[comp.emacs] The impossible happens - SIGIOT on Sun-3

mkhaw@teknowledge-vaxc.ARPA (Mike Khaw) (07/01/88)

We have a user (no, really, it wasn't me) who got a core dump with a
message about "IOT instruction" while running GNU Emacs 18.41 on a Sun
3/50 running SunOS 3.5.

The only reference I can find to IOT in the man files is in sigvec(2),
where it says under SIGIOT:

	IOT trap (not generated on Suns)

!!! What's going on here?

Thanks,
Mike Khaw
-- 
internet: mkhaw@teknowledge.arpa
uucp:	  {uunet|sun|ucbvax|decwrl|uw-beaver}!mkhaw%teknowledge.arpa
hardcopy: Teknowledge Inc, 1850 Embarcadero Rd, POB 10119, Palo Alto, CA 94303

jr@bbn.com (John Robinson) (07/01/88)

In article <23550@teknowledge-vaxc.ARPA>, mkhaw@teknowledge-vaxc (Mike Khaw) writes:
>We have a user (no, really, it wasn't me) who got a core dump with a
>message about "IOT instruction" while running GNU Emacs 18.41 on a Sun
>3/50 running SunOS 3.5.

When emacs reaches an impossible condition, it calls the C library
function abort.  If you look at abort(3), you'll see that it generates
a SIGIOT signal, which causes emacs to core dump.

To debug it, run dbx (or your favorite debugger) on the core file and
look at the stack.  Have the source handy.

I encountered an abort running under X11.  But since the X11 code
originated about version 18.49, I assume you aren't.

You might want to try to get the current distribution.
-- 

rich@oxtrap (K. Richard Magill) (07/06/88)

In article <23550@teknowledge-vaxc.ARPA>, mkhaw@teknowledge-vaxc (Mike Khaw) writes:
>We have a user (no, really, it wasn't me) who got a core dump with a
>message about "IOT instruction" while running GNU Emacs 18.41 on a Sun
>3/50 running SunOS 3.5.


Floating point binaries on a non-floating point sun?




rich.
--