Sender: RUPP AT gnat DOT com
Message-ID: <3342C9A0.7099BBEB@gnat.com>
Date: Wed, 02 Apr 1997 21:03:28 -0800
From: Douglas Rupp <rupp AT gnat DOT com>
Organization: Ada Core Technologies
MIME-Version: 1.0
To: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
CC: Douglas Rupp <rupp AT gmat DOT com>, djgpp-workers AT delorie DOT com
Subject: Re: GDB and SIGALRM
References: <Pine DOT SUN DOT 3 DOT 91 DOT 970403073912 DOT 25849A-100000 AT is>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Eli Zaretskii wrote:
> 
> > What would be involved in getting a program that raises SIGALRM to work
> > with GDB?
> 
> You need to patch the DJGPP debugger support and go32targ.c from GDB.
> Roberth Hoehne has posted the necessary patches here in the past.
> Those patches are already incorporated into RHIDE and RHGDB, so if
> it's a one-shot job, you might as well just use RHGDB instead of
> downloading GDB sources and rebuilding it.
> 
> DJ, is it hard to make djgpp-workers archives available for searching
> through the Web?  I understand that the mail gets accumulated, but I
> couldn't find a way to search it, and sometimes I need to look for
> past correspondence here.

Robert told me:
 With no one of the current DJGPP debuggers you can debug programs,
 which gets and handles signals. Under GDB you will get probably
 undefined results and under RHIDE/RHGDB you will get your
 program gracefull terminated with an mesage about the signal.

--
This doesn't help.  I need SIGALRM to be passed on to the program
being debugged.  GNAT uses SIGALRM to implement delay statements and
so any program that uses these can't be debugged.