[PATCH 1/4] perf ftrace: Show error message when fails to set ftrace files

[PATCH 1/4] perf ftrace: Show error message when fails to set ftrace files

[Namhyung Kim]

It'd be better for debugging to show an error message when it fails to
setup ftrace for some reason.

Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
 tools/perf/builtin-ftrace.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c
index 9e0b35cd0eea..966a94fa8200 100644
--- a/tools/perf/builtin-ftrace.c
+++ b/tools/perf/builtin-ftrace.c
@@ -61,6 +61,7 @@ static int __write_tracing_file(const char *name, const char *val, bool append)
 	int fd, ret = -1;
 	ssize_t size = strlen(val);
 	int flags = O_WRONLY;
+	char errbuf[512];
 
 	file = get_tracing_file(name);
 	if (!file) {
@@ -75,14 +76,16 @@ static int __write_tracing_file(const char *name, const char *val, bool append)
 
 	fd = open(file, flags);
 	if (fd < 0) {
-		pr_debug("cannot open tracing file: %s\n", name);
+		pr_debug("cannot open tracing file: %s: %s\n",
+			 name, str_error_r(errno, errbuf, sizeof(errbuf)));
 		goto out;
 	}
 
 	if (write(fd, val, size) == size)
 		ret = 0;
 	else
-		pr_debug("write '%s' to tracing/%s failed\n", val, name);
+		pr_debug("write '%s' to tracing/%s failed: %s\n",
+			 val, name, str_error_r(errno, errbuf, sizeof(errbuf)));
 
 	close(fd);
 out:
-- 
2.13.0

[PATCH 2/4] perf ftrace: Move setup_pager before opening trace_pipe

[Namhyung Kim]

The perf ftrace fails to reset tracer after finish the recording like
below:

  $ sudo perf ftrace -v hello
  write 'nop' to tracing/current_tracer failed: Device or resource busy
  ...

This is because the trace_pipe file is open in pager process.  Move the
pager setup before opening the file.

Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Fixes: 583359646fde ("perf ftrace: Use pager for displaying result")
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
 tools/perf/builtin-ftrace.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c
index 966a94fa8200..982b98ee639e 100644
--- a/tools/perf/builtin-ftrace.c
+++ b/tools/perf/builtin-ftrace.c
@@ -231,6 +231,8 @@ static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 		goto out_reset;
 	}
 
+	setup_pager();
+
 	trace_file = get_tracing_file("trace_pipe");
 	if (!trace_file) {
 		pr_err("failed to open trace_pipe\n");
@@ -254,8 +256,6 @@ static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 		goto out_close_fd;
 	}
 
-	setup_pager();
-
 	perf_evlist__start_workload(ftrace->evlist);
 
 	while (!done) {
-- 
2.13.0

[tip:perf/core] perf ftrace: Move setup_pager before opening trace_pipe

[tip-bot for Namhyung Kim]

Commit-ID:  29681bc5bb4326c2f9eac5dc68d8fad3e88b4bb5
Gitweb:     http://git.kernel.org/tip/29681bc5bb4326c2f9eac5dc68d8fad3e88b4bb5
Author:     Namhyung Kim <namhyung@kernel.org>
AuthorDate: Sun, 18 Jun 2017 23:23:00 +0900
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 19 Jun 2017 22:05:52 -0300

perf ftrace: Move setup_pager before opening trace_pipe

The 'perf ftrace' command fails to reset tracer after finishing
recording like below:

  $ sudo perf ftrace -v hello
  write 'nop' to tracing/current_tracer failed: Device or resource busy
  ...

This is because the trace_pipe file is open in pager process.  Move the
pager setup to before opening the file.

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: kernel-team@lge.com
Fixes: 583359646fde ("perf ftrace: Use pager for displaying result")
Link: http://lkml.kernel.org/r/20170618142302.25390-2-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-ftrace.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c
index 966a94f..982b98e 100644
--- a/tools/perf/builtin-ftrace.c
+++ b/tools/perf/builtin-ftrace.c
@@ -231,6 +231,8 @@ static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 		goto out_reset;
 	}
 
+	setup_pager();
+
 	trace_file = get_tracing_file("trace_pipe");
 	if (!trace_file) {
 		pr_err("failed to open trace_pipe\n");
@@ -254,8 +256,6 @@ static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 		goto out_close_fd;
 	}
 
-	setup_pager();
-
 	perf_evlist__start_workload(ftrace->evlist);
 
 	while (!done) {

[PATCH 3/4] perf ftrace: Add option for function filtering

[Namhyung Kim]

The -T/--trace-funcs and -N/--notrace-funcs options are to specify
functions to enable/disable tracing dynamically.

The -G/--graph-funcs and -g/--nograph-funcs options are to set filters
for function graph tracer.

For example, to trace fault handling functions only:

  $ sudo perf ftrace -T *fault hello
   0)               |  __do_page_fault() {
   0)               |    handle_mm_fault() {
   0)   2.117 us    |      __handle_mm_fault();
   0)   3.627 us    |    }
   0)   7.811 us    |  }
   0)               |  __do_page_fault() {
   0)               |    handle_mm_fault() {
   0)   2.014 us    |      __handle_mm_fault();
   0)   2.424 us    |    }
   0)   2.951 us    |  }
   ...

To trace all functions executed in __do_page_fault:

  $ sudo perf ftrace -G __do_page_fault hello
   3)               |  __do_page_fault() {
   3)   0.060 us    |    down_read_trylock();
   3)               |    find_vma() {
   3)   0.075 us    |      vmacache_find();
   3)   0.053 us    |      vmacache_update();
   3)   1.246 us    |    }
   3)               |    handle_mm_fault() {
   3)   0.063 us    |      __rcu_read_lock();
   3)   0.056 us    |      mem_cgroup_from_task();
   3)   0.057 us    |      __rcu_read_unlock();
   3)               |      __handle_mm_fault() {
   3)               |        filemap_map_pages() {
   3)   0.058 us    |          __rcu_read_lock();
   3)               |          alloc_set_pte() {
   ...

But don't want to show details in handle_mm_fault:

  $ sudo perf ftrace -G __do_page_fault -g handle_mm_fault hello
   3)               |  __do_page_fault() {
   3)   0.049 us    |    down_read_trylock();
   3)               |    find_vma() {
   3)   0.048 us    |      vmacache_find();
   3)   0.041 us    |      vmacache_update();
   3)   0.680 us    |    }
   3)   0.036 us    |    up_read();
   3)   4.547 us    |  } /* __do_page_fault */
   ...

Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
 tools/perf/Documentation/perf-ftrace.txt |  30 ++++++++
 tools/perf/builtin-ftrace.c              | 117 +++++++++++++++++++++++++++++--
 2 files changed, 141 insertions(+), 6 deletions(-)

diff --git a/tools/perf/Documentation/perf-ftrace.txt b/tools/perf/Documentation/perf-ftrace.txt
index 6e6a8b22c859..78d6126ca485 100644
--- a/tools/perf/Documentation/perf-ftrace.txt
+++ b/tools/perf/Documentation/perf-ftrace.txt
@@ -48,6 +48,36 @@ OPTIONS
 	Ranges of CPUs are specified with -: 0-2.
 	Default is to trace on all online CPUs.
 
+-T::
+--trace-funcs=::
+	Only trace functions given by the argument.  Multiple functions
+	can be given by using this option more than once.  The function
+	argument also can be a glob pattern.  It will be passed to
+	'set_ftrace_filter' in tracefs.
+
+-N::
+--notrace-funcs=::
+	Do not trace functions given by the argument.  Like -T option,
+	this can be used more than once to specify multiple functions
+	(or glob patterns).  It will be passed to 'set_ftrace_notrace'
+	in tracefs.
+
+-G::
+--graph-funcs=::
+	Set graph filter on the given function (or a glob pattern).
+	This is useful for the function_graph tracer only and enables
+	tracing for functions executed from the given function.
+	This can be used more than once to specify multiple functions.
+	It will be passed to 'set_graph_function' in tracefs.
+
+-g::
+--nograph-funcs=::
+	Set graph notrace filter on the given function (or a glob pattern).
+	Like -G option, this is useful for the function_graph tracer only
+	and disables tracing for function executed from the given function.
+	This can be used more than once to specify multiple functions.
+	It will be passed to 'set_graph_notrace' in tracefs.
+
 
 SEE ALSO
 --------
diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c
index 982b98ee639e..3285375ce3c2 100644
--- a/tools/perf/builtin-ftrace.c
+++ b/tools/perf/builtin-ftrace.c
@@ -28,9 +28,18 @@
 #define DEFAULT_TRACER  "function_graph"
 
 struct perf_ftrace {
-	struct perf_evlist *evlist;
-	struct target target;
-	const char *tracer;
+	struct perf_evlist	*evlist;
+	struct target		target;
+	const char		*tracer;
+	struct list_head	filters;
+	struct list_head	notrace;
+	struct list_head	graph_funcs;
+	struct list_head	nograph_funcs;
+};
+
+struct filter_entry {
+	struct list_head	list;
+	char			name[];
 };
 
 static bool done;
@@ -104,6 +113,7 @@ static int append_tracing_file(const char *name, const char *val)
 }
 
 static int reset_tracing_cpu(void);
+static void reset_tracing_filters(void);
 
 static int reset_tracing_files(struct perf_ftrace *ftrace __maybe_unused)
 {
@@ -119,6 +129,7 @@ static int reset_tracing_files(struct perf_ftrace *ftrace __maybe_unused)
 	if (reset_tracing_cpu() < 0)
 		return -1;
 
+	reset_tracing_filters();
 	return 0;
 }
 
@@ -184,6 +195,48 @@ static int reset_tracing_cpu(void)
 	return ret;
 }
 
+static int __set_tracing_filter(const char *filter_file, struct list_head *funcs)
+{
+	struct filter_entry *pos;
+
+	list_for_each_entry(pos, funcs, list) {
+		if (append_tracing_file(filter_file, pos->name) < 0)
+			return -1;
+	}
+
+	return 0;
+}
+
+static int set_tracing_filters(struct perf_ftrace *ftrace)
+{
+	int ret;
+
+	ret = __set_tracing_filter("set_ftrace_filter", &ftrace->filters);
+	if (ret < 0)
+		return ret;
+
+	ret = __set_tracing_filter("set_ftrace_notrace", &ftrace->notrace);
+	if (ret < 0)
+		return ret;
+
+	ret = __set_tracing_filter("set_graph_function", &ftrace->graph_funcs);
+	if (ret < 0)
+		return ret;
+
+	/* old kernels do not have this filter */
+	__set_tracing_filter("set_graph_notrace", &ftrace->nograph_funcs);
+
+	return ret;
+}
+
+static void reset_tracing_filters(void)
+{
+	write_tracing_file("set_ftrace_filter", " ");
+	write_tracing_file("set_ftrace_notrace", " ");
+	write_tracing_file("set_graph_function", " ");
+	write_tracing_file("set_graph_notrace", " ");
+}
+
 static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 {
 	char *trace_file;
@@ -226,6 +279,11 @@ static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 		goto out_reset;
 	}
 
+	if (set_tracing_filters(ftrace) < 0) {
+		pr_err("failed to set tracing filters\n");
+		goto out_reset;
+	}
+
 	if (write_tracing_file("current_tracer", ftrace->tracer) < 0) {
 		pr_err("failed to set current_tracer to %s\n", ftrace->tracer);
 		goto out_reset;
@@ -310,6 +368,32 @@ static int perf_ftrace_config(const char *var, const char *value, void *cb)
 	return -1;
 }
 
+static int parse_filter_func(const struct option *opt, const char *str,
+			     int unset __maybe_unused)
+{
+	struct list_head *head = opt->value;
+	struct filter_entry *entry;
+
+	entry = malloc(sizeof(*entry) + strlen(str) + 1);
+	if (entry == NULL)
+		return -ENOMEM;
+
+	strcpy(entry->name, str);
+	list_add_tail(&entry->list, head);
+
+	return 0;
+}
+
+static void delete_filter_func(struct list_head *head)
+{
+	struct filter_entry *pos, *tmp;
+
+	list_for_each_entry_safe(pos, tmp, head, list) {
+		list_del(&pos->list);
+		free(pos);
+	}
+}
+
 int cmd_ftrace(int argc, const char **argv)
 {
 	int ret;
@@ -333,9 +417,22 @@ int cmd_ftrace(int argc, const char **argv)
 		    "system-wide collection from all CPUs"),
 	OPT_STRING('C', "cpu", &ftrace.target.cpu_list, "cpu",
 		    "list of cpus to monitor"),
+	OPT_CALLBACK('T', "trace-funcs", &ftrace.filters, "func",
+		     "trace given functions only", parse_filter_func),
+	OPT_CALLBACK('N', "notrace-funcs", &ftrace.notrace, "func",
+		     "do not trace given functions", parse_filter_func),
+	OPT_CALLBACK('G', "graph-funcs", &ftrace.graph_funcs, "func",
+		     "Set graph filter on given functions", parse_filter_func),
+	OPT_CALLBACK('g', "nograph-funcs", &ftrace.nograph_funcs, "func",
+		     "Set nograph filter on given functions", parse_filter_func),
 	OPT_END()
 	};
 
+	INIT_LIST_HEAD(&ftrace.filters);
+	INIT_LIST_HEAD(&ftrace.notrace);
+	INIT_LIST_HEAD(&ftrace.graph_funcs);
+	INIT_LIST_HEAD(&ftrace.nograph_funcs);
+
 	ret = perf_config(perf_ftrace_config, &ftrace);
 	if (ret < 0)
 		return -1;
@@ -351,12 +448,14 @@ int cmd_ftrace(int argc, const char **argv)
 
 		target__strerror(&ftrace.target, ret, errbuf, 512);
 		pr_err("%s\n", errbuf);
-		return -EINVAL;
+		goto out_delete_filters;
 	}
 
 	ftrace.evlist = perf_evlist__new();
-	if (ftrace.evlist == NULL)
-		return -ENOMEM;
+	if (ftrace.evlist == NULL) {
+		ret = -ENOMEM;
+		goto out_delete_filters;
+	}
 
 	ret = perf_evlist__create_maps(ftrace.evlist, &ftrace.target);
 	if (ret < 0)
@@ -367,5 +466,11 @@ int cmd_ftrace(int argc, const char **argv)
 out_delete_evlist:
 	perf_evlist__delete(ftrace.evlist);
 
+out_delete_filters:
+	delete_filter_func(&ftrace.filters);
+	delete_filter_func(&ftrace.notrace);
+	delete_filter_func(&ftrace.graph_funcs);
+	delete_filter_func(&ftrace.nograph_funcs);
+
 	return ret;
 }
-- 
2.13.0

[tip:perf/core] perf ftrace: Add option for function filtering

[tip-bot for Namhyung Kim]

Commit-ID:  78b83e8b12b4467540ca501c7c019e9d46051957
Gitweb:     http://git.kernel.org/tip/78b83e8b12b4467540ca501c7c019e9d46051957
Author:     Namhyung Kim <namhyung@kernel.org>
AuthorDate: Sun, 18 Jun 2017 23:23:01 +0900
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 19 Jun 2017 22:05:53 -0300

perf ftrace: Add option for function filtering

The -T/--trace-funcs and -N/--notrace-funcs options are to specify
functions to enable/disable tracing dynamically.

The -G/--graph-funcs and -g/--nograph-funcs options are to set filters
for function graph tracer.

For example, to trace fault handling functions only:

  $ sudo perf ftrace -T *fault hello
   0)               |  __do_page_fault() {
   0)               |    handle_mm_fault() {
   0)   2.117 us    |      __handle_mm_fault();
   0)   3.627 us    |    }
   0)   7.811 us    |  }
   0)               |  __do_page_fault() {
   0)               |    handle_mm_fault() {
   0)   2.014 us    |      __handle_mm_fault();
   0)   2.424 us    |    }
   0)   2.951 us    |  }
   ...

To trace all functions executed in __do_page_fault:

  $ sudo perf ftrace -G __do_page_fault hello
   2)               |  __do_page_fault() {
   3)   0.060 us    |    down_read_trylock();
   3)               |    find_vma() {
   3)   0.075 us    |      vmacache_find();
   3)   0.053 us    |      vmacache_update();
   3)   1.246 us    |    }
   3)               |    handle_mm_fault() {
   3)   0.063 us    |      __rcu_read_lock();
   3)   0.056 us    |      mem_cgroup_from_task();
   3)   0.057 us    |      __rcu_read_unlock();
   3)               |      __handle_mm_fault() {
   3)               |        filemap_map_pages() {
   3)   0.058 us    |          __rcu_read_lock();
   3)               |          alloc_set_pte() {
   ...

But don't want to show details in handle_mm_fault:

  $ sudo perf ftrace -G __do_page_fault -g handle_mm_fault hello
   3)               |  __do_page_fault() {
   3)   0.049 us    |    down_read_trylock();
   3)               |    find_vma() {
   3)   0.048 us    |      vmacache_find();
   3)   0.041 us    |      vmacache_update();
   3)   0.680 us    |    }
   3)   0.036 us    |    up_read();
   3)   4.547 us    |  } /* __do_page_fault */
   ...

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: kernel-team@lge.com
Link: http://lkml.kernel.org/r/20170618142302.25390-3-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-ftrace.txt |  30 ++++++++
 tools/perf/builtin-ftrace.c              | 117 +++++++++++++++++++++++++++++--
 2 files changed, 141 insertions(+), 6 deletions(-)

diff --git a/tools/perf/Documentation/perf-ftrace.txt b/tools/perf/Documentation/perf-ftrace.txt
index 6e6a8b2..78d6126 100644
--- a/tools/perf/Documentation/perf-ftrace.txt
+++ b/tools/perf/Documentation/perf-ftrace.txt
@@ -48,6 +48,36 @@ OPTIONS
 	Ranges of CPUs are specified with -: 0-2.
 	Default is to trace on all online CPUs.
 
+-T::
+--trace-funcs=::
+	Only trace functions given by the argument.  Multiple functions
+	can be given by using this option more than once.  The function
+	argument also can be a glob pattern.  It will be passed to
+	'set_ftrace_filter' in tracefs.
+
+-N::
+--notrace-funcs=::
+	Do not trace functions given by the argument.  Like -T option,
+	this can be used more than once to specify multiple functions
+	(or glob patterns).  It will be passed to 'set_ftrace_notrace'
+	in tracefs.
+
+-G::
+--graph-funcs=::
+	Set graph filter on the given function (or a glob pattern).
+	This is useful for the function_graph tracer only and enables
+	tracing for functions executed from the given function.
+	This can be used more than once to specify multiple functions.
+	It will be passed to 'set_graph_function' in tracefs.
+
+-g::
+--nograph-funcs=::
+	Set graph notrace filter on the given function (or a glob pattern).
+	Like -G option, this is useful for the function_graph tracer only
+	and disables tracing for function executed from the given function.
+	This can be used more than once to specify multiple functions.
+	It will be passed to 'set_graph_notrace' in tracefs.
+
 
 SEE ALSO
 --------
diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c
index 982b98e..3285375 100644
--- a/tools/perf/builtin-ftrace.c
+++ b/tools/perf/builtin-ftrace.c
@@ -28,9 +28,18 @@
 #define DEFAULT_TRACER  "function_graph"
 
 struct perf_ftrace {
-	struct perf_evlist *evlist;
-	struct target target;
-	const char *tracer;
+	struct perf_evlist	*evlist;
+	struct target		target;
+	const char		*tracer;
+	struct list_head	filters;
+	struct list_head	notrace;
+	struct list_head	graph_funcs;
+	struct list_head	nograph_funcs;
+};
+
+struct filter_entry {
+	struct list_head	list;
+	char			name[];
 };
 
 static bool done;
@@ -104,6 +113,7 @@ static int append_tracing_file(const char *name, const char *val)
 }
 
 static int reset_tracing_cpu(void);
+static void reset_tracing_filters(void);
 
 static int reset_tracing_files(struct perf_ftrace *ftrace __maybe_unused)
 {
@@ -119,6 +129,7 @@ static int reset_tracing_files(struct perf_ftrace *ftrace __maybe_unused)
 	if (reset_tracing_cpu() < 0)
 		return -1;
 
+	reset_tracing_filters();
 	return 0;
 }
 
@@ -184,6 +195,48 @@ static int reset_tracing_cpu(void)
 	return ret;
 }
 
+static int __set_tracing_filter(const char *filter_file, struct list_head *funcs)
+{
+	struct filter_entry *pos;
+
+	list_for_each_entry(pos, funcs, list) {
+		if (append_tracing_file(filter_file, pos->name) < 0)
+			return -1;
+	}
+
+	return 0;
+}
+
+static int set_tracing_filters(struct perf_ftrace *ftrace)
+{
+	int ret;
+
+	ret = __set_tracing_filter("set_ftrace_filter", &ftrace->filters);
+	if (ret < 0)
+		return ret;
+
+	ret = __set_tracing_filter("set_ftrace_notrace", &ftrace->notrace);
+	if (ret < 0)
+		return ret;
+
+	ret = __set_tracing_filter("set_graph_function", &ftrace->graph_funcs);
+	if (ret < 0)
+		return ret;
+
+	/* old kernels do not have this filter */
+	__set_tracing_filter("set_graph_notrace", &ftrace->nograph_funcs);
+
+	return ret;
+}
+
+static void reset_tracing_filters(void)
+{
+	write_tracing_file("set_ftrace_filter", " ");
+	write_tracing_file("set_ftrace_notrace", " ");
+	write_tracing_file("set_graph_function", " ");
+	write_tracing_file("set_graph_notrace", " ");
+}
+
 static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 {
 	char *trace_file;
@@ -226,6 +279,11 @@ static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 		goto out_reset;
 	}
 
+	if (set_tracing_filters(ftrace) < 0) {
+		pr_err("failed to set tracing filters\n");
+		goto out_reset;
+	}
+
 	if (write_tracing_file("current_tracer", ftrace->tracer) < 0) {
 		pr_err("failed to set current_tracer to %s\n", ftrace->tracer);
 		goto out_reset;
@@ -310,6 +368,32 @@ static int perf_ftrace_config(const char *var, const char *value, void *cb)
 	return -1;
 }
 
+static int parse_filter_func(const struct option *opt, const char *str,
+			     int unset __maybe_unused)
+{
+	struct list_head *head = opt->value;
+	struct filter_entry *entry;
+
+	entry = malloc(sizeof(*entry) + strlen(str) + 1);
+	if (entry == NULL)
+		return -ENOMEM;
+
+	strcpy(entry->name, str);
+	list_add_tail(&entry->list, head);
+
+	return 0;
+}
+
+static void delete_filter_func(struct list_head *head)
+{
+	struct filter_entry *pos, *tmp;
+
+	list_for_each_entry_safe(pos, tmp, head, list) {
+		list_del(&pos->list);
+		free(pos);
+	}
+}
+
 int cmd_ftrace(int argc, const char **argv)
 {
 	int ret;
@@ -333,9 +417,22 @@ int cmd_ftrace(int argc, const char **argv)
 		    "system-wide collection from all CPUs"),
 	OPT_STRING('C', "cpu", &ftrace.target.cpu_list, "cpu",
 		    "list of cpus to monitor"),
+	OPT_CALLBACK('T', "trace-funcs", &ftrace.filters, "func",
+		     "trace given functions only", parse_filter_func),
+	OPT_CALLBACK('N', "notrace-funcs", &ftrace.notrace, "func",
+		     "do not trace given functions", parse_filter_func),
+	OPT_CALLBACK('G', "graph-funcs", &ftrace.graph_funcs, "func",
+		     "Set graph filter on given functions", parse_filter_func),
+	OPT_CALLBACK('g', "nograph-funcs", &ftrace.nograph_funcs, "func",
+		     "Set nograph filter on given functions", parse_filter_func),
 	OPT_END()
 	};
 
+	INIT_LIST_HEAD(&ftrace.filters);
+	INIT_LIST_HEAD(&ftrace.notrace);
+	INIT_LIST_HEAD(&ftrace.graph_funcs);
+	INIT_LIST_HEAD(&ftrace.nograph_funcs);
+
 	ret = perf_config(perf_ftrace_config, &ftrace);
 	if (ret < 0)
 		return -1;
@@ -351,12 +448,14 @@ int cmd_ftrace(int argc, const char **argv)
 
 		target__strerror(&ftrace.target, ret, errbuf, 512);
 		pr_err("%s\n", errbuf);
-		return -EINVAL;
+		goto out_delete_filters;
 	}
 
 	ftrace.evlist = perf_evlist__new();
-	if (ftrace.evlist == NULL)
-		return -ENOMEM;
+	if (ftrace.evlist == NULL) {
+		ret = -ENOMEM;
+		goto out_delete_filters;
+	}
 
 	ret = perf_evlist__create_maps(ftrace.evlist, &ftrace.target);
 	if (ret < 0)
@@ -367,5 +466,11 @@ int cmd_ftrace(int argc, const char **argv)
 out_delete_evlist:
 	perf_evlist__delete(ftrace.evlist);
 
+out_delete_filters:
+	delete_filter_func(&ftrace.filters);
+	delete_filter_func(&ftrace.notrace);
+	delete_filter_func(&ftrace.graph_funcs);
+	delete_filter_func(&ftrace.nograph_funcs);
+
 	return ret;
 }

[PATCH 4/4] perf ftrace: Add -D option for depth filter

[Namhyung Kim]

The -D/--graph-depth option is to set max graph depth.  The following
example traces max 2-depth of page fault handler.

  $ sudo perf ftrace -G __do_page_fault -D 2 -- hello
   ...
   0)               |  __do_page_fault() {
   0)   0.063 us    |    down_read_trylock();
   0)   0.251 us    |    find_vma();
   0)   5.374 us    |    handle_mm_fault();
   0)   0.054 us    |    up_read();
   0)   7.463 us    |  }
   ...

Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
 tools/perf/Documentation/perf-ftrace.txt |  3 +++
 tools/perf/builtin-ftrace.c              | 31 +++++++++++++++++++++++++++++++
 2 files changed, 34 insertions(+)

diff --git a/tools/perf/Documentation/perf-ftrace.txt b/tools/perf/Documentation/perf-ftrace.txt
index 78d6126ca485..721a447f046e 100644
--- a/tools/perf/Documentation/perf-ftrace.txt
+++ b/tools/perf/Documentation/perf-ftrace.txt
@@ -78,6 +78,9 @@ OPTIONS
 	This can be used more than once to specify multiple functions.
 	It will be passed to 'set_graph_notrace' in tracefs.
 
+-D::
+--graph-depth=::
+	Set max depth for function graph tracer to follow
 
 SEE ALSO
 --------
diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c
index 3285375ce3c2..dd26c62c9893 100644
--- a/tools/perf/builtin-ftrace.c
+++ b/tools/perf/builtin-ftrace.c
@@ -35,6 +35,7 @@ struct perf_ftrace {
 	struct list_head	notrace;
 	struct list_head	graph_funcs;
 	struct list_head	nograph_funcs;
+	int			graph_depth;
 };
 
 struct filter_entry {
@@ -129,6 +130,9 @@ static int reset_tracing_files(struct perf_ftrace *ftrace __maybe_unused)
 	if (reset_tracing_cpu() < 0)
 		return -1;
 
+	if (write_tracing_file("max_graph_depth", "0") < 0)
+		return -1;
+
 	reset_tracing_filters();
 	return 0;
 }
@@ -237,6 +241,26 @@ static void reset_tracing_filters(void)
 	write_tracing_file("set_graph_notrace", " ");
 }
 
+static int set_tracing_depth(struct perf_ftrace *ftrace)
+{
+	char buf[16];
+
+	if (ftrace->graph_depth == 0)
+		return 0;
+
+	if (ftrace->graph_depth < 0) {
+		pr_err("invalid graph depth: %d\n", ftrace->graph_depth);
+		return -1;
+	}
+
+	snprintf(buf, sizeof(buf), "%d", ftrace->graph_depth);
+
+	if (write_tracing_file("max_graph_depth", buf) < 0)
+		return -1;
+
+	return 0;
+}
+
 static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 {
 	char *trace_file;
@@ -284,6 +308,11 @@ static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 		goto out_reset;
 	}
 
+	if (set_tracing_depth(ftrace) < 0) {
+		pr_err("failed to set graph depth\n");
+		goto out_reset;
+	}
+
 	if (write_tracing_file("current_tracer", ftrace->tracer) < 0) {
 		pr_err("failed to set current_tracer to %s\n", ftrace->tracer);
 		goto out_reset;
@@ -425,6 +454,8 @@ int cmd_ftrace(int argc, const char **argv)
 		     "Set graph filter on given functions", parse_filter_func),
 	OPT_CALLBACK('g', "nograph-funcs", &ftrace.nograph_funcs, "func",
 		     "Set nograph filter on given functions", parse_filter_func),
+	OPT_INTEGER('D', "graph-depth", &ftrace.graph_depth,
+		    "Max depth for function graph tracer"),
 	OPT_END()
 	};
 
-- 
2.13.0

[tip:perf/core] perf ftrace: Add -D option for depth filter

[tip-bot for Namhyung Kim]

Commit-ID:  1096c35aa821cc4789a64232a0e210bb87a0e5e8
Gitweb:     http://git.kernel.org/tip/1096c35aa821cc4789a64232a0e210bb87a0e5e8
Author:     Namhyung Kim <namhyung@kernel.org>
AuthorDate: Sun, 18 Jun 2017 23:23:02 +0900
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 19 Jun 2017 22:05:54 -0300

perf ftrace: Add -D option for depth filter

The -D/--graph-depth option is to set max graph depth.  The following
example traces max 2-depth of page fault handler.

  $ sudo perf ftrace -G __do_page_fault -D 2 -- hello
   ...
   0)               |  __do_page_fault() {
   0)   0.063 us    |    down_read_trylock();
   0)   0.251 us    |    find_vma();
   0)   5.374 us    |    handle_mm_fault();
   0)   0.054 us    |    up_read();
   0)   7.463 us    |  }
   ...

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: kernel-team@lge.com
Link: http://lkml.kernel.org/r/20170618142302.25390-4-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-ftrace.txt |  3 +++
 tools/perf/builtin-ftrace.c              | 31 +++++++++++++++++++++++++++++++
 2 files changed, 34 insertions(+)

diff --git a/tools/perf/Documentation/perf-ftrace.txt b/tools/perf/Documentation/perf-ftrace.txt
index 78d6126..721a447 100644
--- a/tools/perf/Documentation/perf-ftrace.txt
+++ b/tools/perf/Documentation/perf-ftrace.txt
@@ -78,6 +78,9 @@ OPTIONS
 	This can be used more than once to specify multiple functions.
 	It will be passed to 'set_graph_notrace' in tracefs.
 
+-D::
+--graph-depth=::
+	Set max depth for function graph tracer to follow
 
 SEE ALSO
 --------
diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c
index 3285375..dd26c62 100644
--- a/tools/perf/builtin-ftrace.c
+++ b/tools/perf/builtin-ftrace.c
@@ -35,6 +35,7 @@ struct perf_ftrace {
 	struct list_head	notrace;
 	struct list_head	graph_funcs;
 	struct list_head	nograph_funcs;
+	int			graph_depth;
 };
 
 struct filter_entry {
@@ -129,6 +130,9 @@ static int reset_tracing_files(struct perf_ftrace *ftrace __maybe_unused)
 	if (reset_tracing_cpu() < 0)
 		return -1;
 
+	if (write_tracing_file("max_graph_depth", "0") < 0)
+		return -1;
+
 	reset_tracing_filters();
 	return 0;
 }
@@ -237,6 +241,26 @@ static void reset_tracing_filters(void)
 	write_tracing_file("set_graph_notrace", " ");
 }
 
+static int set_tracing_depth(struct perf_ftrace *ftrace)
+{
+	char buf[16];
+
+	if (ftrace->graph_depth == 0)
+		return 0;
+
+	if (ftrace->graph_depth < 0) {
+		pr_err("invalid graph depth: %d\n", ftrace->graph_depth);
+		return -1;
+	}
+
+	snprintf(buf, sizeof(buf), "%d", ftrace->graph_depth);
+
+	if (write_tracing_file("max_graph_depth", buf) < 0)
+		return -1;
+
+	return 0;
+}
+
 static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 {
 	char *trace_file;
@@ -284,6 +308,11 @@ static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
 		goto out_reset;
 	}
 
+	if (set_tracing_depth(ftrace) < 0) {
+		pr_err("failed to set graph depth\n");
+		goto out_reset;
+	}
+
 	if (write_tracing_file("current_tracer", ftrace->tracer) < 0) {
 		pr_err("failed to set current_tracer to %s\n", ftrace->tracer);
 		goto out_reset;
@@ -425,6 +454,8 @@ int cmd_ftrace(int argc, const char **argv)
 		     "Set graph filter on given functions", parse_filter_func),
 	OPT_CALLBACK('g', "nograph-funcs", &ftrace.nograph_funcs, "func",
 		     "Set nograph filter on given functions", parse_filter_func),
+	OPT_INTEGER('D', "graph-depth", &ftrace.graph_depth,
+		    "Max depth for function graph tracer"),
 	OPT_END()
 	};
 

Re: [PATCH 1/4] perf ftrace: Show error message when fails to set ftrace files

[Arnaldo Carvalho de Melo]

Em Sun, Jun 18, 2017 at 11:22:59PM +0900, Namhyung Kim escreveu:
> It'd be better for debugging to show an error message when it fails to
> setup ftrace for some reason.

Thanks a lot, tested the series, excellent ftrace subset to support in
'perf ftrace'! :-)

- Arnaldo
 
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Cc: Frederic Weisbecker <fweisbec@gmail.com>
> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
> ---
>  tools/perf/builtin-ftrace.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c
> index 9e0b35cd0eea..966a94fa8200 100644
> --- a/tools/perf/builtin-ftrace.c
> +++ b/tools/perf/builtin-ftrace.c
> @@ -61,6 +61,7 @@ static int __write_tracing_file(const char *name, const char *val, bool append)
>  	int fd, ret = -1;
>  	ssize_t size = strlen(val);
>  	int flags = O_WRONLY;
> +	char errbuf[512];
>  
>  	file = get_tracing_file(name);
>  	if (!file) {
> @@ -75,14 +76,16 @@ static int __write_tracing_file(const char *name, const char *val, bool append)
>  
>  	fd = open(file, flags);
>  	if (fd < 0) {
> -		pr_debug("cannot open tracing file: %s\n", name);
> +		pr_debug("cannot open tracing file: %s: %s\n",
> +			 name, str_error_r(errno, errbuf, sizeof(errbuf)));
>  		goto out;
>  	}
>  
>  	if (write(fd, val, size) == size)
>  		ret = 0;
>  	else
> -		pr_debug("write '%s' to tracing/%s failed\n", val, name);
> +		pr_debug("write '%s' to tracing/%s failed: %s\n",
> +			 val, name, str_error_r(errno, errbuf, sizeof(errbuf)));
>  
>  	close(fd);
>  out:
> -- 
> 2.13.0

[tip:perf/core] perf ftrace: Show error message when fails to set ftrace files

[tip-bot for Namhyung Kim]

Commit-ID:  e7bd9ba20a9ec7024a0566a93c22b9571a48939a
Gitweb:     http://git.kernel.org/tip/e7bd9ba20a9ec7024a0566a93c22b9571a48939a
Author:     Namhyung Kim <namhyung@kernel.org>
AuthorDate: Sun, 18 Jun 2017 23:22:59 +0900
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Mon, 19 Jun 2017 22:05:51 -0300

perf ftrace: Show error message when fails to set ftrace files

It'd be better for debugging to show an error message when it fails to
setup ftrace for some reason.

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: kernel-team@lge.com
Link: http://lkml.kernel.org/r/20170618142302.25390-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-ftrace.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c
index 9e0b35c..966a94f 100644
--- a/tools/perf/builtin-ftrace.c
+++ b/tools/perf/builtin-ftrace.c
@@ -61,6 +61,7 @@ static int __write_tracing_file(const char *name, const char *val, bool append)
 	int fd, ret = -1;
 	ssize_t size = strlen(val);
 	int flags = O_WRONLY;
+	char errbuf[512];
 
 	file = get_tracing_file(name);
 	if (!file) {
@@ -75,14 +76,16 @@ static int __write_tracing_file(const char *name, const char *val, bool append)
 
 	fd = open(file, flags);
 	if (fd < 0) {
-		pr_debug("cannot open tracing file: %s\n", name);
+		pr_debug("cannot open tracing file: %s: %s\n",
+			 name, str_error_r(errno, errbuf, sizeof(errbuf)));
 		goto out;
 	}
 
 	if (write(fd, val, size) == size)
 		ret = 0;
 	else
-		pr_debug("write '%s' to tracing/%s failed\n", val, name);
+		pr_debug("write '%s' to tracing/%s failed: %s\n",
+			 val, name, str_error_r(errno, errbuf, sizeof(errbuf)));
 
 	close(fd);
 out: