Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!ucsd!chem.ucsd.edu!tps
From: tps@chem.ucsd.edu (Tom Stockfisch)
Newsgroups: comp.lang.c
Subject: Re: checking for overflow in C
Message-ID: <472@chem.ucsd.EDU>
Date: 17 May 89 05:18:58 GMT
References: <13367@dartvax.Dartmouth.EDU> <1989May6.224226.22085@utzoo.uucp> <1989May9.183140.1770@utzoo.uucp> <8172@june.cs.washington.edu> <1079@atanasoff.cs.iastate.edu> <1989May12.154417.21344@utzoo.uucp> <10256@smoke.BRL.MIL>
Reply-To: tps@chem.ucsd.edu (Tom Stockfisch)
Organization: Chemistry Dept, UC San Diego
Lines: 17

In article <10256@smoke.BRL.MIL> gwyn@brl.arpa (Doug Gwyn) writes:
_A few years ago, a system manager here, responding to naive "customer"
_pressure to generate floating exceptions on our Gould PN series machines,
_modified the run-time start-off module to enable exceptions.  What he
_didn't appreciate was that the C compiler generated code that required
_integer overflow to be ignored, and the Gould hardware tied integer
_and floating exceptions to the same enable bit.  Consequently non-floating
_applications started dying due to unexpected SIGFPEs.  We were not amused.

Aren't unsigned int operations guaranteed not to generate exceptions?
It seems like the C compiler could/should have done things differently.

So, what machines/systems does SIGFPE not work reliably on?

-- 

|| Tom Stockfisch, UCSD Chemistry	tps@chem.ucsd.edu