bug#71289: 30.0.50; cmcheckmagic aborts when tty_write_glyphs writes "Garbage collecting..." in some cases

[Daniel Clemente <n142857@gmail.com>]

> >
> > > Bug#71343 seems to describe a GUI frame (if not, what does the
> > > window-manager have to do with it?).  While here we are talking about
> > > TTY frames, right?  And killing frames is not involved in this bug
> > > report, right?  Or what am I missing?
> >
> > This one (bug 71289) seems to be about crashes involving GC messages
> > and resizing the terminal.
>
> AFAIR, almost all of the bugs you submitted have something to do with
> GC messages, so the above doesn't help in any way to make this one
> stand out, or am I confused?

Bug 71343 has nothing to do with GC.

This one (bug 71289) does. I don't know what you mean by stand out; it
doesn't need to stand out.
But each bug is different, e.g.:
- 71176 was a bug about GC messages not working well when
opening+closing frames, in some conditions
- this one (71289) is about GC messages not working well when resizing
the terminal, in some conditions


> > In all reports I'm using TTY frames, i.e. Emacs compiled without X
> > support; it's not GUI frames.  But I'm running them inside a urxvt
> > window in X with a window manager.  I wanted a simpler and smaller
> > Emacs without any X support, but I still run it in X.
>
> So how is the window manager relevant here?

Here (bug 71289) you need a window manager to resize the window (the X
window that is running emacsclient).
The particular window manager you use isn't important, in fact I saw
the same errors happen in wmii and icewm.  Saying this is important,
because I was afraid that the problems could come from wmii (wmii
failing to communicate that an X window has been resized). But the
window manager isn't at fault in this bug 71289.