(/ 1 0) kills emacs on a DecStation running Ultrix

Article View: gnu.emacs.bug

Article #98534

#98534

From: Boaz Ben-Zvi
Date: Tue, 14 May 1991 22:03

14 lines
1427 bytes

Dividing by zero (in elisp) killed my Emacs (18.5[67]) on a DecStation 5000 
running Ultrix 3.1 and produced the message: "Trace/BPT trap" .
  Looking at the code ("src/data.c"), it seems that emacs expects a SIGFPE
to be generated when the division by zero happens. I tried using either "cc"
or "gcc" on various DecStations and versions of Ultrix, but they all produced
the same signal: "Trace/BPT trap (User Breakpoint (used by debuggers))"
  I guess the solution is to fix  init_data()  to use  SIGTRAP (???).

-- Boaz

========================================================
*   Boaz Ben-Zvi   (boaz@neutron.lcs.mit.edu)          *
*   MIT: NE43-526   Tel: (617) 253-6212                *
========================================================

Technical Details

Message-ID: <1991May14.220341.22670@mintaka.lcs.mit.edu>

Path: arc1-ca.pugleaf.net!gmdzi!unido!mcsun!uunet!tut.cis.ohio-state.edu!unreplyable!garbage

Show Full Headers

Complete Article Headers

X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: fb2f1,dfc575f1982cfcc4,start
X-Google-Attributes: gidfb2f1,public
X-Google-ArrivalTime: 1991-05-14 15:03:41 PST
Path: arc1-ca.pugleaf.net!gmdzi!unido!mcsun!uunet!tut.cis.ohio-state.edu!unreplyable!garbage
From: boaz@CONCERTO.LCS.MIT.EDU (Boaz Ben-Zvi)
Newsgroups: gnu.emacs.bug
Subject: (/ 1 0) kills emacs on a DecStation running Ultrix
Message-ID: <1991May14.220341.22670@mintaka.lcs.mit.edu>
Date: 14 May 91 22:03:41 GMT
Sender: daemon@tut.cis.ohio-state.edu
Reply-To: boaz@lcs.mit.edu
Distribution: gnu
Organization: Laboratory for Computer Science, MIT
Lines: 14
Approved: bug-gnu-emacs@prep.ai.mit.edu