21 Jul, 2017
1 commit
-
Replacing prefixcmp(), same purpose, inverted result, so standardize on
the kernel variant, to reduce silly differences among tools/ and the
kernel sources, making it easier for people to work in both codebases.And then doing:
if (strstarts(option, "no-"))
Looks clearer than doing:
if (!prefixcmp(option, "no-"))
To figure out if option starts witn "no-".
Cc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Josh Poimboeuf
Cc: Namhyung Kim
Cc: Rusty Russell
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-kaei42gi7lpa8subwtv7eug8@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
20 Jun, 2017
4 commits
-
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
Tested-by: Arnaldo Carvalho de Melo
Cc: Frederic Weisbecker
Cc: Jiri Olsa
Cc: Masami Hiramatsu
Cc: Peter Zijlstra
Cc: Steven Rostedt
Cc: kernel-team@lge.com
Link: http://lkml.kernel.org/r/20170618142302.25390-4-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
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
Tested-by: Arnaldo Carvalho de Melo
Cc: Frederic Weisbecker
Cc: Jiri Olsa
Cc: Masami Hiramatsu
Cc: Peter Zijlstra
Cc: Steven Rostedt
Cc: kernel-team@lge.com
Link: http://lkml.kernel.org/r/20170618142302.25390-3-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
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
Cc: Frederic Weisbecker
Cc: Jiri Olsa
Cc: Masami Hiramatsu
Cc: Peter Zijlstra
Cc: Steven Rostedt
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 -
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
Cc: Frederic Weisbecker
Cc: Jiri Olsa
Cc: Masami Hiramatsu
Cc: Peter Zijlstra
Cc: Steven Rostedt
Cc: kernel-team@lge.com
Link: http://lkml.kernel.org/r/20170618142302.25390-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
25 Apr, 2017
1 commit
-
Not needed in this header, added to the places that need poll(), wait()
and a few other prototypes.Link: http://lkml.kernel.org/n/tip-i39c7b6xmo1vwd9wxp6fmkl0@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
20 Apr, 2017
2 commits
-
Instead of getting it out of luck from util.h, where it isn't needed at
all.Cc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-0bqugg5lc5ksla1v4m0dnmc1@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Removing it from util.h, part of an effort to disentangle the includes
hell, that makes changes to util.h or something included by it to cause
a complete rebuild of the tools.Cc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-ztrjy52q1rqcchuy3rubfgt2@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
27 Mar, 2017
1 commit
-
We got it from the git sources but never used it for anything, with the
place where this would be somehow used remaining:static int run_builtin(struct cmd_struct *p, int argc, const char **argv)
{
prefix = NULL;
if (p->option & RUN_SETUP)
prefix = NULL; /* setup_perf_directory(); */Ditch it.
Cc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-uw5swz05vol0qpr32c5lpvus@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
04 Mar, 2017
3 commits
-
It's convenient to use the pager when seeing many lines of result.
Note that setup_pager() should be called after perf_evlist__prepare_workload()
since they can interfere each other regarding shared stdio streams.Signed-off-by: Namhyung Kim
Tested-by: Arnaldo Carvalho de Melo
Cc: Frederic Weisbecker
Cc: Jiri Olsa
Cc: Peter Zijlstra
Cc: Steven Rostedt
Cc: kernel-team@lge.com
Link: http://lkml.kernel.org/r/20170224011251.14946-4-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
The -a/--all-cpus and -C/--cpu option is for controlling tracing cpus.
Signed-off-by: Namhyung Kim
Cc: Frederic Weisbecker
Cc: Jiri Olsa
Cc: Peter Zijlstra
Cc: Steven Rostedt
Cc: kernel-team@lge.com
Link: http://lkml.kernel.org/r/20170224011251.14946-3-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
The -p (--pid) option enables to trace existing process by its pid.
Committer notes:
Testing it:
Using the function_graph tracer on a process that is just waiting for user
input and thus will make 'perf ftrace' sit there waiting for that, then press
any key on that mutt session and see what happens:# perf ftrace -t function_graph -p `pidof mutt` | head -40
2) 1.038 us | switch_mm_irqs_off();
------------------------------------------
2) -0 => mutt-3595
------------------------------------------2) | finish_task_switch() {
2) | smp_irq_work_interrupt() {
2) | irq_enter() {
2) 0.180 us | rcu_irq_enter();
2) 1.248 us | }
2) | __wake_up() {
2) 0.126 us | _raw_spin_lock_irqsave();
2) | __wake_up_common() {
2) | pollwake() {
2) | default_wake_function() {
2) | try_to_wake_up() {
2) 0.662 us | _raw_spin_lock_irqsave();
2) | select_task_rq_fair() {
2) 1.719 us | effective_load.isra.41();
2) 1.343 us | effective_load.isra.41();
2) | select_idle_sibling() {
2) 0.331 us | idle_cpu();
2) 1.458 us | }
2) 8.350 us | }
2) 0.200 us | _raw_spin_lock();
2) | ttwu_do_activate() {
2) | activate_task() {
2) 0.136 us | update_rq_clock.part.77();
2) | enqueue_task_fair() {
2) | enqueue_entity() {
2) 0.146 us | update_curr();
2) 0.330 us | account_entity_enqueue();
2) 0.280 us | update_cfs_shares();
2) 0.321 us | place_entity();
2) 0.206 us | __enqueue_entity();
2) 6.926 us | }
2) | enqueue_entity() {
2) 0.105 us | update_curr();
2) 0.175 us | account_entity_enqueue();
2) 0.531 us | update_cfs_shares();
#Signed-off-by: Namhyung Kim
Tested-by: Arnaldo Carvalho de Melo
Cc: Frederic Weisbecker
Cc: Jiri Olsa
Cc: Peter Zijlstra
Cc: Steven Rostedt
Cc: kernel-team@lge.com
Link: http://lkml.kernel.org/r/20170224011251.14946-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
01 Feb, 2017
1 commit
-
Currently 'perf ftrace' command allows selecting 'function_graph' or
'function', defaulting to 'function_graph'.Add ftrace.tracer config option to select the default tracer:
# cat ~/.perfconfig
[ftrace]
tracer = function# perf ftrace usleep 123456 | head -10
-14450 [002] d... 10089.284231: finish_task_switch -14450 [002] .... 10089.284232: finish_wait -14450 [002] .... 10089.284232: mutex_lock -14450 [002] .... 10089.284232: _cond_resched -0 [000] d... 3855.820847: switch_mm_irqs_off -18550 [000] d... 3855.820849: finish_task_switch -18550 [000] d... 3855.820851: smp_irq_work_interrupt -18550 [000] d... 3855.820851: irq_enter -18550 [000] d... 3855.820851: rcu_irq_enter
Tested-by: Arnaldo Carvalho de Melo
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Peter Zijlstra
Cc: Wang Nan
Link: http://lkml.kernel.org/r/1485862711-20216-3-git-send-email-treeze.taeung@gmail.com
[ Added missign space in error message, changed the logic to make it more compact and less error prone ]
Signed-off-by: Arnaldo Carvalho de Melo
27 Jan, 2017
1 commit
-
As a result of commit a3497642c261 ("perf ftrace: Make 'function_graph'
be the default tracer") the ftrace.tracer variable can't be NULL but the
other code setting default tracer remained.Signed-off-by: Taeung Song
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Peter Zijlstra
Cc: Wang Nan
Link: http://lkml.kernel.org/r/1485423339-22780-1-git-send-email-treeze.taeung@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo
26 Jan, 2017
2 commits
-
So that we can suppress the '-t function_graph' and get a more compact command
line:# perf ftrace usleep 123456 | grep raw_spin_lock | sort -k2 -nr | head -5
2) 0.555 us | _raw_spin_lock();
2) 0.516 us | _raw_spin_lock();
2) 0.410 us | _raw_spin_lock_irq();
2) 0.374 us | _raw_spin_lock_irqsave();
#Tested-by: Masami Hiramatsu
Cc: Adrian Hunter
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Jeremy Eder
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Cc: Steven Rostedt
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-ss9xgx5htpxcv86x42pnh3m6@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
The 'perf ftrace' command is a simple wrapper of kernel's ftrace
functionality. It only supports single thread tracing currently and
just reads trace_pipe in text and then write it to stdout.Committer notes:
Testing it:
# perf ftrace -f function_graph usleep 123456
2) | SyS_nanosleep() {
2) | _copy_from_user() {
2) 0.900 us | }
2) 1.354 us | }
2) | hrtimer_nanosleep() {
2) 0.062 us | __hrtimer_init();
2) | do_nanosleep() {
2) | hrtimer_start_range_ns() {
2) 5.025 us | }
2) | schedule() {
2) 0.125 us | rcu_note_context_switch();
2) 0.057 us | _raw_spin_lock();
2) | deactivate_task() {
2) 0.369 us | update_rq_clock.part.77();
2) | dequeue_task_fair() {
2) + 22.453 us | }
2) + 23.736 us | }
2) | pick_next_task_fair() {
2) + 47.167 us | }
2) | pick_next_task_idle() {
2) 4.462 us | }
------------------------------------------
2) usleep-20387 => -0
------------------------------------------2) 0.806 us | switch_mm_irqs_off();
------------------------------------------
2) -0 => usleep-20387
------------------------------------------2) 0.151 us | finish_task_switch();
2) @ 123597.2 us | }
2) 0.037 us | _cond_resched();
2) | hrtimer_try_to_cancel() {
2) 0.064 us | hrtimer_active();
2) 0.353 us | }
2) @ 123605.3 us | }
2) @ 123606.2 us | }
2) @ 123608.3 us | } /* SyS_nanosleep */
2) | __do_page_fault() {
Signed-off-by: Namhyung Kim
Tested-by: Arnaldo Carvalho de Melo
Tested-by: Masami Hiramatsu
Cc: Frederic Weisbecker
Cc: Jeremy Eder
Cc: Jiri Olsa ,
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Cc: Steven Rostedt
Link: http://lkml.kernel.org/n/tip-r1hgmsj4dxny8arn3o9mw512@git.kernel.org
[ Various foward port fixes, add man page ]
Signed-off-by: Arnaldo Carvalho de Melo
