[EMAIL PROTECTED] wrote:
On Tue, Jan 22, 2008 at 02:09:18PM -0800, Andrew Lentvorski wrote:
No, the original Lisp-M was meant for Lisp, period. It wasn't until
they started to fail that they ported to mainstream processors.
Well it is a shame they didn't last.
Why? They made incredibly crappy business decisions. Having good
engineers tied down in bad companies is not good for the economy.
Better for the companies to go bankrupt and release the engineers to
somewhere better.
And, BTW, don't underestimate "mainstream" processors. The conventional
"wisdom" for a long time was that stack machines were always the way to
go even for VM's. Suddenly, VM's are starting to switch to
register-based VM's and are getting big performance gains.
Every assembly I've ever learned (x86 and ARM) had a stack *and* registers.
You seem to be drawing a distinction as if some CPUs have one but not the
other. Please elaborate.
Stack is slow. Registers are fast. Nobody computes using the stack
unless they absolutely have to. Stack is really only used as a place to
spill registers.
Even all of those variables that C claims are going onto the "stack" are
really only allocated in registers if the compiler can get away with it.
-a
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
