bug#23928: 25.0.95; Performance regression observable with smartparens

[Eli Zaretskii]

> Cc: npostavs@users.sourceforge.net, aaronjensen@gmail.com,
>  23928@debbugs.gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Sun, 10 Jul 2016 22:56:03 +0300
> 
> On 07/10/2016 10:20 PM, Eli Zaretskii wrote:
> 
> > It only causes trouble if done in the innermost loops of your
> > program.  Otherwise, the cost is almost negligible.
> 
> I see. Then I think we should figure out why it's become more expensive 
> in Emacs 25 (if that is indeed the case).

It isn't yet clear that this is what happened.  Can someone time this
in both versions compiled with the same optimization options?