Mail Archives: djgpp/1997/12/27/08:07:32
On 26 Dec 97, Salvador Eduardo Tropea (SET) was found to have
commented thusly:
> "S. M. Halloran" <mitch AT duzen DOT com DOT tr> wrote:
>
> > I am seeing a W95 exception every time rhIDE exits. The machine has
> > 16 MB RAM and 120+ MB free uncompressed drive space. Clearly I have
> > a configuration problem. I am reporting the information regarding
> > the exception which the system provides below.
> > ------------------
> > WINOLDAP caused an invalid page fault in
> > module VGAFULL.3GR at 0001:00000c3f.
> ^^^^^^
> It looks like a problem with the video drivers. I guess is a bug in RHIDE or in
> your video drivers. My experience says that the drivers for W95 are usually
> FULL of bugs. Be sure you have the last version from the manufacturer. A friend
> had hard problems with Allegro programs running in W95 with a Trident
> TGUI9440AGi the problem was that he was trying to use the drivers shipped with
> W95. Isn't the only case I know.
>
> > Registers:
> > EAX=000071b0 CS=0c87 EIP=00000c3f EFLGS=00000212
> > EBX=000200ac SS=259f ESP=000045ca EBP=000046ba
> > ECX=00000020 DS=259f ESI=8281c000 FS=207e
> > EDX=8281bfa2 ES=01b3 EDI=82814000 GS=0327
> > Bytes at CS:EIP:
> > f3 67 26 a7 75 13 66 59 66 2b f2 8b c6 d1 e8 2b
> > Stack dump:
> > 0001004f 00000038 bfa00bc6 3fa08281 071b8281 00000000 202002b4
> > 20202020 20202020 20202020 20202020 20202020 20202020 20202020
> > 20202020 20202020
> > ------------------
> >
> > Speculations as well as correct analyses welcome.
> I sent mine ;-) Does it happends with all the RHIDE video modes? Did you tried
> -G1?
>
I was thinking that you (or another) might have suggested
that something was going on with the video driver; the card is an
S3Trio 64V2 DX (775), and S3 Inc. provides its own EXE to run
under DOS to identify the particular chip, and that was used to
select the driver, which I obtained off the Internet from their
web/ftp site not less than a month ago. If there is something wrong
with the driver therefore, then S3 hasn't discovered it yet.
But at any rate, I seem to have solved the problem. Your astute
perception as to whether it happened in all video modes led me to
discover than rhIDE exits without the page fault in 80x25 in a DOS
window. I also noticed that the window remained instead of
automagically closing. The mode giving the problems was 80x43 text
in a DOS window (with Windows vidmode set at 800x600x16-bit color), a
mode I became familiar with when using Turbo C++ 3.0. In the
Properties box, I set Close on Exit in the General tab, and the
problem seemed to go away when rhIDE was opened (it opened in
80x25 mode) and closed in 80x43 mode. However, I did one other thing
at the same time--violating the
change-one-thing-at-a-time-to-pinpoint-the-problem rule...I also
increased the DPMI-memory request that DOS makes since I needed a VM
of 15-20 MB for this particular program. So what change could have
possibly had the positive effect?
If I have time---who does--and if someone reminds me to do so in the
future, I promise to be more systematic in identifying the modes and
eliciting the exceptions for this particular setup. But right now,
the boss is setting deadlines on matters of higher priority....
I have just been using rhIDE and djgpp for a couple of weeks now, so
I really am not familiar with all the wonderful features these things
have to offer. I am attempting to take a crash course in gcc (as
djgpp) and rhIDE...and trying to find out what documents I am
missing, especially the one that discusses the meaning of the
compiler error and warning message from gcc/djgpp that appear in
rhIDE's Message window (by the way, is there a reason why Alt-W + M
and Alt-W + W do not call up by default the Message and Watch
windows, resp., as is the case in Borland's Turbo IDE?)
Best regards,
Mitch Halloran
Research (Bio)chemist
Duzen Laboratories Group
Ankara TURKEY
mitch AT duzen DOT com DOT tr
other job title: Sequoia's (dob 12-20-95) daddy
- Raw text -