Re: Dynamic modules: MODULE_HANDLE_SIGNALS etc.

[Eli Zaretskii]

> Cc: address@hidden, address@hidden, address@hidden,
>  address@hidden
> From: Paul Eggert <address@hidden>
> Date: Sun, 20 Dec 2015 13:40:36 -0800
> 
> Eli Zaretskii wrote:
> >> From: Philipp Stephani <address@hidden>
> >> Date: Sun, 20 Dec 2015 18:34:03 +0000
> >> Cc: address@hidden, address@hidden, address@hidden,
> >>    address@hidden, address@hidden
> >>
> >>      Thanks, but please mention in these comments that emacs-module.c
> >>      relies on these requirements.
> >>
> >> Done. Attached new patch.
> >
> > Thanks, pushed.
> 
> These newly-added comments are incorrect. Any of these functions can exit 
> nonlocally due to stack overflow. I assume it's OK to revert them?

Then we either cannot use them in emacs-module.c, or should place
comments there to the effect that we disregard that possibility (and
maybe augment the comments in lisp.h saying that stack overflow is the
only allowed reason for non-local exits).  Which one do we prefer?