Message-ID:	<348EF16F.2A5@cs.com>
Date:	Wed, 10 Dec 1997 19:45:51 +0000
From:	"John M. Aldrich" <fighteer AT cs DOT com>
Reply-To:	fighteer AT cs DOT com
Organization:	Two pounds of chaos and a pinch of salt
MIME-Version:	1.0
To:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
CC:	djgpp AT delorie DOT com
Subject:	Re: RHIDE Debugging Idea...
References:	<Pine DOT SUN DOT 3 DOT 91 DOT 971210111442 DOT 10111E-100000 AT is>

Eli Zaretskii wrote:
> 
> On Tue, 9 Dec 1997, John M. Aldrich wrote:
> 
> > The good
> > news is that there already exists a function in the DJGPP libc named
> > __dpmi_set_debug_watchpoint(), which calls the DPMI function 0x0b00.  It
> > accepts a pointer to a given location in your program and installs a
> > breakpoint at that location when the program is run under a debugger.
> 
> How is that different from just setting a breakpoint (a hardware one, if
> you need that, with the hbreak command) using the debugger itself?

My concern isn't strictly the difference between the two; it's that the
function doesn't seem to work at all.  :-)  If it did work as promised,
it would solve the original poster's problem.  (Whether that problem
needs to be solved in that manner is a different question entirely.)

Another possibility is to include a command to set the breakpoint in the
'gdb.ini' file that gets used with that particular program.  This would
automatically enable the requisite breakpoint when gdb reads that
program, although it would be deleted or preserved just like any other.

-- 
---------------------------------------------------------------------
|      John M. Aldrich       | A singer in a smoky room / Smell of  |
|       aka Fighteer I       | wine and cheap perfume / For a smile |
|   mailto:fighteer AT cs DOT com   | they can share the night / It goes   |
| http://www.cs.com/fighteer | on and on and on...                  |
---------------------------------------------------------------------

- Raw text -