Hello Michael,Arkady, and everybody else

WITHOUT PROGRAMMING OR TESTING !!

 some thoughts:

being pressed by arkady, I rethought the issue.

I think, that even IF EMM385 returns A000, the old kernels
will not merge a000 UMB into lower RAM.

old kernel, old a001 EMM386:

   10000 - 9ffff   available to programs
   9ffff - a0000   possibly  used as UMB link, but not available
   a0000 - a000f   hidden by emm386 'change a000 to a001'
   a0010 - affff   available UMB
...

old kernel, NEW a000 EMM386:

   10000 - 9ffff   available to programs
   9ffff - a0000   possibly  used as UMB link, but not available
   a0000 - affff   available UMB
...

AFAICS, the 9ffff - a0000 range is used as UMB link, but in itself
never available.
the problems I saw are
    enable_UMB()
    dos_alloc() returns  8000-afff
    disable_UMB()

however dos_alloc() returning  8000-afff should never be happen,
as the UMB link area is used.
    

arkady turbo kernel, NEW a000 EMM386:

   10000 - affff   available to programs

if the a000-afff area is indeed merged into lower memory and
UMB area starts at Cxxxx, this might even work.

however I still would prefer arkady to recompile emm386
and tell us if that works, instead of using mental input bandwidth of
all people subscribed to this list.

Finally, an idea that might even be used:

I found that using EMM386 /IA000-Afff is dangerous.

But the Numega SofyIce people had the idea to write a program

  LOADBIG MyProgram MyArguments

that would
     merge A000-Afff into low memory
       run MyProgram MyArguments
     unmerge A000-Afff from low memory

so I could run a very normal system; but on request (for make/compile
jobs mostly) I could enjoy 64/96 K more memory.

tom




     
     
  




  



















-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com
_______________________________________________
Freedos-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/freedos-devel

Reply via email to