From: "Felix" <natter AT metronet DOT de>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Dissasember debugging bug?
Date: Wed, 11 Nov 1998 16:45:41 +0100
Organization: -
Lines: 38
Message-ID: <72cas3$k2q$1@news.metronet.de>
References: <19981109181515 DOT 00812 DOT 00001533 AT ng-fb2 DOT aol DOT com>
NNTP-Posting-Host: bonn2.pop.metronet.de
X-Newsreader: Microsoft Outlook Express 4.72.3110.5
X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

hi,
I've experienced the same Problem, but it only happens when you enter a
sub-level function and then go on single-stepping with F8, if you use F7 all
the time, it works fine
I still think this is a little bug, but it's easy to live with, so what the
hell...

good luck,
Felix Natter (if anybody knows how to change "my" name using MS-Outlook
Express, please tell me!)

>Everything comples and runs fine, but try placing a breakpoint on the
printf
>line, running the program, then opening the dissasembler.  Press F8 to get
to
>the "call AddFour" instruction, then press F7.  This will bring you to the
>disasembly of the AddFour funciton.  Press F8, and the program crashes.
>However, if you move the cursor down to say, "mov eax, [ebp + x_AddFour",
and
>press F4, everything works fine!
>
>I am suspecting the problem is that rhide debugger debugger is not setting
some
>registers properly when it transfers control back to the program.  But I'm
not
>experienced programmer, just an amatuer that's pulling his hairs out by the
>roots.  I've got a routine in a larger program where I first noticed this
>problem.  Doing any complicated ASM whatsoever is impossible without a
debugger
>that works.
>
>Of course, I could be being stupid, so if I am, please let me know.  If you
>don't mind, could you email me with a solition if one exists?
>(mrwiggley AT aol DOT com)
>
>Thanks everbody.