🚀 go-pugleaf

I have a problem that I hope at least one of you out there could shed some 
light on. We have a program that functions very well when run on machines 
with all other DOSs than Microsoft's own DOS 5. It runs ok on DOS 3.3, 4 and 
on OEM versions of DOS 5, but *not* on the original MSDOS 5.

I have not had the opportunity to do much debugging on the program yet, but 
it is a fairly simple one written in Microsoft C, mostly doing some file 
handling stuff. When run in a DOS shell under Windows, it is aborted by 
Windows with a message saying "...illegal instruction, close down and 
reboot...". Windows hangs when exiting, making me suspect that some DOS code 
or data area is overwritten.

A call to Microsoft's MSDOS technical support did not do much good. The MS 
guy believed that the OEM versions were all based on Microsoft's code, with 
only minor changes. Furthermore, no revisions of MSDOS 5 have ever been 
issued (i.e. no 5.01 etc). If this is the case, then my application should 
run equally good or bad on all DOSs, which it obviously does not.

There is a bug in my application, no doubt, but with some rather odd 
properties that I partly will have to explain to the users who cannot 
run the program even on plain vanilla machines with the original MSDOS 
5 installed. So, could somebody help me clear up on these DOS matters a 
little bit? For example:

	- Who licences DOS code from who?
	- Who uses IBMBIOS.COM/IBMDOS.COM and who uses IO.SYS/MSDOS.SYS ?
	- Are there any known functional differences between the different 
	  DOS brands? Any of the OEMs having bug-fixed the original 
	  MSDOS code?
	- If the program fails on MSDOS 5, what other OEM DOSs should I 
	  expect it to fail with?

Thanks,
Bj›rn.

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
   Bjorn Myrland                  *  bjorn.myrland@sipaa.sintef.no
   SINTEF Safety and Reliability  *  N-7034 Trondheim, NORWAY
 -------------------------------------------------------------------
   SINTEF - The Foundation for Scientific and Industrial Research
                at the Norwegian Institute of technology
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *