Re: [PATCH] kselftests: ftrace: limit the executing time by reading from cached trace

[Masami Hiramatsu]

On Tue, 26 Oct 2021 09:15:34 -0400
Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:

> On Tue, 26 Oct 2021 21:13:31 +0900
> Masami Hiramatsu <mhiramat@xxxxxxxxxx> wrote:
> 
> > > > No, since each testcase must be run under the clean state. Would we need to
> > > > recover the settings?  
> > > 
> > > I would at least put it back to the default. If someone runs the tests,
> > > it should at least put it back to what it was at boot. Otherwise,
> > > someone might run the tests, and then wonder why events are being
> > > dropped when they are reading the trace.  
> > 
> > Umm, we may need to have a knob to reset the ftrace options...
> > Can we warn such user that if the ftracetest finds that the current
> > value is not the same what it sets?
> 
> You mean before we set pause-on-trace, make sure that it was cleared?
> 
> That could work too, and then just set everything back to what we expected
> it to be at the start.

No, it costs much higher (to save the default settings and recover)
OK, then just make it recover to the build default setting.