Hi!

1-Июн-2005 10:29 [EMAIL PROTECTED] (Michael Devore) wrote to
[email protected]:

MD> Back in the early days with mostly BIOS calls being made, some applications
MD> would only save the low part of 16-bit registers because the BIOS was
MD> "supposed to" change one 8-bit register of the pair.

     ...and they (AL/AH) may be counted as separate registers.

MD> Drivers, too.  You
MD> know what?  Some BIOSes and drivers changed the whole 16-bit register and
MD> broke those overly thrifty applications.  So the BIOS was literally buggy

     Yes! Corporate programmers make many bugs, some of them was documented
as "feature" (like INT25/26 stack patching). About BIOS: some editions was
broke even unpaired registers, like BP, and this is also bug. :(

MD> and the applications were right -- and it was still a dumb thing for the
MD> applications to do.  I speak from experience having done exactly that dumb
MD> thing once and having learned my lesson.

     Yes, in general applications may (and sometime should) overcare about
possible bugs in external (including system/hardware) code. But me talk
about precise case (EMM386, which modify unspecified (part of) register),
and this case may (and should be) fixed, notwithstanding if some/most
extenders contains (or not) extra protection again server bugs.

MD> But you may be looking for a debate since you're basically quoting back my
MD> message and negating it with your own viewpoint.  I am uninterested in
MD> arguing semantics, specs, and coding philosophy right now.  It's going to
MD> be fixed regardless.

     Perfect!

MD> While muttering under my breath.




-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r 
_______________________________________________
Freedos-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/freedos-devel

Reply via email to