🚀 go-pugleaf

wilson@cs.utexas.edu (Paul Wilson) writes:

> [...whole lot of banging on pointer-swizzling cut out...]
>
> This is why I think that increasing the address bus size unnecessarily
> is crazy.  64 bits is okay, since we'll have main memories too big
> for 32 bit buses---but beyond that is lunacy.  (We can even make 32
> bit buses work well for very large main memories, if we're clever.)
> 
> Why put something that almost by definition goes unused almost all the
> time right smack in the critical path (and data paths) of your hardware?
> Wouldn't you rather use those paths for communicating useful information?
> Pointer swizzling and (other) forms of compression let you optimize
> the use of your data paths, rather than throwing expensive hardware
> at the problem.

Ok, whatzitake to make this happen on a "real" machine?
One that can be sold to a "naive" end-user?
And that's secure against a malicious programmer?
(at least as secure as current Unices & VM systems are now ;-)

Asking to use a special compiler & library (Texas-style) doesn't
necessarily cut it, because the malious programmer can side-step the
compiler.  It's OK if Mr.Bad can only crash his processes, but it's
Not OK for Mr.Bad to bring the system down by scribbiling clever
values into pointer slots and paging things in & out.

Do you need hardware help (bit-per-word specifying pointers)?  
OS help?  What?  (Nothing??)

Cliff

(not that I'm volunteering or accusing, just curious)

--
Cliff Click                  Compiler Researcher & Designer
RISC Software, Motorola      PowerPC Compilers
cliffc@risc.sps.mot.com      (512) 891-7240