As far as I am concerned, a stack machine is a great idea. I've dabbled here and there with CPU designs. Some of the most "efficient" I've come up with were stack machines. What's nice about them is that since the operands are implicit, you don't have to pack them into the instruction word. Instead of a 32-bit instruction, you can have an 8-bit instruction, and that means you can execute 4 times as many instructions for your memory/cache bandwidth. What always got in my way with the stack machines had to do with things like context switches and supervisor-mode stuff that required seeing the CPU in a less stack-oriented way. In THIS case, however, we won't have context switches to worry about or any of that.
Ok, what's the instruction set now? :) On 5/1/05, Ray Heasman <[EMAIL PROTECTED]> wrote: > Hi Timothy, > On Sun, 2005-05-01 at 18:38 -0400, Timothy Miller wrote: > > Let's figure out what instructions we need. I'll start vague, and we > > can get more specific as we decide what's important and what's not. > > There are a lot of hobbiests out there that have done tiny > microcontrollers before. Perhaps we could leverage some of them? > > If we are really trying to save gates, then I would suggest a stack > machine rather than a register-based design. I have seen complete stack > machines done in very few lines of HDL. Also, their word length can be > pretty independent of the rest of the design. > > A further advantage is that it's easy to build a Forth engine on top of > a stack machine, if people don't want to program it in assembler. (It is > also maybe worth noting that all PCI firmware is supposed to be written > in Forth, because it is trivial to support it on any CPU.) > > Here is a link to some stack machines out there: > > http://www.ultratechnology.com/chips.htm > > Whatever you decide, this is the sort of thing I like playing with, so I > suspect I will be involved with programming it, at some point. :-) > > Cheers, > Ray > > _______________________________________________ Open-graphics mailing list [email protected] http://lists.duskglow.com/mailman/listinfo/open-graphics List service provided by Duskglow Consulting, LLC (www.duskglow.com)
