From: muta...@gmail.com   
      
   On Saturday, August 7, 2021 at 12:47:43 AM UTC+10, Scott Lurndal wrote:   
      
   > >Ok. But that wasn't available on the 8086. Maybe there was   
   > >some way "fake paging" could have been created. Regardless,   
   > >we ended up with the "relocation (base) register" option, which   
   > >is presumably what a segment register is considered to be.   
      
   > I really don't understand your obsession with the 8086.   
      
   It is fascinating.   
      
   Imagine maxing out the 1 MiB and then you rock up with   
   an 80386 with 2 MiB of memory and suddenly you double   
   the amount of available memory.   
      
   The 8086 was hiding its secrets - a pivot in computer   
   science.   
      
   > >> Simple fact is that program which has more than 1M of code   
   > >> will not run on 1M machine.   
   > >   
   > >This is true, but I would have stopped right here.   
      
   > Actually, this is not true. Back in the day, long before   
   > the 8086/8088 were released, real operating systems were   
   > supporting applications larger than the maximum segment   
   > size using secondary backing storage (i.e. swap the segment   
   > out and replace it with another segment).   
      
   Ok, there are two things I wish to avoid - overlays and paging.   
      
   I'd rather bide my time waiting for real memory to come   
   online so that I can keep a conceptually simple system.   
      
   Noting that the end I need less than 32 MiB to do what   
   I want, which is get my toolchain to rebuild itself.   
      
   I don't mind if that compilation procedure, that uses more   
   than 16 MiB, runs using paging on MVS/XA or whatever.   
   I just don't want my own code to have such kludges. Just   
   as I don't want my own code to be written in assembler,   
   even though someone needed to do that. I'd rather have   
   generated assembler and print that out to be typed back   
   in if we have a "redo".   
      
   One of my greatest concerns at the moment is the fact   
   that the memory management code isn't really done   
   properly for PDOS/86 because I used large memory   
   model instead of huge memory model under Watcom.   
      
   BFN. Paul.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)