Re: [PATCH 1/3] tracing: Update event filters for multibuffer

[Tom Zanussi <tom.zanussi@linux.intel.com>]

On Tue, 2013-11-05 at 17:27 -0500, Steven Rostedt wrote:
> On Thu, 24 Oct 2013 08:34:17 -0500
> Tom Zanussi <tom.zanussi@linux.intel.com> wrote:
> 
> 
> > @@ -1577,6 +1577,7 @@ static void event_remove(struct ftrace_event_call *call)
> >  		if (file->event_call != call)
> >  			continue;
> >  		ftrace_event_enable_disable(file, 0);
> > +		destroy_preds(file);
> >  		/*
> >  		 * The do_for_each_event_file() is
> >  		 * a double loop. After finding the call for this
> > @@ -1700,7 +1701,7 @@ static void __trace_remove_event_call(struct ftrace_event_call *call)
> >  {
> >  	event_remove(call);
> >  	trace_destroy_fields(call);
> > -	destroy_preds(call);
> > +	destroy_call_preds(call);
> 
> A small nit, but I don't believe we need this anymore.
> 
> First, what event that requires a call filter can be removed?
> 
> Second, if one could be removed, the previous call to "event_remove"
> would remove the filter for us, as destroy_preds(file) calls
> destroy_call_preds() if the USE_CALL_FILTER flag is set.
> 
> I'll keep it in for now, but may remove it later.
> 

Good catch.  Yeah, that destroy_call_preds() call is redundant - feel
free to remove it, or I can..

Thanks,

Tom

> -- Steve
> 
> 
> >  }
> >  
>