[comp.lang.c++] Strange run-time error with Borland C++2.0 IDE

mcmillan@anaconda.cis.ohio-state.edu (Harold McMillan) (04/08/91)

I just got Borland C++ 2.0 and a brand-new computer - a Gateway 2000 386
- 33MHz with 4Mb RAM.  I consistently get the following error after a
few cycles of compling and running the CIRCLE.PRJ demo program.  The
error occurs when I try to run the program within the BCX environment.

GP at 0020 4FCA EC 0148
GP at 0020 59A0 EC 05C4
System: Cannot recover from exception
System: Kernel cannot continue
Error: Cannot load BCX.OVY because real-mode version is already running

I assure you that I have not been running BC when this occurs.
I am running BCX with no command-line options.
At this point, the ONLY device driver that I have
loaded is a mouse, but this error also occurs when I have QEMM and/or
PC-CACHE loaded.

I realize that this problem concerns a particular compiler, and not
the general topic of this group, but I was hoping that a nuber of you
are users of this compiler.

- Hal McMillan
--
Hal McMillan                | "It's an ideal principle, which can be verified
mcmillan@cis.ohio-state.edu |  only under ideal conditions.  Which means never.
72627.642@CompuServe.com    |  But it's still true."

jim@se-sd.SanDiego.NCR.COM (Jim Ruehlin, CFC User) (04/09/91)

In <104577@tut.cis.ohio-state.edu> mcmillan@anaconda.cis.ohio-state.edu (Harold McMillan) writes:

>I just got Borland C++ 2.0 and a brand-new computer - a Gateway 2000 386
>- 33MHz with 4Mb RAM.  I consistently get the following error after a
>few cycles of compling and running the CIRCLE.PRJ demo program.  The
>error occurs when I try to run the program within the BCX environment.

>GP at 0020 4FCA EC 0148
>GP at 0020 59A0 EC 05C4
>System: Cannot recover from exception
>System: Kernel cannot continue
>Error: Cannot load BCX.OVY because real-mode version is already running
>--

I've gotten the same error when linking in BCX.  Usually, it will occur
after I've re-compiled everything.  I run BCX in Windows, standard mode.

This is the only gross bug I've run across so far.

- Jim Ruehlin

ahodgson@athena.mit.edu (Antony Hodgson) (04/09/91)

In article <104577@tut.cis.ohio-state.edu> mcmillan@anaconda.cis.ohio-state.edu (Harold McMillan) writes:
>I just got Borland C++ 2.0 and a brand-new computer - a Gateway 2000 386
>- 33MHz with 4Mb RAM.  I consistently get the following error after a
>few cycles of compling and running the CIRCLE.PRJ demo program.  The
>error occurs when I try to run the program within the BCX environment.
>
>GP at 0020 4FCA EC 0148
>GP at 0020 59A0 EC 05C4
>System: Cannot recover from exception
>System: Kernel cannot continue
>Error: Cannot load BCX.OVY because real-mode version is already running
>
I had a very similar problem when I put a watch on (char*)0,4m to watch
for NULL pointer assignment.  The general protection fault occurs as
soon as I try to enter the watch - even before I run the program.

Tony Hodgson
ahodgson@hstbme.mit.edu

mcmillan@harpsichord.cis.ohio-state.edu (Harold McMillan) (04/09/91)

It seems that the problem with the GP exception  in Borland C++ 2.0
that I posted earlier may have to do with the Microsoft mouse driver.
Someone sent me mail to this effect, and I have not seen the problem
recur without the mouse driver loaded.  Now, how on earth did Borland
design BCX to be incompatible with the Microsoft Mouse, or did they
just see this as a way to get back at Microsoft?

	Hal
--
Hal McMillan                | "It's an ideal principle, which can be verified
mcmillan@cis.ohio-state.edu |  only under ideal conditions.  Which means never.
72627.642@CompuServe.com    |  But it's still true."

randall@Virginia.EDU (Randall Atkinson) (04/10/91)

In article <105414@tut.cis.ohio-state.edu> mcmillan@harpsichord.cis.ohio-state.edu (Harold McMillan) writes:
>It seems that the problem with the GP exception  in Borland C++ 2.0
>that I posted earlier may have to do with the Microsoft mouse driver.
>Someone sent me mail to this effect, and I have not seen the problem
>recur without the mouse driver loaded.  Now, how on earth did Borland
>design BCX to be incompatible with the Microsoft Mouse, or did they
>just see this as a way to get back at Microsoft?

Gee, why jump to unwarranted conclusions ?  The problem is that
OLDER versions of the MS Mouse driver aren't compatible.  BC++ is
fully compatible with the current MS Mouse driver.  The fix reported
by others on the net is to get a current MS Mouse device driver...

daveo@hpcvra.cv.hp.com. (Dave Ochs) (04/12/91)

/ hpcvra.cv.hp.com:comp.lang.c++ / randall@Virginia.EDU (Randall Atkinson) /  1:19 pm  Apr  9, 1991 /

>Gee, why jump to unwarranted conclusions ?  The problem is that
>OLDER versions of the MS Mouse driver aren't compatible.  BC++ is
>fully compatible with the current MS Mouse driver.  The fix reported
>by others on the net is to get a current MS Mouse device driver...

What versions are you referring to?  Versions...file sizes/dates...

Dave Ochs
Hewlett-Packard
daveo@cv.hp.com