Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!cbmvax!vu-vlsi!colin
From: colin@vu-vlsi.UUCP
Newsgroups: comp.lang.c,comp.sys.ibm.pc
Subject: Re: Help with ^C in MS-DOS
Message-ID: <667@vu-vlsi.UUCP>
Date: Tue, 24-Mar-87 16:00:24 EST
Article-I.D.: vu-vlsi.667
Posted: Tue Mar 24 16:00:24 1987
Date-Received: Thu, 26-Mar-87 02:44:02 EST
References: <234@cup.portal.com> <1922@hoptoad.uucp>
Reply-To: colin@vu-vlsi.UUCP (Colin Kelley)
Distribution: world
Organization: Villanova Univ. EE Dept.
Lines: 26
Xref: utgpu comp.lang.c:1315 comp.sys.ibm.pc:2518

In article <1922@hoptoad.uucp> pozar@hoptoad.UUCP (Tim Pozar) writes:
>In article <234@cup.portal.com> RichardKitts@cup.portal.com writes:
>>
>>   I am trying to prevent a ^C break in MS-DOS using MSC 4.0.
>
>   There are a couple of ways to handle a ^C.
>   MSC provides the best way.  Look up signal() (page 350 Run Time
>Library Manual).  Signal() will let you define what should go on when
>a ^C is received by the programme.  In this way you can define how the 
>programme should go through an orderly exit of the routine or the programme
>itself.  The example on page 352 is a good example of this.
>        Tim Pozar

Beware that MS-DOS (starting with version 3.0) uses its own stack, and as
a result, the stack segment will not be correct when the signal handler
is started.  (Don't the compiler writers talk to the MS-DOS people???)
The symptom of this problem is a "stack overflow" message when the ^C
handler is invoked.  I've got a very short assembly language routine to get
around this--it first restores the stack segment, then calls the C signal
handler.  I can mail you the routine if you want it.  (It's not worth posting
since I just posted it a few months ago...)

	-Colin Kelley  ..{cbmvax,pyrnj,bpa}!vu-vlsi!colin

>terrorist cryptography DES drugs cipher secret decode NSA CIA NRO IRS
>coke crack pot LSD russian missile atom nuclear assassinate libyan RSA