Re: [PATCH 01/10] perf hists: Consolidate __hists__add_*entry()

[Rodrigo Campos <rodrigo@sdfg.com.ar>]

On Wed, Nov 06, 2013 at 05:16:04PM +0900, Namhyung Kim wrote:
> Hi Rodrigo,
> 
> On Tue, 5 Nov 2013 21:52:43 +0000, Rodrigo Campos wrote:
> > On Tue, Nov 05, 2013 at 05:09:52PM -0300, Arnaldo Carvalho de Melo wrote:
> >> @@ -486,15 +425,15 @@ struct hist_entry *__hists__add_entry(struct hists *hists,
> >>  		.ip	= al->addr,
> >>  		.level	= al->level,
> >>  		.stat = {
> >> -			.period	= period,
> >>  			.nr_events = 1,
> >> +			.period	= period,
> >>  			.weight = weight,
> >>  		},
> >
> > Isn't this seems unrelated and unneeded ?
> >
> > The "period" field is before the "nr_events" field in the struct, so maybe is
> > more clear to leave it as it was ?  The actual relative order (it has some more
> > fields) in the struct is: period, weigth, nr_events. Might be better if they
> > match that order here ? Although not sure since we are using the fields with
> > name and is clear enough.
> 
> Yes, it just a small unrelated cosmetic change.  I don't think the order
> matters much - it just makes my eyes a bit more comfortable. :)
> 
> IOW, I changed it since _add_branch_entry() and _add_mem_entry() do it
> slightly different order.  So I decided to clean it up and putting
> nr_events at first looked reasonable to me.

Great, sounds reasonable to me too :)




Thangs again,
Rodrigo