[GIT PULL 00/12] perf/core improvements and fixes

[GIT PULL 00/12] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

Hi Ingo,

	Please consider pulling, way more to process in the next days, with the
patchkit for eBPF looking good, perf stat stuff from Jiri and some new hardware
stuff from Andi.

- Arnaldo

The following changes since commit 75f80859b130a1cc84e59e71295ce2dd51fe1c81:

  perf/x86/intel/pebs: Robustify PEBS buffer drain (2015-08-04 10:17:01 +0200)

are available in the git repository at:

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

for you to fetch changes up to f151f53aa4f54a647353e1935e4c6cef7f094dd4:

  perf tools: Fix build errors with mipsel-linux-uclibc compiler (2015-08-05 16:56:16 -0300)

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

New features:

- Deref sys_enter pointer args with contents from probe:vfs_getname, showing
  pathnames instead of pointers in many syscalls in 'perf trace' (Arnaldo Carvalho de Melo)

- Make 'perf trace' write to stderr by default, just like 'strace' (Milian Woff)

Infrastructure:

- color_vfprintf() fixes (Andi Kleen, Jiri Olsa)

- Allow enabling/disabling PERF_SAMPLE_TIME per event (Kan Liang)

- Fix build errors with mipsel-linux-uclibc compiler (Petri Gynther)

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

----------------------------------------------------------------
Andi Kleen (1):
      perf tools: Do not include escape sequences in color_vfprintf return

Arnaldo Carvalho de Melo (6):
      perf script: No tracepoints? Don't call libtraceevent.
      perf trace: Do not show syscall tracepoint filter in the --no-syscalls case
      perf trace: Remember if the vfs_getname tracepoint/kprobe is in place
      perf trace: Use a constant for the syscall formatting buffer
      perf trace: Deref sys_enter pointer args with contents from probe:vfs_getname
      perf trace: Use vfs_getname syscall arg beautifier in more syscalls

Jiri Olsa (1):
      perf tools: Remove trail argument to color vsprintf

Kan Liang (2):
      perf tools: Per-event time support
      perf tools: Refine parse/config callchain functions

Milian Wolff (1):
      perf trace: Write to stderr by default

Petri Gynther (1):
      perf tools: Fix build errors with mipsel-linux-uclibc compiler

 tools/build/feature/test-glibc.c         |  11 ++
 tools/perf/Documentation/perf-record.txt |   4 +-
 tools/perf/builtin-record.c              |   2 +-
 tools/perf/builtin-script.c              |   3 +-
 tools/perf/builtin-trace.c               | 175 +++++++++++++++++++++++++++----
 tools/perf/util/callchain.c              |  14 +--
 tools/perf/util/callchain.h              |   2 +-
 tools/perf/util/cloexec.h                |   2 +-
 tools/perf/util/color.c                  |  21 +---
 tools/perf/util/color.h                  |   1 -
 tools/perf/util/evsel.c                  |  25 +++--
 tools/perf/util/evsel.h                  |   2 +
 tools/perf/util/parse-events.c           |  12 +++
 tools/perf/util/parse-events.h           |   1 +
 tools/perf/util/parse-events.l           |   1 +
 tools/perf/util/pmu.c                    |   2 +-
 16 files changed, 220 insertions(+), 58 deletions(-)

[PATCH 01/12] perf script: No tracepoints? Don't call libtraceevent.

[Arnaldo Carvalho de Melo]

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

The libtraceevent handler (session->tevent) is only initialized when
there are tracepoints in a perf.data event list, so do not call
pevent_set_function_resolve() in those cases, fixing a segfault.

Reported-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Jiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-xyynkucl5p4bcs13zi4i4b1f@git.kernel.org
Report-link: http://lkml.kernel.org/r/20150803174113.GA20282@krava.redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-script.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index bd31380122f9..7912feb9a024 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -1861,7 +1861,8 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 	else
 		symbol_conf.use_callchain = false;
 
-	if (pevent_set_function_resolver(session->tevent.pevent,
+	if (session->tevent.pevent &&
+	    pevent_set_function_resolver(session->tevent.pevent,
 					 machine__resolve_kernel_addr,
 					 &session->machines.host) < 0) {
 		pr_err("%s: failed to set libtraceevent function resolver\n", __func__);
-- 
2.1.0

[PATCH 02/12] perf trace: Do not show syscall tracepoint filter in the --no-syscalls case

[Arnaldo Carvalho de Melo]

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

We were accessing trace->syscalls.events members even when that struct
wasn't initialized, i.e. --no-syscalls was specified on the command
line, fix it to show that, still in debug mode, when we have an event
qualifier list, i.e. when we actually are doing subset syscall tracing.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Milian Wolff <mail@milianw.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Fixes: 19867b6186f3 ("perf trace: Use event filters for the event qualifier list")
Link: http://lkml.kernel.org/n/tip-7980ym6vujgh3yiai0cqzc88@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 06cfa93c0305..3cfca93309ee 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -2394,9 +2394,10 @@ static int trace__run(struct trace *trace, int argc, const char **argv)
 		err = trace__set_ev_qualifier_filter(trace);
 		if (err < 0)
 			goto out_errno;
-	}
 
-	pr_debug("%s\n", trace->syscalls.events.sys_exit->filter);
+		pr_debug("event qualifier tracepoint filter: %s\n",
+			 trace->syscalls.events.sys_exit->filter);
+	}
 
 	err = perf_evlist__apply_filters(evlist, &evsel);
 	if (err < 0)
-- 
2.1.0

[PATCH 03/12] perf trace: Remember if the vfs_getname tracepoint/kprobe is in place

[Arnaldo Carvalho de Melo]

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

So that we can later decide if we will store where to expand the
pathname once we are handling vfs_getname or if we should instead
just go on and straight away print the pointer.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Milian Wolff <mail@milianw.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-ytxk5s5jpc50wahffmlxgxuw@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 3cfca93309ee..aa1e2888c81d 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -1312,6 +1312,7 @@ struct trace {
 	bool			show_tool_stats;
 	bool			trace_syscalls;
 	bool			force;
+	bool			vfs_getname;
 	int			trace_pgfaults;
 };
 
@@ -2188,19 +2189,20 @@ static int trace__record(struct trace *trace, int argc, const char **argv)
 
 static size_t trace__fprintf_thread_summary(struct trace *trace, FILE *fp);
 
-static void perf_evlist__add_vfs_getname(struct perf_evlist *evlist)
+static bool perf_evlist__add_vfs_getname(struct perf_evlist *evlist)
 {
 	struct perf_evsel *evsel = perf_evsel__newtp("probe", "vfs_getname");
 	if (evsel == NULL)
-		return;
+		return false;
 
 	if (perf_evsel__field(evsel, "pathname") == NULL) {
 		perf_evsel__delete(evsel);
-		return;
+		return false;
 	}
 
 	evsel->handler = trace__vfs_getname;
 	perf_evlist__add(evlist, evsel);
+	return true;
 }
 
 static int perf_evlist__add_pgfault(struct perf_evlist *evlist,
@@ -2330,7 +2332,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv)
 		goto out_error_raw_syscalls;
 
 	if (trace->trace_syscalls)
-		perf_evlist__add_vfs_getname(evlist);
+		trace->vfs_getname = perf_evlist__add_vfs_getname(evlist);
 
 	if ((trace->trace_pgfaults & TRACE_PFMAJ) &&
 	    perf_evlist__add_pgfault(evlist, PERF_COUNT_SW_PAGE_FAULTS_MAJ)) {
-- 
2.1.0

[PATCH 04/12] perf trace: Use a constant for the syscall formatting buffer

[Arnaldo Carvalho de Melo]

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

We were using it as a magic number, 1024, fix that.

Eventually we need to stop doing it per line, and do it per
arg, traversing the args at output time, to avoid the memmove()
calls that will be used in the next cset to replace pointers
present at raw_syscalls:sys_enter time with its contents that
appear at probe:vfs_getname time, before raw_syscalls:sys_exit
time.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Milian Wolff <mail@milianw.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-4sz3wid39egay1pp8qmbur4u@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index aa1e2888c81d..12d6fc0227b1 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -1262,6 +1262,8 @@ fail:
 #define TRACE_PFMAJ		(1 << 0)
 #define TRACE_PFMIN		(1 << 1)
 
+static const size_t trace__entry_str_size = 2048;
+
 struct trace {
 	struct perf_tool	tool;
 	struct {
@@ -1822,7 +1824,7 @@ static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel,
 	args = perf_evsel__sc_tp_ptr(evsel, args, sample);
 
 	if (ttrace->entry_str == NULL) {
-		ttrace->entry_str = malloc(1024);
+		ttrace->entry_str = malloc(trace__entry_str_size);
 		if (!ttrace->entry_str)
 			goto out_put;
 	}
@@ -1832,9 +1834,9 @@ static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel,
 
 	ttrace->entry_time = sample->time;
 	msg = ttrace->entry_str;
-	printed += scnprintf(msg + printed, 1024 - printed, "%s(", sc->name);
+	printed += scnprintf(msg + printed, trace__entry_str_size - printed, "%s(", sc->name);
 
-	printed += syscall__scnprintf_args(sc, msg + printed, 1024 - printed,
+	printed += syscall__scnprintf_args(sc, msg + printed, trace__entry_str_size - printed,
 					   args, trace, thread);
 
 	if (sc->is_exit) {
-- 
2.1.0

[PATCH 05/12] perf trace: Deref sys_enter pointer args with contents from probe:vfs_getname

[Arnaldo Carvalho de Melo]

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

To work like strace and dereference syscall pointer args we need to
insert probes (or tracepoints) right after we copy those bytes from
userspace.

Since we're formatting the syscall args at raw_syscalls:sys_enter time,
we need to have a formatter that just stores the position where, later,
when we get the probe:vfs_getname, we can insert the pointer contents.

Now, if a probe:vfs_getname with this format is in place:

 # perf probe -l
  probe:vfs_getname (on getname_flags:72@/home/git/linux/fs/namei.c with pathname)

That was, in this case, put in place with:

 # perf probe 'vfs_getname=getname_flags:72 pathname=filename:string'
 Added new event:
  probe:vfs_getname    (on getname_flags:72 with pathname=filename:string)

 You can now use it in all perf tools, such as:

	perf record -e probe:vfs_getname -aR sleep 1
 #

Then 'perf trace' will notice that and do the pointer -> contents
expansion:

 # trace -e open touch /tmp/bla
  0.165 (0.010 ms): touch/17752 open(filename: /etc/ld.so.cache, flags: CLOEXEC) = 3
  0.195 (0.011 ms): touch/17752 open(filename: /lib64/libc.so.6, flags: CLOEXEC) = 3
  0.512 (0.012 ms): touch/17752 open(filename: /usr/lib/locale/locale-archive, flags: CLOEXEC) = 3
  0.582 (0.012 ms): touch/17752 open(filename: /tmp/bla, flags: CREAT|NOCTTY|NONBLOCK|WRONLY, mode: 438) = 3
 #

Roughly equivalent to strace's output:

 # strace -rT -e open touch /tmp/bla
  0.000000 open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 3 <0.000039>
  0.000317 open("/lib64/libc.so.6", O_RDONLY|O_CLOEXEC) = 3 <0.000102>
  0.001461 open("/usr/lib/locale/locale-archive", O_RDONLY|O_CLOEXEC) = 3 <0.000072>
  0.000405 open("/tmp/bla", O_WRONLY|O_CREAT|O_NOCTTY|O_NONBLOCK, 0666) = 3 <0.000055>
  0.000641 +++ exited with 0 +++
 #

Now we need to either look for at all syscalls that are marked as
pointers and have some well known names ("filename", "pathname", etc)
and set the arg formatter to the one used for the "open" syscall in this
patch.

This implementation works for syscalls with just a string being copied
from userspace, for matching syscalls with more than one string being
copied via the same probe/trace point (vfs_getname) we need to extend
the vfs_getname probe spec to include the pointer too, but there are
some problems with that in 'perf probe' or the kernel kprobes code, need
to investigate before considering supporting multiple strings per
syscall.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Milian Wolff <mail@milianw.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-xvuwx6nuj8cf389kf9s2ue2s@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 76 +++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 75 insertions(+), 1 deletion(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 12d6fc0227b1..0255dd892662 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -744,6 +744,11 @@ static size_t syscall_arg__scnprintf_access_mode(char *bf, size_t size,
 
 #define SCA_ACCMODE syscall_arg__scnprintf_access_mode
 
+static size_t syscall_arg__scnprintf_filename(char *bf, size_t size,
+					      struct syscall_arg *arg);
+
+#define SCA_FILENAME syscall_arg__scnprintf_filename
+
 static size_t syscall_arg__scnprintf_open_flags(char *bf, size_t size,
 					       struct syscall_arg *arg)
 {
@@ -1088,7 +1093,8 @@ static struct syscall_fmt {
 	{ .name	    = "newfstatat", .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
 	{ .name	    = "open",	    .errmsg = true,
-	  .arg_scnprintf = { [1] = SCA_OPEN_FLAGS, /* flags */ }, },
+	  .arg_scnprintf = { [0] = SCA_FILENAME,   /* filename */
+			     [1] = SCA_OPEN_FLAGS, /* flags */ }, },
 	{ .name	    = "open_by_handle_at", .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
 			     [2] = SCA_OPEN_FLAGS, /* flags */ }, },
@@ -1208,6 +1214,11 @@ static size_t fprintf_duration(unsigned long t, FILE *fp)
 	return printed + fprintf(fp, "): ");
 }
 
+/**
+ * filename.ptr: The filename char pointer that will be vfs_getname'd
+ * filename.entry_str_pos: Where to insert the string translated from
+ *                         filename.ptr by the vfs_getname tracepoint/kprobe.
+ */
 struct thread_trace {
 	u64		  entry_time;
 	u64		  exit_time;
@@ -1216,6 +1227,10 @@ struct thread_trace {
 	unsigned long	  pfmaj, pfmin;
 	char		  *entry_str;
 	double		  runtime_ms;
+        struct {
+		unsigned long ptr;
+		int           entry_str_pos;
+	} filename;
 	struct {
 		int	  max;
 		char	  **table;
@@ -1418,6 +1433,27 @@ static size_t syscall_arg__scnprintf_close_fd(char *bf, size_t size,
 	return printed;
 }
 
+static void thread__set_filename_pos(struct thread *thread, const char *bf,
+				     unsigned long ptr)
+{
+	struct thread_trace *ttrace = thread__priv(thread);
+
+	ttrace->filename.ptr = ptr;
+	ttrace->filename.entry_str_pos = bf - ttrace->entry_str;
+}
+
+static size_t syscall_arg__scnprintf_filename(char *bf, size_t size,
+					      struct syscall_arg *arg)
+{
+	unsigned long ptr = arg->val;
+
+	if (!arg->trace->vfs_getname)
+		return scnprintf(bf, size, "%#x", ptr);
+
+	thread__set_filename_pos(arg->thread, bf, ptr);
+	return 0;
+}
+
 static bool trace__filter_duration(struct trace *trace, double t)
 {
 	return t < (trace->duration_filter * NSEC_PER_MSEC);
@@ -1938,7 +1974,45 @@ static int trace__vfs_getname(struct trace *trace, struct perf_evsel *evsel,
 			      union perf_event *event __maybe_unused,
 			      struct perf_sample *sample)
 {
+	struct thread *thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
+	struct thread_trace *ttrace;
+	size_t filename_len, entry_str_len, to_move;
+	ssize_t remaining_space;
+	char *pos;
+	const char *filename;
+
 	trace->last_vfs_getname = perf_evsel__rawptr(evsel, sample, "pathname");
+
+	if (!thread)
+		goto out;
+
+	ttrace = thread__priv(thread);
+	if (!ttrace)
+		goto out;
+
+	if (!ttrace->filename.ptr)
+		goto out;
+
+	entry_str_len = strlen(ttrace->entry_str);
+	remaining_space = trace__entry_str_size - entry_str_len - 1; /* \0 */
+	if (remaining_space <= 0)
+		goto out;
+
+	filename = trace->last_vfs_getname;
+	filename_len = strlen(filename);
+	if (filename_len > (size_t)remaining_space) {
+		filename += filename_len - remaining_space;
+		filename_len = remaining_space;
+	}
+
+	to_move = entry_str_len - ttrace->filename.entry_str_pos + 1; /* \0 */
+	pos = ttrace->entry_str + ttrace->filename.entry_str_pos;
+	memmove(pos + filename_len, pos, to_move);
+	memcpy(pos, filename, filename_len);
+
+	ttrace->filename.ptr = 0;
+	ttrace->filename.entry_str_pos = 0;
+out:
 	return 0;
 }
 
-- 
2.1.0

[PATCH 06/12] perf trace: Use vfs_getname syscall arg beautifier in more syscalls

[Arnaldo Carvalho de Melo]

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

Those were covered and tested in this cset:

 access, chdir, chmod, chown, chroot, creat, getxattr,
 inotify_add_watch, lchown, lgetxattr, listxattr,
 lsetxattr, mkdir, mkdirat, mknod, rmdir, faccessat,
 newfstatat, openat, readlink, readlinkat, removexattr,
 setxattr, statfs, swapon, swapoff, truncate, unlinkat,
 utime, utimes, utimensat.

E.g.:

  # trace -e statfs,access,mkdir mkdir /tmp/bla
   0.285 (0.020 ms): mkdir/2799 access(filename: /etc/ld.so.preload, mode: R         ) = -1 ENOENT No such file or directory
   1.070 (0.032 ms): mkdir/2799 statfs(pathname: /sys/fs/selinux, buf: 0x7ffeafbdc930) = 0
   1.087 (0.013 ms): mkdir/2799 statfs(pathname: /sys/fs/selinux, buf: 0x7ffeafbdc820) = 0
   1.189 (0.014 ms): mkdir/2799 access(filename: /etc/selinux/config                 ) = 0
   1.905 (0.610 ms): mkdir/2799 mkdir(pathname: /tmp/bla, mode: 511                  ) = 0
  #

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Milian Wolff <mail@milianw.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-wbqtnlktquun3wtpjdz3okul@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

  and an empty message aborts the commit.
---
 tools/perf/builtin-trace.c | 74 ++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 65 insertions(+), 9 deletions(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 0255dd892662..98d423efdaa9 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -976,14 +976,23 @@ static struct syscall_fmt {
 	bool	   hexret;
 } syscall_fmts[] = {
 	{ .name	    = "access",	    .errmsg = true,
-	  .arg_scnprintf = { [1] = SCA_ACCMODE, /* mode */ }, },
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */
+			     [1] = SCA_ACCMODE,  /* mode */ }, },
 	{ .name	    = "arch_prctl", .errmsg = true, .alias = "prctl", },
 	{ .name	    = "brk",	    .hexret = true,
 	  .arg_scnprintf = { [0] = SCA_HEX, /* brk */ }, },
+	{ .name	    = "chdir",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
+	{ .name	    = "chmod",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
+	{ .name	    = "chroot",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
 	{ .name     = "clock_gettime",  .errmsg = true, STRARRAY(0, clk_id, clockid), },
 	{ .name	    = "close",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_CLOSE_FD, /* fd */ }, },
 	{ .name	    = "connect",    .errmsg = true, },
+	{ .name	    = "creat",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
 	{ .name	    = "dup",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
 	{ .name	    = "dup2",	    .errmsg = true,
@@ -994,7 +1003,8 @@ static struct syscall_fmt {
 	{ .name	    = "eventfd2",   .errmsg = true,
 	  .arg_scnprintf = { [1] = SCA_EFD_FLAGS, /* flags */ }, },
 	{ .name	    = "faccessat",  .errmsg = true,
-	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
+			     [1] = SCA_FILENAME, /* filename */ }, },
 	{ .name	    = "fadvise64",  .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
 	{ .name	    = "fallocate",  .errmsg = true,
@@ -1008,7 +1018,8 @@ static struct syscall_fmt {
 	{ .name	    = "fchown",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
 	{ .name	    = "fchownat",   .errmsg = true,
-	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
+			     [1] = SCA_FILENAME, /* filename */ }, },
 	{ .name	    = "fcntl",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
 			     [1] = SCA_STRARRAY, /* cmd */ },
@@ -1023,7 +1034,8 @@ static struct syscall_fmt {
 	{ .name	    = "fstat",	    .errmsg = true, .alias = "newfstat",
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
 	{ .name	    = "fstatat",    .errmsg = true, .alias = "newfstatat",
-	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
+			     [1] = SCA_FILENAME, /* filename */ }, },
 	{ .name	    = "fstatfs",    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
 	{ .name	    = "fsync",    .errmsg = true,
@@ -1040,6 +1052,10 @@ static struct syscall_fmt {
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
 	{ .name	    = "getitimer",  .errmsg = true, STRARRAY(0, which, itimers), },
 	{ .name	    = "getrlimit",  .errmsg = true, STRARRAY(0, resource, rlimit_resources), },
+	{ .name	    = "getxattr",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
+	{ .name	    = "inotify_add_watch",	    .errmsg = true,
+	  .arg_scnprintf = { [1] = SCA_FILENAME, /* pathname */ }, },
 	{ .name	    = "ioctl",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
 #if defined(__i386__) || defined(__x86_64__)
@@ -1054,18 +1070,33 @@ static struct syscall_fmt {
 #endif
 	{ .name	    = "kill",	    .errmsg = true,
 	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
+	{ .name	    = "lchown",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
+	{ .name	    = "lgetxattr",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
 	{ .name	    = "linkat",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
+	{ .name	    = "listxattr",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
 	{ .name	    = "lseek",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
 			     [2] = SCA_STRARRAY, /* whence */ },
 	  .arg_parm	 = { [2] = &strarray__whences, /* whence */ }, },
+	{ .name	    = "lsetxattr",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
 	{ .name	    = "lstat",	    .errmsg = true, .alias = "newlstat", },
+	{ .name	    = "lsxattr",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
 	{ .name     = "madvise",    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_HEX,	 /* start */
 			     [2] = SCA_MADV_BHV, /* behavior */ }, },
+	{ .name	    = "mkdir",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
 	{ .name	    = "mkdirat",    .errmsg = true,
-	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
+			     [1] = SCA_FILENAME, /* pathname */ }, },
+	{ .name	    = "mknod",      .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
 	{ .name	    = "mknodat",    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
 	{ .name	    = "mlock",	    .errmsg = true,
@@ -1091,7 +1122,8 @@ static struct syscall_fmt {
 	{ .name	    = "name_to_handle_at", .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
 	{ .name	    = "newfstatat", .errmsg = true,
-	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
+			     [1] = SCA_FILENAME, /* filename */ }, },
 	{ .name	    = "open",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FILENAME,   /* filename */
 			     [1] = SCA_OPEN_FLAGS, /* flags */ }, },
@@ -1100,6 +1132,7 @@ static struct syscall_fmt {
 			     [2] = SCA_OPEN_FLAGS, /* flags */ }, },
 	{ .name	    = "openat",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
+			     [1] = SCA_FILENAME, /* filename */
 			     [2] = SCA_OPEN_FLAGS, /* flags */ }, },
 	{ .name	    = "perf_event_open", .errmsg = true,
 	  .arg_scnprintf = { [1] = SCA_INT, /* pid */
@@ -1121,8 +1154,11 @@ static struct syscall_fmt {
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
 	{ .name	    = "read",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "readlink",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* path */ }, },
 	{ .name	    = "readlinkat", .errmsg = true,
-	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
+			     [1] = SCA_FILENAME, /* pathname */ }, },
 	{ .name	    = "readv",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
 	{ .name	    = "recvfrom",   .errmsg = true,
@@ -1131,8 +1167,12 @@ static struct syscall_fmt {
 	  .arg_scnprintf = { [3] = SCA_MSG_FLAGS, /* flags */ }, },
 	{ .name	    = "recvmsg",    .errmsg = true,
 	  .arg_scnprintf = { [2] = SCA_MSG_FLAGS, /* flags */ }, },
+	{ .name	    = "removexattr", .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
 	{ .name	    = "renameat",   .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	{ .name	    = "rmdir",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
 	{ .name	    = "rt_sigaction", .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_SIGNUM, /* sig */ }, },
 	{ .name	    = "rt_sigprocmask",  .errmsg = true, STRARRAY(0, how, sighow), },
@@ -1149,6 +1189,8 @@ static struct syscall_fmt {
 	  .arg_scnprintf = { [3] = SCA_MSG_FLAGS, /* flags */ }, },
 	{ .name	    = "setitimer",  .errmsg = true, STRARRAY(0, which, itimers), },
 	{ .name	    = "setrlimit",  .errmsg = true, STRARRAY(0, resource, rlimit_resources), },
+	{ .name	    = "setxattr",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
 	{ .name	    = "shutdown",   .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
 	{ .name	    = "socket",	    .errmsg = true,
@@ -1160,17 +1202,31 @@ static struct syscall_fmt {
 			     [1] = SCA_SK_TYPE, /* type */ },
 	  .arg_parm	 = { [0] = &strarray__socket_families, /* family */ }, },
 	{ .name	    = "stat",	    .errmsg = true, .alias = "newstat", },
+	{ .name	    = "statfs",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
+	{ .name	    = "swapoff",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* specialfile */ }, },
+	{ .name	    = "swapon",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* specialfile */ }, },
 	{ .name	    = "symlinkat",  .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
 	{ .name	    = "tgkill",	    .errmsg = true,
 	  .arg_scnprintf = { [2] = SCA_SIGNUM, /* sig */ }, },
 	{ .name	    = "tkill",	    .errmsg = true,
 	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
+	{ .name	    = "truncate",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* path */ }, },
 	{ .name	    = "uname",	    .errmsg = true, .alias = "newuname", },
 	{ .name	    = "unlinkat",   .errmsg = true,
-	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
+			     [1] = SCA_FILENAME, /* pathname */ }, },
+	{ .name	    = "utime",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
 	{ .name	    = "utimensat",  .errmsg = true,
-	  .arg_scnprintf = { [0] = SCA_FDAT, /* dirfd */ }, },
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dirfd */
+			     [1] = SCA_FILENAME, /* filename */ }, },
+	{ .name	    = "utimes",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
 	{ .name	    = "write",	    .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
 	{ .name	    = "writev",	    .errmsg = true,
-- 
2.1.0

[PATCH 07/12] perf tools: Per-event time support

[Arnaldo Carvalho de Melo]

From: Kan Liang <kan.liang@intel.com>

This patchkit adds the ability to turn off time stamps per event.

One usaful case for partial time is to work with per-event callgraph to
enable "PEBS threshold > 1" (https://lkml.org/lkml/2015/5/10/196), which
can significantly reduce the sampling overhead.

The event samples with time stamps off will not be ordered.

Signed-off-by: Kan Liang <kan.liang@intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1438677022-34296-2-git-send-email-kan.liang@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-record.txt |  4 +++-
 tools/perf/util/evsel.c                  | 14 +++++++++++---
 tools/perf/util/evsel.h                  |  2 ++
 tools/perf/util/parse-events.c           | 12 ++++++++++++
 tools/perf/util/parse-events.h           |  1 +
 tools/perf/util/parse-events.l           |  1 +
 tools/perf/util/pmu.c                    |  2 +-
 7 files changed, 31 insertions(+), 5 deletions(-)

diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index ac41350ca485..0d852d1bc90f 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -49,7 +49,9 @@ OPTIONS
 	  These params can be used to overload default config values per event.
 	  Here is a list of the params.
 	  - 'period': Set event sampling period
-
+	  - 'time': Disable/enable time stamping. Acceptable values are 1 for
+		    enabling time stamping. 0 for disabling time stamping.
+		    The default is 1.
 	  Note: If user explicitly sets options which conflict with the params,
 	  the value set by the params will be overridden.
 
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 7d3acba5a512..7febfe255703 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -587,15 +587,23 @@ perf_evsel__config_callgraph(struct perf_evsel *evsel,
 	}
 }
 
-static void apply_config_terms(struct perf_event_attr *attr __maybe_unused,
-			       struct list_head *config_terms)
+static void apply_config_terms(struct perf_evsel *evsel)
 {
 	struct perf_evsel_config_term *term;
+	struct list_head *config_terms = &evsel->config_terms;
+	struct perf_event_attr *attr = &evsel->attr;
 
 	list_for_each_entry(term, config_terms, list) {
 		switch (term->type) {
 		case PERF_EVSEL__CONFIG_TERM_PERIOD:
 			attr->sample_period = term->val.period;
+			break;
+		case PERF_EVSEL__CONFIG_TERM_TIME:
+			if (term->val.time)
+				perf_evsel__set_sample_bit(evsel, TIME);
+			else
+				perf_evsel__reset_sample_bit(evsel, TIME);
+			break;
 		default:
 			break;
 		}
@@ -798,7 +806,7 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts)
 	 * Apply event specific term settings,
 	 * it overloads any global configuration.
 	 */
-	apply_config_terms(attr, &evsel->config_terms);
+	apply_config_terms(evsel);
 }
 
 static int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index a7d2175358b8..6a129081f3ad 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -39,6 +39,7 @@ struct cgroup_sel;
 */
 enum {
 	PERF_EVSEL__CONFIG_TERM_PERIOD,
+	PERF_EVSEL__CONFIG_TERM_TIME,
 	PERF_EVSEL__CONFIG_TERM_MAX,
 };
 
@@ -47,6 +48,7 @@ struct perf_evsel_config_term {
 	int	type;
 	union {
 		u64	period;
+		bool	time;
 	} val;
 };
 
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 09bee93fd0ec..a6cb9afc20e2 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -603,6 +603,14 @@ do {									   \
 		 * attr->branch_sample_type = term->val.num;
 		 */
 		break;
+	case PARSE_EVENTS__TERM_TYPE_TIME:
+		CHECK_TYPE_VAL(NUM);
+		if (term->val.num > 1) {
+			err->str = strdup("expected 0 or 1");
+			err->idx = term->err_val;
+			return -EINVAL;
+		}
+		break;
 	case PARSE_EVENTS__TERM_TYPE_NAME:
 		CHECK_TYPE_VAL(STR);
 		break;
@@ -650,6 +658,10 @@ do {								\
 		switch (term->type_term) {
 		case PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD:
 			ADD_CONFIG_TERM(PERIOD, period, term->val.num);
+			break;
+		case PARSE_EVENTS__TERM_TYPE_TIME:
+			ADD_CONFIG_TERM(TIME, time, term->val.num);
+			break;
 		default:
 			break;
 		}
diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h
index 2063048a4354..e6f9aacc1cce 100644
--- a/tools/perf/util/parse-events.h
+++ b/tools/perf/util/parse-events.h
@@ -63,6 +63,7 @@ enum {
 	PARSE_EVENTS__TERM_TYPE_NAME,
 	PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD,
 	PARSE_EVENTS__TERM_TYPE_BRANCH_SAMPLE_TYPE,
+	PARSE_EVENTS__TERM_TYPE_TIME,
 };
 
 struct parse_events_term {
diff --git a/tools/perf/util/parse-events.l b/tools/perf/util/parse-events.l
index 13cef3c65565..f5427505ae77 100644
--- a/tools/perf/util/parse-events.l
+++ b/tools/perf/util/parse-events.l
@@ -183,6 +183,7 @@ config2			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_CONFIG2); }
 name			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_NAME); }
 period			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD); }
 branch_type		{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_BRANCH_SAMPLE_TYPE); }
+time			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_TIME); }
 ,			{ return ','; }
 "/"			{ BEGIN(INITIAL); return '/'; }
 {name_minus}		{ return str(yyscanner, PE_NAME); }
diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
index 7bcb8c315615..b615cdf211d6 100644
--- a/tools/perf/util/pmu.c
+++ b/tools/perf/util/pmu.c
@@ -607,7 +607,7 @@ static char *formats_error_string(struct list_head *formats)
 {
 	struct perf_pmu_format *format;
 	char *err, *str;
-	static const char *static_terms = "config,config1,config2,name,period,branch_type\n";
+	static const char *static_terms = "config,config1,config2,name,period,branch_type,time\n";
 	unsigned i = 0;
 
 	if (!asprintf(&str, "valid terms:"))
-- 
2.1.0

[PATCH 08/12] perf tools: Refine parse/config callchain functions

[Arnaldo Carvalho de Melo]

From: Kan Liang <kan.liang@intel.com>

Pass global callchain_param into parse_callchain_record_opt and
perf_evsel__config_callgraph as parameter. So we can reuse these
functions to parse/config local param for callchain.

Signed-off-by: Kan Liang <kan.liang@intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1438677022-34296-3-git-send-email-kan.liang@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-record.c |  2 +-
 tools/perf/util/callchain.c | 14 +++++++-------
 tools/perf/util/callchain.h |  2 +-
 tools/perf/util/evsel.c     | 11 ++++++-----
 4 files changed, 15 insertions(+), 14 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index f51131b11ad7..25cf6b404e8a 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -779,7 +779,7 @@ int record_parse_callchain_opt(const struct option *opt,
 		return 0;
 	}
 
-	ret = parse_callchain_record_opt(arg);
+	ret = parse_callchain_record_opt(arg, &callchain_param);
 	if (!ret)
 		callchain_debug();
 
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
index 9f643ee77001..931cca8e6ae8 100644
--- a/tools/perf/util/callchain.c
+++ b/tools/perf/util/callchain.c
@@ -53,7 +53,7 @@ static int get_stack_size(const char *str, unsigned long *_size)
 }
 #endif /* HAVE_DWARF_UNWIND_SUPPORT */
 
-int parse_callchain_record_opt(const char *arg)
+int parse_callchain_record_opt(const char *arg, struct callchain_param *param)
 {
 	char *tok, *name, *saveptr = NULL;
 	char *buf;
@@ -73,7 +73,7 @@ int parse_callchain_record_opt(const char *arg)
 		/* Framepointer style */
 		if (!strncmp(name, "fp", sizeof("fp"))) {
 			if (!strtok_r(NULL, ",", &saveptr)) {
-				callchain_param.record_mode = CALLCHAIN_FP;
+				param->record_mode = CALLCHAIN_FP;
 				ret = 0;
 			} else
 				pr_err("callchain: No more arguments "
@@ -86,20 +86,20 @@ int parse_callchain_record_opt(const char *arg)
 			const unsigned long default_stack_dump_size = 8192;
 
 			ret = 0;
-			callchain_param.record_mode = CALLCHAIN_DWARF;
-			callchain_param.dump_size = default_stack_dump_size;
+			param->record_mode = CALLCHAIN_DWARF;
+			param->dump_size = default_stack_dump_size;
 
 			tok = strtok_r(NULL, ",", &saveptr);
 			if (tok) {
 				unsigned long size = 0;
 
 				ret = get_stack_size(tok, &size);
-				callchain_param.dump_size = size;
+				param->dump_size = size;
 			}
 #endif /* HAVE_DWARF_UNWIND_SUPPORT */
 		} else if (!strncmp(name, "lbr", sizeof("lbr"))) {
 			if (!strtok_r(NULL, ",", &saveptr)) {
-				callchain_param.record_mode = CALLCHAIN_LBR;
+				param->record_mode = CALLCHAIN_LBR;
 				ret = 0;
 			} else
 				pr_err("callchain: No more arguments "
@@ -219,7 +219,7 @@ int perf_callchain_config(const char *var, const char *value)
 	var += sizeof("call-graph.") - 1;
 
 	if (!strcmp(var, "record-mode"))
-		return parse_callchain_record_opt(value);
+		return parse_callchain_record_opt(value, &callchain_param);
 #ifdef HAVE_DWARF_UNWIND_SUPPORT
 	if (!strcmp(var, "dump-size")) {
 		unsigned long size = 0;
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
index 679c2c6d8ade..68a32c2fe87a 100644
--- a/tools/perf/util/callchain.h
+++ b/tools/perf/util/callchain.h
@@ -177,7 +177,7 @@ int fill_callchain_info(struct addr_location *al, struct callchain_cursor_node *
 			bool hide_unresolved);
 
 extern const char record_callchain_help[];
-int parse_callchain_record_opt(const char *arg);
+int parse_callchain_record_opt(const char *arg, struct callchain_param *param);
 int parse_callchain_report_opt(const char *arg);
 int perf_callchain_config(const char *var, const char *value);
 
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 7febfe255703..f572f469a30d 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -545,14 +545,15 @@ int perf_evsel__group_desc(struct perf_evsel *evsel, char *buf, size_t size)
 
 static void
 perf_evsel__config_callgraph(struct perf_evsel *evsel,
-			     struct record_opts *opts)
+			     struct record_opts *opts,
+			     struct callchain_param *param)
 {
 	bool function = perf_evsel__is_function_event(evsel);
 	struct perf_event_attr *attr = &evsel->attr;
 
 	perf_evsel__set_sample_bit(evsel, CALLCHAIN);
 
-	if (callchain_param.record_mode == CALLCHAIN_LBR) {
+	if (param->record_mode == CALLCHAIN_LBR) {
 		if (!opts->branch_stack) {
 			if (attr->exclude_user) {
 				pr_warning("LBR callstack option is only available "
@@ -568,12 +569,12 @@ perf_evsel__config_callgraph(struct perf_evsel *evsel,
 				    "Falling back to framepointers.\n");
 	}
 
-	if (callchain_param.record_mode == CALLCHAIN_DWARF) {
+	if (param->record_mode == CALLCHAIN_DWARF) {
 		if (!function) {
 			perf_evsel__set_sample_bit(evsel, REGS_USER);
 			perf_evsel__set_sample_bit(evsel, STACK_USER);
 			attr->sample_regs_user = PERF_REGS_MASK;
-			attr->sample_stack_user = callchain_param.dump_size;
+			attr->sample_stack_user = param->dump_size;
 			attr->exclude_callchain_user = 1;
 		} else {
 			pr_info("Cannot use DWARF unwind for function trace event,"
@@ -714,7 +715,7 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts)
 		evsel->attr.exclude_callchain_user = 1;
 
 	if (callchain_param.enabled && !evsel->no_aux_samples)
-		perf_evsel__config_callgraph(evsel, opts);
+		perf_evsel__config_callgraph(evsel, opts, &callchain_param);
 
 	if (opts->sample_intr_regs) {
 		attr->sample_regs_intr = PERF_REGS_MASK;
-- 
2.1.0

[PATCH 09/12] perf tools: Remove trail argument to color vsprintf

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@redhat.com>

Seems like it's always '\n' through color_fprintf_ln, which is not used
at all, removing.. ;-)

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1438649408-20807-2-git-send-email-andi@firstfloor.org
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/color.c | 16 ++--------------
 tools/perf/util/color.h |  1 -
 2 files changed, 2 insertions(+), 15 deletions(-)

diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
index 55355b3d4f85..ff051d24a0bb 100644
--- a/tools/perf/util/color.c
+++ b/tools/perf/util/color.c
@@ -68,7 +68,7 @@ static int __color_vsnprintf(char *bf, size_t size, const char *color,
 }
 
 static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
-		va_list args, const char *trail)
+		va_list args)
 {
 	int r = 0;
 
@@ -87,8 +87,6 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
 	r += vfprintf(fp, fmt, args);
 	if (perf_use_color_default && *color)
 		r += fprintf(fp, "%s", PERF_COLOR_RESET);
-	if (trail)
-		r += fprintf(fp, "%s", trail);
 	return r;
 }
 
@@ -100,7 +98,7 @@ int color_vsnprintf(char *bf, size_t size, const char *color,
 
 int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args)
 {
-	return __color_vfprintf(fp, color, fmt, args, NULL);
+	return __color_vfprintf(fp, color, fmt, args);
 }
 
 int color_snprintf(char *bf, size_t size, const char *color,
@@ -126,16 +124,6 @@ int color_fprintf(FILE *fp, const char *color, const char *fmt, ...)
 	return r;
 }
 
-int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...)
-{
-	va_list args;
-	int r;
-	va_start(args, fmt);
-	r = __color_vfprintf(fp, color, fmt, args, "\n");
-	va_end(args);
-	return r;
-}
-
 /*
  * This function splits the buffer by newlines and colors the lines individually.
  *
diff --git a/tools/perf/util/color.h b/tools/perf/util/color.h
index 38146f922c54..a93997f16dec 100644
--- a/tools/perf/util/color.h
+++ b/tools/perf/util/color.h
@@ -35,7 +35,6 @@ int color_vsnprintf(char *bf, size_t size, const char *color,
 int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args);
 int color_fprintf(FILE *fp, const char *color, const char *fmt, ...);
 int color_snprintf(char *bf, size_t size, const char *color, const char *fmt, ...);
-int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...);
 int color_fwrite_lines(FILE *fp, const char *color, size_t count, const char *buf);
 int value_color_snprintf(char *bf, size_t size, const char *fmt, double value);
 int percent_color_snprintf(char *bf, size_t size, const char *fmt, ...);
-- 
2.1.0

[PATCH 10/12] perf tools: Do not include escape sequences in color_vfprintf return

[Arnaldo Carvalho de Melo]

From: Andi Kleen <ak@linux.intel.com>

color_vprintf was including the length of the invisible escape sequences
in its return argument. Don't include them to make the return value
usable for indentation calculations.

v2: Add comment, rebase

Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1438649408-20807-3-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/color.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
index ff051d24a0bb..9b9565416f90 100644
--- a/tools/perf/util/color.c
+++ b/tools/perf/util/color.c
@@ -67,6 +67,7 @@ static int __color_vsnprintf(char *bf, size_t size, const char *color,
 	return r;
 }
 
+/* Colors are not included in return value */
 static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
 		va_list args)
 {
@@ -83,10 +84,10 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
 	}
 
 	if (perf_use_color_default && *color)
-		r += fprintf(fp, "%s", color);
+		fprintf(fp, "%s", color);
 	r += vfprintf(fp, fmt, args);
 	if (perf_use_color_default && *color)
-		r += fprintf(fp, "%s", PERF_COLOR_RESET);
+		fprintf(fp, "%s", PERF_COLOR_RESET);
 	return r;
 }
 
-- 
2.1.0

[PATCH 11/12] perf trace: Write to stderr by default

[Arnaldo Carvalho de Melo]

From: Milian Wolff <milian.wolff@kdab.com>

Without this patch, it is cumbersome to read the trace output but
ignoring the normal, potentially verbose, output of the debuggee.  One
common example is doing something like the following:

 perf trace -s find /tmp > /dev/null

Without this patch, the trace summary will be lost. Now, it will still
be printed at the end. This behavior is also applied by strace.

Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: David Ahern <dsahern@gmail.com>
Link: http://lkml.kernel.org/n/tip-tqnks6y2cnvm5f9g2dsfr7zl@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 98d423efdaa9..a47497011c93 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -2965,7 +2965,7 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
 			.mmap_pages    = UINT_MAX,
 			.proc_map_timeout  = 500,
 		},
-		.output = stdout,
+		.output = stderr,
 		.show_comm = true,
 		.trace_syscalls = true,
 	};
-- 
2.1.0

[PATCH 12/12] perf tools: Fix build errors with mipsel-linux-uclibc compiler

[Arnaldo Carvalho de Melo]

From: Petri Gynther <pgynther@google.com>

linux/tools$ make ARCH=mips CROSS_COMPILE=mipsel-linux- perf
...
config/Makefile:256: *** No gnu/libc-version.h found, please install
glibc-dev[el].  Stop.
make[1]: *** [all] Error 2
make: *** [perf] Error 2

...
In file included from builtin-sched.c:13:0:
util/cloexec.h:8:12: error: redundant redeclaration of ‘sched_getcpu’
 [-Werror=redundant-decls]
 extern int sched_getcpu(void) __THROW;

mipsel-buildroot-linux-uclibc/sysroot/usr/include/bits/sched.h:88:12:
 note: previous declaration of ‘sched_getcpu’ was here
 extern int sched_getcpu (void) __THROW;

uclibc info:
sysroot/usr/include/bits/uClibc_config.h
__UCLIBC_MAJOR__ 0
__UCLIBC_MINOR__ 9
__UCLIBC_SUBLEVEL__ 33

sysroot/usr/include/features.h
__UCLIBC__ 1
__GLIBC__ 2
__GLIBC_MINOR__ 2

Signed-off-by: Petri Gynther <pgynther@google.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1438735081-24131-1-git-send-email-pgynther@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/build/feature/test-glibc.c | 11 +++++++++++
 tools/perf/util/cloexec.h        |  2 +-
 2 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/tools/build/feature/test-glibc.c b/tools/build/feature/test-glibc.c
index b0820345cd98..9367f7586676 100644
--- a/tools/build/feature/test-glibc.c
+++ b/tools/build/feature/test-glibc.c
@@ -1,8 +1,19 @@
+#include <stdlib.h>
+
+#if !defined(__UCLIBC__)
 #include <gnu/libc-version.h>
+#else
+#define XSTR(s) STR(s)
+#define STR(s) #s
+#endif
 
 int main(void)
 {
+#if !defined(__UCLIBC__)
 	const char *version = gnu_get_libc_version();
+#else
+	const char *version = XSTR(__GLIBC__) "." XSTR(__GLIBC_MINOR__);
+#endif
 
 	return (long)version;
 }
diff --git a/tools/perf/util/cloexec.h b/tools/perf/util/cloexec.h
index 68888c29b04a..3bee6773ddb0 100644
--- a/tools/perf/util/cloexec.h
+++ b/tools/perf/util/cloexec.h
@@ -4,7 +4,7 @@
 unsigned long perf_event_open_cloexec_flag(void);
 
 #ifdef __GLIBC_PREREQ
-#if !__GLIBC_PREREQ(2, 6)
+#if !__GLIBC_PREREQ(2, 6) && !defined(__UCLIBC__)
 extern int sched_getcpu(void) __THROW;
 #endif
 #endif
-- 
2.1.0