Hi Eli, thanks. It seems to be working on latest git master. I think it is
a good idea to cherry-pick this into the emacs-25 branch, since it is also
affected.

On Sat, Mar 11, 2017 at 5:33 AM, Eli Zaretskii <eliz@gnu.org> wrote:

> > Date: Fri, 10 Mar 2017 15:37:27 +0200
> > From: Eli Zaretskii <eliz@gnu.org>
> > Cc: emacs-devel@gnu.org
> >
> > > From: Filipe Silva <filipe.silva@gmail.com>
> > > Date: Fri, 10 Mar 2017 10:16:04 -0300
> > > Cc: Emacs developers <emacs-devel@gnu.org>
> > >
> > > Eli, Here's the recipe from emacs -Q
> > >
> > > https://github.com/ninrod/coredump-recipe
> >
> > Thanks, I will look into that soon.
>
> I think I fixed this on the master branch, please test.  (My master
> build is compiled with all kinds of runtime checks, so it hit an
> assertion violation before the place where you reported an abort.)
>
> Btw, vim-empty-lines-mode.el plays with fire here: in
> vim-empty-lines-count-screen-lines it narrows the buffer to the region
> just before point-max, while having an overlay affecting the display
> that ends at point-max.  This generally hits on all kinds of dark
> corners, so it would be better not to do stuff like that.
>