Re: [RFC patch 29/32] trace event asoc remove semicolons

[Steven Rostedt <rostedt@goodmis.org>]

On Tue, 2011-05-03 at 10:06 -0400, Mathieu Desnoyers wrote:

> This depends on a preliminary ftrace cleanup patch I just forwarded to
> you. LKML has eaten that patch due to too many CC. [ insert favorite
> expression of frustration here ]. But I think the best way to proceed
> would be to get a Acked-by from you and let Steven pull the patch
> through the tracing tree.

Sure, we can do that (for the semi-colon clean up).

> 
> > 
> > > +#ifndef _TRACE_ASOC_DEF_
> > > +#define _TRACE_ASOC_DEF_
> > > +
> > >  struct snd_soc_jack;
> > >  struct snd_soc_codec;
> > >  struct snd_soc_card;
> > >  struct snd_soc_dapm_widget;
> > >  
> > > +#endif /* _TRACE_ASOC_DEF_ */
> > 
> > This is an unrelated change and should be split out.  Frankly I'm not
> > that this is an improvement anyway, the struct declarations aren't going
> > to do any harm...
> 
> Yes, this change is related, because these declarations will cause harm.
> The structure declaration cannot be put within an array declaration,
> because then they won't be simply "ignored": they will try to declare
> structures within an array, which is not valid. This is why I did this
> change in the same patch.

No, Mark is right in this case. This IS a separate change, and deserves
a separate patch. This patch is to remove the semi-colons. This change
should be separated out and stated it is for the array. In fact, you
should probably have two different patch sets. One for the semi-colon
clean up, which I'm fine with. And another set that does more to the
patch, like this change. If I do the array change and find that it
doesn't help much, I wont want this change. But I'm willing to keep the
semi-colon change.

Yes, when I convert ftrace and perf to use arrays, then this will be
needed, but it is not needed for this series. I haven't looked at the
other patches yet (besides the first one), but if you have more of this
in other patches, please remove it. You can make another patch set that
I can use before the ftrace/perf updates.

-- Steve