >> > > collect2: error: ld returned 1 exit status
>>> >> > > make[1]: *** [perf] Error 1
>>> >> > > make[1]: *** Waiting for unfinished jobs....
>>> >> > > make: *** [all] Error 2
>>> >>
>>> >> > > I haven't checked this one.. will do tomorrow
>>> >>
>>> >> > we need to plug libunwind flags/libs only if
>>> >> > the $(feature-libunwind) is enabled..
>>> >>
>>> >> > NO_LIBUNWIND - user's decision not to link with libunwind or
>>> >> >                architecture that does not support it
>>> >>
>>> >> > $(feature-libunwind) - if it's actually installed
>>> >>
>>> >> > attached change fixies that for me, feel free to use/merge it
>>> >>
>>> >> Argh, I used tests/make on one machine where those two patches by Jean
>>> >> were not applied, then rebased on another, the one I use to submit,
>>> >> those got included but not tests/make tested, which probably explains
>>> >> why this got thru :-\
>>> >>
>>> >> Jean, can you please check that this works for you on ARM too?
>>> >
>>> > I just noticed that this patch breaks the feature detection mechanism,
>>> > after it is applied it is back performing all tests at every make call,
>>> > this needs rethinking, so I'm dropping both.
>>> Oh I am sorry about that. I tested on ARM with and without the
>>> LIBUNWIND_DIR option set.
>>> Let me rethink/rework this and come back to you with a proper fix.
>>>
>>> >
>>> > Ingo, please disregard, yet again, my latest pull request, sigh.
>>> >
>>> > Jiri, this could be something for tests/make, till then I'll try
>>> > to check this manually.
>>>
>>> One question though: are you OK with the principle of having
>>> per-feature check flags? This brings two things to the feature
>>> detection and build:
>>> 1. the ability to specify specific flags for the feature check, which
>>> is not possible on the current code,
>>> 2. a simplification in the Makefiles.
>>
>> looks good to me
>>
>> jirka