[PATCH] selftests/ftrace: Always enable ftrace_enabled for ftrace testcases

[PATCH] selftests/ftrace: Always enable ftrace_enabled for ftrace testcases

[Xiao Yang]

A lot of ftrace testcases get failure if ftrace_enabled is disabled by default
because ftrace_enabled is a big on/off switch for the whole function tracer.

Signed-off-by: Xiao Yang <yangx.jy@cn.fujitsu.com>
---
 tools/testing/selftests/ftrace/test.d/functions | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/tools/testing/selftests/ftrace/test.d/functions b/tools/testing/selftests/ftrace/test.d/functions
index 5d4550591ff9..54c18275bd7f 100644
--- a/tools/testing/selftests/ftrace/test.d/functions
+++ b/tools/testing/selftests/ftrace/test.d/functions
@@ -1,3 +1,6 @@
+enable_ftrace() { # enable function tracer
+    echo 1 > /proc/sys/kernel/ftrace_enabled
+}
 
 clear_trace() { # reset trace output
     echo > trace
@@ -88,6 +91,7 @@ initialize_ftrace() { # Reset ftrace to initial-state
 # As the initial state, ftrace will be set to nop tracer,
 # no events, no triggers, no filters, no function filters,
 # no probes, and tracing on.
+    enable_ftrace
     disable_tracing
     reset_tracer
     reset_trigger
-- 
2.23.0

Re: [PATCH] selftests/ftrace: Always enable ftrace_enabled for ftrace testcases

[Steven Rostedt]

On Fri, 3 Apr 2020 16:29:29 +0800
Xiao Yang <yangx.jy@cn.fujitsu.com> wrote:

> A lot of ftrace testcases get failure if ftrace_enabled is disabled by default
> because ftrace_enabled is a big on/off switch for the whole function tracer.

Why would it be disabled by default? Anyway, I'm thinking failure is the
right answer if ftrace_enable is off.

-- Steve

Re: [PATCH] selftests/ftrace: Always enable ftrace_enabled for ftrace testcases

[Xiao Yang]

On 2020/4/3 21:22, Steven Rostedt wrote:
> On Fri, 3 Apr 2020 16:29:29 +0800
> Xiao Yang<yangx.jy@cn.fujitsu.com>  wrote:
>
>> A lot of ftrace testcases get failure if ftrace_enabled is disabled by default
>> because ftrace_enabled is a big on/off switch for the whole function tracer.
>
Hi Steve

Thanks for your quick reply.

> Why would it be disabled by default? Anyway, I'm thinking failure is the
> right answer if ftrace_enable is off.

Sorry, it seems better to say that ftrace_enabled is disabled by user.

I got 75 pass and 19 failure if ftrace_enable is off. Just have some 
questions:

1) Is it necessary to run all testcases if ftrace_enable is off?
2) Is is expected to get pass if ftrace_enable is off?
    (I can understand that failure is expected if ftrace_enable is off)

Attach the test log:
------------------------------------------------
[root@Fedora-31 20200403-141956]# cat ftracetest.log
Fri 03 Apr 2020 02:19:56 PM CST
=== Ftrace unit tests ===
[1] Basic trace file check      [PASS]
[2] Basic test for tracers      [PASS]
[3] Basic trace clock test      [PASS]
[4] Basic event tracing check   [PASS]
[5] Change the ringbuffer size  [PASS]
[6] Snapshot and tracing setting        [PASS]
[7] trace_pipe and trace_marker [PASS]
[8] Test ftrace direct functions against tracers        [UNRESOLVED]
[9] Test ftrace direct functions against kprobes        [UNRESOLVED]
[10] Generic dynamic event - add/remove kprobe events   [PASS]
[11] Generic dynamic event - add/remove synthetic events        [PASS]
[12] Generic dynamic event - selective clear (compatibility)    [PASS]
[13] Generic dynamic event - generic clear event        [PASS]
[14] event tracing - enable/disable with event level files      [PASS]
[15] event tracing - restricts events based on pid      [PASS]
[16] event tracing - enable/disable with subsystem level files  [PASS]
[17] event tracing - enable/disable with top level files        [PASS]
[18] Test trace_printk from module      [UNRESOLVED]
[19] ftrace - function graph filters with stack tracer  [FAIL]
[20] ftrace - function graph filters    [FAIL]
[21] ftrace - function trace with cpumask       [PASS]
[22] ftrace - test for function event triggers  [FAIL]
[23] ftrace - function glob filters     [PASS]
[24] ftrace - function pid filters      [FAIL]
[25] ftrace - stacktrace filter command [FAIL]
[26] ftrace - function trace on module  [UNRESOLVED]
[27] ftrace - function profiler with function tracing   [FAIL]
[28] ftrace - function profiling        [PASS]
[29] ftrace - test reading of set_ftrace_filter [PASS]
[30] ftrace - Max stack tracer  [PASS]
[31] ftrace - test for function traceon/off triggers    [FAIL]
[32] ftrace - test tracing error log support    [PASS]
[33] Test creation and deletion of trace instances while setting an 
event       [PASS]
[34] Test creation and deletion of trace instances      [PASS]
[35] Kprobe dynamic event - adding and removing [PASS]
[36] Kprobe dynamic event - busy event check    [PASS]
[37] Kprobe event with comm arguments   [FAIL]
[38] Kprobe event string type argument  [FAIL]
[39] Kprobe event symbol argument       [FAIL]
[40] Kprobe event argument syntax       [PASS]
[41] Kprobe dynamic event with arguments        [FAIL]
[42] Kprobes event arguments with types [FAIL]
[43] Kprobe event user-memory access    [FAIL]
[44] Kprobe event auto/manual naming    [PASS]
[45] Kprobe dynamic event with function tracer  [FAIL]
[46] Kprobe dynamic event - probing module      [UNRESOLVED]
[47] Create/delete multiprobe on kprobe event   [PASS]
[48] Kprobe event parser error log check        [PASS]
[49] Kretprobe dynamic event with arguments     [FAIL]
[50] Kretprobe dynamic event with maxactive     [PASS]
[51] Register/unregister many kprobe events     [PASS]
[52] Kprobe events - probe points       [PASS]
[53] Kprobe dynamic event - adding and removing [PASS]
[54] Uprobe event parser error log check        [PASS]
[55] test for the preemptirqsoff tracer [UNSUPPORTED]
[56] Meta-selftest      [UNRESOLVED]
[57] Test wakeup RT tracer      [PASS]
[58] Test wakeup tracer [PASS]
[59] event trigger - test inter-event histogram trigger expected fail 
actions   [XFAIL]
[60] event trigger - test field variable support        [PASS]
[61] event trigger - test inter-event combined histogram trigger 
[PASS]
[62] event trigger - test multiple actions on hist trigger      [PASS]
[63] event trigger - test inter-event histogram trigger onchange action 
[PASS]
[64] event trigger - test inter-event histogram trigger onmatch action 
[PASS]
[65] event trigger - test inter-event histogram trigger onmatch-onmax 
action    [PASS]
[66] event trigger - test inter-event histogram trigger onmax action 
[PASS]
[67] event trigger - test inter-event histogram trigger snapshot action 
[PASS]
[68] event trigger - test synthetic event create remove [PASS]
[69] event trigger - test synthetic_events syntax parser        [PASS]
[70] event trigger - test inter-event histogram trigger trace action 
[PASS]
[71] event trigger - test event enable/disable trigger  [PASS]
[72] event trigger - test trigger filter        [PASS]
[73] event trigger - test histogram modifiers   [PASS]
[74] event trigger - test histogram parser errors       [PASS]
[75] event trigger - test histogram trigger     [PASS]
[76] event trigger - test multiple histogram triggers   [PASS]
[77] event trigger - test snapshot-trigger      [PASS]
[78] event trigger - test stacktrace-trigger    [PASS]
[79] trace_marker trigger - test histogram trigger      [PASS]
[80] trace_marker trigger - test snapshot trigger       [PASS]
[81] trace_marker trigger - test histogram with synthetic event against 
kernel event    [PASS]
[82] trace_marker trigger - test histogram with synthetic event [PASS]
[83] event trigger - test traceon/off trigger   [PASS]
[84] (instance)  Basic test for tracers [PASS]
[85] (instance)  Basic trace clock test [PASS]
[86] (instance)  Change the ringbuffer size     [PASS]
[87] (instance)  Snapshot and tracing setting   [PASS]
[88] (instance)  trace_pipe and trace_marker    [PASS]
[89] (instance)  event tracing - enable/disable with event level files 
[PASS]
[90] (instance)  event tracing - restricts events based on pid  [PASS]
[91] (instance)  event tracing - enable/disable with subsystem level 
files      [PASS]
[92] (instance)  ftrace - test for function event triggers      [FAIL]
[93] (instance)  ftrace - function pid filters  [FAIL]
[94] (instance)  ftrace - stacktrace filter command     [FAIL]
[95] (instance)  ftrace - test for function traceon/off triggers 
[FAIL]
[96] (instance)  event trigger - test event enable/disable trigger 
[PASS]
[97] (instance)  event trigger - test trigger filter    [PASS]
[98] (instance)  event trigger - test histogram modifiers       [PASS]
[99] (instance)  event trigger - test histogram trigger [PASS]
[100] (instance)  event trigger - test multiple histogram triggers 
[PASS]
[101] (instance)  trace_marker trigger - test histogram trigger [PASS]
[102] (instance)  trace_marker trigger - test snapshot trigger  [PASS]

# of passed:  75
# of failed:  19
# of unresolved:  6
# of untested:  0
# of unsupported:  1
# of xfailed:  1
# of undefined(test bug):  0
------------------------------------------------

Thanks,
Xiao Yang

>
> -- Steve
>
>
> .
>

Re: [PATCH] selftests/ftrace: Always enable ftrace_enabled for ftrace testcases

[Steven Rostedt]

On Fri, 3 Apr 2020 23:01:41 +0800
Xiao Yang <yangx.jy@cn.fujitsu.com> wrote:

> > Why would it be disabled by default? Anyway, I'm thinking failure is the
> > right answer if ftrace_enable is off.  
> 
> Sorry, it seems better to say that ftrace_enabled is disabled by user.

But then why would you run the tests after disabling it?

> 
> I got 75 pass and 19 failure if ftrace_enable is off. Just have some 
> questions:
> 
> 1) Is it necessary to run all testcases if ftrace_enable is off?
> 2) Is is expected to get pass if ftrace_enable is off?
>     (I can understand that failure is expected if ftrace_enable is off)

Because its a way to test the tests!

-- Steve

Re: [PATCH] selftests/ftrace: Always enable ftrace_enabled for ftrace testcases

[Xiao Yang]

On 2020/4/3 23:15, Steven Rostedt wrote:
>>> Why would it be disabled by default? Anyway, I'm thinking failure is the
>>> >  >  right answer if ftrace_enable is off.
>> >
>> >  Sorry, it seems better to say that ftrace_enabled is disabled by user.
> But then why would you run the tests after disabling it?
Hi Steven,

I disabled it accidentally.

>
>> >
>> >  I got 75 pass and 19 failure if ftrace_enable is off. Just have some
>> >  questions:
>> >
>> >  1) Is it necessary to run all testcases if ftrace_enable is off?
>> >  2) Is is expected to get pass if ftrace_enable is off?
>> >       (I can understand that failure is expected if ftrace_enable is off)
> Because its a way to test the tests!

I think that all tests depend on ftrace_enabled and running them should 
report "not supported" or enable it automatically when it is disabled.

Thanks,
Xiao Yang

Re: [PATCH] selftests/ftrace: Always enable ftrace_enabled for ftrace testcases

[Steven Rostedt]

On Fri, 3 Apr 2020 23:36:39 +0800
Xiao Yang <yangx.jy@cn.fujitsu.com> wrote:

> >> >  1) Is it necessary to run all testcases if ftrace_enable is off?
> >> >  2) Is is expected to get pass if ftrace_enable is off?
> >> >       (I can understand that failure is expected if ftrace_enable is off)  
> > Because its a way to test the tests!  
> 
> I think that all tests depend on ftrace_enabled and running them should 
> report "not supported" or enable it automatically when it is disabled.

As I'm considering just removing ftrace_enable, I'm going to keep the tests
as is.

Thanks,

-- Steve

Re: [PATCH] selftests/ftrace: Always enable ftrace_enabled for ftrace testcases

[Xiao Yang]

On 2020/4/4 0:31, Steven Rostedt wrote:
> As I'm considering just removing ftrace_enable, I'm going to keep the tests
> as is.
Hi Steven,

Thanks for your explanation.
It is reasonable for me to keep the tests.

Thanks,
Xiao Yang