Re: Scrolling commands and skipping redisplay, was: Re: emacs rendering comparisson between emacs23 and emacs26.3

[Eli Zaretskii]

> From: Richard Stallman <address@hidden>
> Cc: address@hidden, address@hidden, address@hidden,
>       address@hidden, address@hidden, address@hidden
> Date: Tue, 21 Apr 2020 23:14:07 -0400
> 
>   > That would mean the last line we scroll into the window will be only
>   > partially visible, and that is a no-no in scrolling.
> 
> It is somewhat undesirable, but I am not convinced it must be
> absolutely rejected.

That's currently a very significant requirement throughout
redisplay_window and its subroutine: any result that leaves the
current line (i.e. the line where point is) not completely visible is
rejected, and we select some other window-start point to avoid that.
It's possible to toss this requirement, of course, but then a lot of
existing code in the display engine will have to be rewritten, because
I won't be surprised if the assumption that this holds is implicitly
made by some parts of the display code.