Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.arch    |    Apparently more than just beeps & boops    |    131,241 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 129,257 of 131,241    |
|    John Savard to All    |
|    By Popular Demand    |
|    05 Aug 25 17:52:58    |
      From: quadibloc@invalid.invalid              I still don't have a dedicated stack pointer, or jump to subroutine       instructions that save the return address on a stack.              But, by reducing the size of displacements for relative addressing in the       standard memory-reference instruction from 11 bits to 10 bits, I have been       able to add the following address modes to standard memory-reference       instructions:              Register indirect       Register indirect with scaled auto-increment       Register indirect with scaled auto-decrement              so one can now push and pop values from registers to a local stack       belonging to a single routine.              The thing I reject is a single big stack that crosses between domains.              And there's one other addressing mode I've also added that this created       room for which had also been requested.              Register specifiers, when they're five bits long, could also point to       pseudo-immediates. So let the instruction point to a 64-bit pseudo-       immediate and use it as an absolute address.              You see, I can be reasonable. Please don't blow up any planets.              John Savard              --- SoupGate-Win32 v1.05        * Origin: you cannot sedate... all the things you hate (1:229/2)    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca