Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Fri, 14 Aug 2020 09:24:40 -0300
From:   Arnaldo Carvalho de Melo <acme@kernel.org>
To:     Changbin Du <changbin.du@gmail.com>
Cc:     Jiri Olsa <jolsa@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>,
        Namhyung Kim <namhyung@kernel.org>,
        Steven Rostedt <rostedt@goodmis.org>,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v8 08/18] perf ftrace: add support for tracing option
 'func_stack_trace'
Message-ID: <20200814122440.GY13995@kernel.org>
References: <20200808023141.14227-1-changbin.du@gmail.com>
 <20200808023141.14227-9-changbin.du@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200808023141.14227-9-changbin.du@gmail.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Em Sat, Aug 08, 2020 at 10:31:31AM +0800, Changbin Du escreveu:
> This adds support to display call trace for function tracer. To do this,
> just specify a '--func-opts call-graph' option.

Applied, thanks for providing the usage example and the output of the
tool, please consider making this more compact and similar to the other
perf tools as:

   # perf ftrace -g -T vfs_read

  # perf report -h -g
  
   Usage: perf report [<options>]
  
      -g, --call-graph <print_type,threshold[,print_limit],order,sort_key[,branch],value>
                            Display call graph (stack chain/backtrace):
  
                                  print_type:     call graph printing style (graph|flat|fractal|folded|none)
                                  threshold:      minimum call graph inclusion threshold (<percent>)
                                  print_limit:    maximum number of call graph entry (<number>)
                                  order:          call graph order (caller|callee)
                                  sort_key:       call graph sort key (function|address)
                                  branch:         include last branch info to call graph (branch)
                                  value:          call graph value (percent|period|count)
  
                                  Default: graph,0.5,caller,function,percent
  
  # 

Is there a way to ask for a limit in the number of levels? That is
similar to /proc/sys/kernel/perf_event_max_stack (system wide) and
perf_event_attr.sample_max_stack (per event)?

One can ask for that per event:

  # perf trace -e sched:*_wakeup/max-stack=2/ -e sched:*switch/max-stack=5/ --max-events=4
     0.000 migration/0/12 sched:sched_switch(prev_comm: "migration/0", prev_pid: 12 (migration/0), prev_prio: 0, prev_state: 1, next_comm: "swapper/0", next_pid: 0, next_prio: 120)
                                       __schedule ([kernel.kallsyms])
                                       __schedule ([kernel.kallsyms])
                                       schedule ([kernel.kallsyms])
                                       smpboot_thread_fn ([kernel.kallsyms])
                                       kthread ([kernel.kallsyms])
     2.381 :0/0 sched:sched_wakeup(comm: "PacerThread", pid: 111637 (PacerThread), prio: 120, success: 1, target_cpu: 0)
                                       ttwu_do_wakeup ([kernel.kallsyms])
                                       ttwu_do_wakeup ([kernel.kallsyms])
     2.387 :0/0 sched:sched_switch(prev_comm: "swapper/0", prev_pid: 0, prev_prio: 120, prev_state: 0, next_comm: "PacerThread", next_pid: 111637 (PacerThread), next_prio: 120)
                                       __schedule ([kernel.kallsyms])
                                       __schedule ([kernel.kallsyms])
                                       schedule_idle ([kernel.kallsyms])
                                       do_idle ([kernel.kallsyms])
                                       cpu_startup_entry ([kernel.kallsyms])
     2.410 PacerThread/111637 sched:sched_switch(prev_comm: "PacerThread", prev_pid: 111637 (PacerThread), prev_prio: 120, prev_state: 1, next_comm: "swapper/0", next_pid: 0, next_prio: 120)
                                       __schedule ([kernel.kallsyms])
                                       __schedule ([kernel.kallsyms])
                                       schedule ([kernel.kallsyms])
                                       futex_wait_queue_me ([kernel.kallsyms])
                                       futex_wait ([kernel.kallsyms])
  #

- Arnaldo
 
> $ sudo perf ftrace -T vfs_read --func-opts call-graph
>  iio-sensor-prox-855   [003]   6168.369657: vfs_read <-ksys_read
>  iio-sensor-prox-855   [003]   6168.369677: <stack trace>
>  => vfs_read
>  => ksys_read
>  => __x64_sys_read
>  => do_syscall_64
>  => entry_SYSCALL_64_after_hwframe
>  ...
> 
> Signed-off-by: Changbin Du <changbin.du@gmail.com>
> 
> ---
> v3: switch to uniform option --func-opts.
> v2: option name '-s' -> '--func-call-graph'
> ---
>  tools/perf/Documentation/perf-ftrace.txt |  4 +++
>  tools/perf/builtin-ftrace.c              | 42 ++++++++++++++++++++++++
>  2 files changed, 46 insertions(+)
> 
> diff --git a/tools/perf/Documentation/perf-ftrace.txt b/tools/perf/Documentation/perf-ftrace.txt
> index c46d0a09b38c..8f08ad0992c2 100644
> --- a/tools/perf/Documentation/perf-ftrace.txt
> +++ b/tools/perf/Documentation/perf-ftrace.txt
> @@ -76,6 +76,10 @@ OPTIONS
>  	specify multiple functions (or glob patterns).  It will be
>  	passed to 'set_ftrace_notrace' in tracefs.
>  
> +--func-opts::
> +	List of options allowed to set:
> +	  call-graph - Display kernel stack trace for function tracer.
> +
>  -G::
>  --graph-funcs=::
>  	Select function_graph tracer and set graph filter on the given
> diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c
> index 07b81d0c1658..469b89748c42 100644
> --- a/tools/perf/builtin-ftrace.c
> +++ b/tools/perf/builtin-ftrace.c
> @@ -27,6 +27,7 @@
>  #include "util/cap.h"
>  #include "util/config.h"
>  #include "util/units.h"
> +#include "util/parse-sublevel-options.h"
>  
>  #define DEFAULT_TRACER  "function_graph"
>  
> @@ -42,6 +43,7 @@ struct perf_ftrace {
>  	int			graph_depth;
>  	unsigned long		percpu_buffer_size;
>  	bool			inherit;
> +	int			func_stack_trace;
>  };
>  
>  struct filter_entry {
> @@ -202,6 +204,7 @@ static void reset_tracing_filters(void);
>  static void reset_tracing_options(struct perf_ftrace *ftrace __maybe_unused)
>  {
>  	write_tracing_option_file("function-fork", "0");
> +	write_tracing_option_file("func_stack_trace", "0");
>  }
>  
>  static int reset_tracing_files(struct perf_ftrace *ftrace __maybe_unused)
> @@ -278,6 +281,17 @@ static int set_tracing_cpu(struct perf_ftrace *ftrace)
>  	return set_tracing_cpumask(cpumap);
>  }
>  
> +static int set_tracing_func_stack_trace(struct perf_ftrace *ftrace)
> +{
> +	if (!ftrace->func_stack_trace)
> +		return 0;
> +
> +	if (write_tracing_option_file("func_stack_trace", "1") < 0)
> +		return -1;
> +
> +	return 0;
> +}
> +
>  static int reset_tracing_cpu(void)
>  {
>  	struct perf_cpu_map *cpumap = perf_cpu_map__new(NULL);
> @@ -426,6 +440,11 @@ static int __cmd_ftrace(struct perf_ftrace *ftrace, int argc, const char **argv)
>  		goto out_reset;
>  	}
>  
> +	if (set_tracing_func_stack_trace(ftrace) < 0) {
> +		pr_err("failed to set tracing option func_stack_trace\n");
> +		goto out_reset;
> +	}
> +
>  	if (set_tracing_filters(ftrace) < 0) {
>  		pr_err("failed to set tracing filters\n");
>  		goto out_reset;
> @@ -598,6 +617,26 @@ static int parse_buffer_size(const struct option *opt,
>  	return -1;
>  }
>  
> +static int parse_func_tracer_opts(const struct option *opt,
> +				  const char *str, int unset)
> +{
> +	int ret;
> +	struct perf_ftrace *ftrace = (struct perf_ftrace *) opt->value;
> +	struct sublevel_option func_tracer_opts[] = {
> +		{ .name = "call-graph",	.value_ptr = &ftrace->func_stack_trace },
> +		{ .name = NULL, }
> +	};
> +
> +	if (unset)
> +		return 0;
> +
> +	ret = perf_parse_sublevel_options(str, func_tracer_opts);
> +	if (ret)
> +		return ret;
> +
> +	return 0;
> +}
> +
>  static void select_tracer(struct perf_ftrace *ftrace)
>  {
>  	bool graph = !list_empty(&ftrace->graph_funcs) ||
> @@ -645,6 +684,9 @@ int cmd_ftrace(int argc, const char **argv)
>  		     parse_filter_func),
>  	OPT_CALLBACK('N', "notrace-funcs", &ftrace.notrace, "func",
>  		     "do not trace given functions", parse_filter_func),
> +	OPT_CALLBACK(0, "func-opts", &ftrace, "options",
> +		     "function tracer options, available options: call-graph",
> +		     parse_func_tracer_opts),
>  	OPT_CALLBACK('G', "graph-funcs", &ftrace.graph_funcs, "func",
>  		     "trace given functions using function_graph tracer",
>  		     parse_filter_func),
> -- 
> 2.25.1
> 

-- 

- Arnaldo