[rfc] Frame based register cache / frame->unwind

[Andrew Cagney]

> Hmm...  I see that the unwind cache has an ``addr'' field.  Does that
> mean that that the prologue analysis function calls
> frame_supply_unwound_register() to set this field?
> 
> Yes, that is what I had in mind.  But see above, it needs tinkering.

To give a ``status report''.  It is flushing out a few, er, 
``problems''.  At present it only works if both ->saved_regs and the 
cache are supplied with values.

One case I've figured out is stack.c:frame_info() which assumes that 
registers can only be saved in memory and, consequently, uses 
->saved_regs directly.  Hmm, I guess much of the output from ``(gdb) 
info frame'' is meaningless on the x86-64?

Another case I'm wondering about is with rs6000's ->alloca_reg.

Sigh.
Andrew