Hi,

On Fri, Jun 21, 2013 at 6:12 AM, Tom Ehlert <[email protected]> wrote:
> am 17. Juni 2013 um 10:00 schrieben Sie:
>
>> I'm having trouble running XPLIQ.EXE (among other rare ones) from
>> XPL0-30.ZIP in FreeDOS. http://www.xpl0.org/
>
> this definitively depends on your *precise* memory layout, which is
> obviously different in 1), 2), 3), 5), 6, DOSBox, DRDOS
>
> no possibility to decide if this is a bug of the freedos loader
> (triggered by this memory layout), or by a bug in the .exe itself.

Well, I did also reproduce this on my laptop. Did you have time to
try? Well, it's not super important, esp. since there are some
apparent workarounds. I really almost don't know what else to do.

>> 5). debug xpliq.exe (then 'g' to run, works because this has its own
>> .EXE loader, right??)
> no. debug uses (most likely ?) the freedos loader

A quick check of its source shows int 21h, 4B01h (load but don't
execute). So I guess you're right. Presumably loading it low is
avoiding the problem.

http://www.delorie.com/djgpp/doc/rbinter/id/51/29.html

>> (Perhaps this is a good
>> excuse to start using 386SWAT, but I don't know if that will run well
>> on this machine.)
> yep.

Well, my fears were correct, 386SWAT does not run (or at least needs
much further tweaking, probably beyond me), it just reboots the
machine entirely. Not just with this program, IIRC I mean just in
general, I can't load it resident at bootup at all on this desktop
(too new, I suppose).

I ended up trying Insight Debugger 1.24 (see BTTR) resident ("/r").
That way I can press Ctrl-Esc to bring it up when it freezes. Assuming
that works like I expect, and based upon a comparison with XMGR.ASM,
it's showing me that it hangs on what appears to be (near) the
beginning of XMGR's "XMSRtn:" procedure. Specifically it loads upon
the part beginning with "push esi // push edi". No idea why, not
enough stack?

Again, I have no idea if it's a FreeDOS bug or XMGR bug or both or
just an unexpected clash. (And I still don't rule out linker / .EXE
header error.)

And this probably isn't high priority, so I don't really want anyone
to waste any huge amount of time on it or stress themselves out. In
other words, feel free to ignore. But if someone does happen to know
how to debug something like this and is really bored ... well, feel
free!    :-)   It's just slightly tedious to keep forgetting about it
and having to reboot (over and over), so I thought it was worth
warning somebody (even if it only happens in semi-rare setups).

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
Freedos-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/freedos-devel

Reply via email to