Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!mit-eddie!uw-beaver!tektronix!cae780!hplabs!sdcrdcf!trwrb!ucla-an!ondine!steve
From: steve@ondine.uucp (Steve Jenkins)
Newsgroups: comp.sys.ibm.pc
Subject: Trapping floating-point exceptions under Turbo C
Message-ID: <88@ucla-an.UUCP>
Date: Tue, 6-Oct-87 17:27:50 EDT
Article-I.D.: ucla-an.88
Posted: Tue Oct  6 17:27:50 1987
Date-Received: Sat, 10-Oct-87 05:59:23 EDT
Sender: news@ucla-an.UUCP
Reply-To: steve@ondine.UUCP (Steve Jenkins)
Distribution: world
Organization: UCLA Dept of Anesthesiology, Los Angeles, California
Lines: 15
Summary: How do you do it?

References:


I would like to get advice from PC users on trapping floating-point
errors under Turbo C -- something along the lines of Unix signal().

I've read about matherr(), but it seems to be called only for errors
occurring in the math library and/or 8087 emulation routines.  I'm
trying to trap runtime exceptions that occur in my code (a programmable
data-acquisition system).  The ideal solution would work with and
without the 8087 -- for now I'll settle for an 8087-specific solution.

Please reply by email.  I'll post a summary of responses to this group.

Thanks.
Steve Jenkins			ucla-an!steve@EE.UCLA.EDU
UCLA Dept of Anesthesiology	{decvax,ihnp4}!hermix!ucla-an!steve
				(213)-825-4364