New register allocator branch created

[Richard Kenner]

    > <<I've worked with the new allocator on both x86, and PowerPC, and while
    > it is orders of magnitude better than the old allocator, at least on x86,
    > >>
    >
    > Can you explain what it means to say that a register allocator is powers
    > of ten better than another allocator. What metric are you using?

    How often it spills and the number of registers needed.

So it uses 1/100 the number of registers?