bug#64596: 30.0.50; On FIXME: in src/buffer.c:1481 (force-mode-line-update)

[Stefan Monnier]

>> > That's not all of the uses of this flag.  Here's one example of other
>> > uses:
>> >
>> >      if (some_windows
>> >          && !f->redisplay
>> >          && !w->redisplay
>> >          && !XBUFFER (w->contents)->text->redisplay)
>> >        continue;
>> >
>> > (This avoids updating the tool bar and the tab bar of the frame, and
>> > there's a similar condition that avoids updating the frame's title.)
>> Right, but notice that what we're fundamentally checking is whether `w`
>> (which is the frame's selected window here) needs to be updated, because
>> that's the one currently "occupying" the tool-bar.
> But testing the frame's redisplay flag will produce false positives,
> because many changes that require to redraw the frame's windows don't
> affect the tool bar or the frame's title.

Of course, we have loads of false positives.  The point of the
`redisplay` flags is just to rule out things we can easily rule out.
It's very coarse.  It could be made finer, of course.

> No, my point is that fset_redisplay shouldn't set
> windows_or_buffers_changed.

Setting `windows_or_buffers_changed` there is currently necessary in
order to make sure we look at the `redisplay` bits (as opposed to
updating only the selected window).


        Stefan