Re: [PATCH v9 04/10] tracing: Add 'snapshot' event trigger command

From: Steven Rostedt
Date: Fri Sep 13 2013 - 21:06:03 EST


On Fri, 13 Sep 2013 16:10:19 -0500
Tom Zanussi <tom.zanussi@xxxxxxxxxxxxxxx> wrote:

> On Fri, 2013-09-13 at 16:01 -0400, Steven Rostedt wrote:
> > On Sat, 7 Sep 2013 10:29:00 -0500
> > Tom Zanussi <tom.zanussi@xxxxxxxxxxxxxxx> wrote:
> >
> >
> > > --- a/kernel/trace/trace_events_trigger.c
> > > +++ b/kernel/trace/trace_events_trigger.c
> > > @@ -696,6 +696,74 @@ static struct event_command trigger_traceoff_cmd = {
> > > .get_trigger_ops = onoff_get_trigger_ops,
> > > };
> > >
> > > +static void
> > > +snapshot_trigger(struct event_trigger_data *data)
> > > +{
> > > + tracing_snapshot();
> > > +}
> >
> > If CONFIG_TRACER_SNAPSHOT is not defined, then we should not bother
> > implementing the snapshot trigger. This should be encapsulated around
> > ifdefs.
>
> OK, I guess I was just trying to avoid the ifdef since
> tracing_snapshot() is already ifdef'ed out (but with a WARN_ONCE()) if
> CONFIG_TRACER_SNAPSHOT isn't defined.
>
> I agree though, it would be better to just ignore all the snapshot
> trigger code if that's the case. Same for the stacktrace trigger,
> though as much as I hate to put big ifdefs in the main code...
>

#ifdef's are OK when they surround entire functions and structures. I
don't think ifdef'ing out these will cause #ifdef's within functions.
That's when things start to look ugly.

-- Steve

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/