From: as@sci.fi   
      
   rridge@csclub.uwaterloo.ca (Ross Ridge) writes:   
      
   > So presumably copy57.5K switches to the memory map where everything is   
   > RAM and does the copy using the remaining 6.5K of RAM to store the code   
   > that does the copying.   
      
   Yes, presumably. But it's just a little bit hard to do I/O if you switch   
   I/O out of your memory space so that's not the thing to do, exactly. One   
   could use RAM to RAM copy to fill the RAM under I/O. But I wonder,   
   Pekka's routine was just 40 bytes using ROM routines so 6.5K (or 5.5K   
   since the program was actually copy 58.5k) RAM seems excessive. The   
   whole operating system is just 8K. So maybe the program skipped the I/O   
   area?   
      
   The other question is, what all do you need to take care of if you're   
   going to get rid of the kernal and still be able to load stuff into RAM   
   below and also get the kernal back later? But with a quick look it looks   
   like disabling interrupts is enough.   
      
   BTW, I think I found the program, it was COPY 58.5K and it's on a disk   
   image at   
   http://www.commodoreserver.com/PublicDiskDetails.asp?DID=BABE16C   
   33E24BE9A89B8743E4F56D2C   
      
   Maybe I'm going to try and figure out exactly how it works. Looks like   
   it's 11 blocks so 2.75K but it might be compressed.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)