entries with just the file path, removing the revision number, date,
author and state.  The actual information is now computed during
compilation and stored in the fasl itself.  (See ticket:48.)

boot-2010-02-1 as the bootstrap file.  You should probably also use
the new -P option for build.sh to generate and update the po files
while building.

  CLOS/defmethod test in cl-bench to run.  (I'm waiting for bug
  reports before adding this everywhere.)
o Add a temp to the MOVE-ARGUMENT vop so we can store things on the
  stack with large offsets.

o MOVE-FROM-SIGNED uses a shorter sequence of instructions with fewer
  branch instructions to figure out if the number is a fixnum.
o MOVE-FROM-UNSIGNED uses shorter sequence of instructions to set the
  bignum correctly.

instructions, when possible.  Also convert some shift instructions to
signx or clruw synthetic instructions.

	have according to CLtS.  Found by Paul Dietz.

	Use boot14.lisp to bootstrap.

	* src/compiler/generic/utils.lisp (fixnumize): Renamed from
	fixnum.

	* src/bootfiles/18e/boot14.lisp: New file.

	* src/code/exports.lisp ("VM"): Export fixnumize.

	* src/assembly/alpha/arith.lisp, src/assembly/alpha/array.lisp:
	* src/assembly/alpha/assem-rtns.lisp, src/assembly/hppa/arith.lisp
	* src/assembly/hppa/array.lisp, src/assembly/hppa/assem-rtns.lisp
	* src/assembly/mips/alloc.lisp, src/assembly/mips/arith.lisp
	* src/assembly/mips/array.lisp, src/assembly/mips/assem-rtns.lisp
	* src/assembly/mips/bit-bash.lisp, src/assembly/ppc/arith.lisp
	* src/assembly/ppc/array.lisp, src/assembly/ppc/assem-rtns.lisp
	* src/assembly/rt/alloc.lisp, src/assembly/rt/arith.lisp
	* src/assembly/rt/array.lisp, src/assembly/rt/assem-rtns.lisp
	* src/assembly/sparc/arith.lisp, src/assembly/sparc/array.lisp
	* src/assembly/sparc/assem-rtns.lisp, src/assembly/x86/arith.lisp
	* src/assembly/x86/array.lisp, src/assembly/x86/assem-rtns.lisp
	* src/compiler/alpha/alloc.lisp, src/compiler/alpha/arith.lisp
	* src/compiler/alpha/array.lisp, src/compiler/alpha/call.lisp
	* src/compiler/alpha/move.lisp, src/compiler/alpha/nlx.lisp
	* src/compiler/alpha/static-fn.lisp, src/compiler/alpha/subprim.lisp
	* src/compiler/alpha/values.lisp, src/compiler/generic/utils.lisp
	* src/compiler/hppa/alloc.lisp, src/compiler/hppa/arith.lisp
	* src/compiler/hppa/array.lisp, src/compiler/hppa/call.lisp
	* src/compiler/hppa/move.lisp, src/compiler/hppa/nlx.lisp
	* src/compiler/hppa/static-fn.lisp, src/compiler/hppa/subprim.lisp
	* src/compiler/hppa/values.lisp, src/compiler/mips/alloc.lisp
	* src/compiler/mips/arith.lisp, src/compiler/mips/array.lisp
	* src/compiler/mips/call.lisp, src/compiler/mips/move.lisp
	* src/compiler/mips/nlx.lisp, src/compiler/mips/static-fn.lisp
	* src/compiler/mips/subprim.lisp, src/compiler/mips/values.lisp
	* src/compiler/ppc/alloc.lisp, src/compiler/ppc/arith.lisp
	* src/compiler/ppc/array.lisp, src/compiler/ppc/call.lisp
	* src/compiler/ppc/move.lisp, src/compiler/ppc/nlx.lisp
	* src/compiler/ppc/static-fn.lisp, src/compiler/ppc/subprim.lisp
	* src/compiler/ppc/values.lisp, src/compiler/rt/alloc.lisp
	* src/compiler/rt/arith.lisp, src/compiler/rt/array.lisp
	* src/compiler/rt/call.lisp, src/compiler/rt/move.lisp
	* src/compiler/rt/nlx.lisp, src/compiler/rt/static-fn.lisp
	* src/compiler/rt/subprim.lisp, src/compiler/rt/values.lisp
	* src/compiler/sparc/alloc.lisp, src/compiler/sparc/arith.lisp
	* src/compiler/sparc/array.lisp, src/compiler/sparc/call.lisp
	* src/compiler/sparc/move.lisp, src/compiler/sparc/nlx.lisp
	* src/compiler/sparc/static-fn.lisp, src/compiler/sparc/subprim.lisp
	* src/compiler/sparc/values.lisp, src/compiler/x86/alloc.lisp
	* src/compiler/x86/arith.lisp, src/compiler/x86/array.lisp
	* src/compiler/x86/call.lisp, src/compiler/x86/cell.lisp
	* src/compiler/x86/macros.lisp, src/compiler/x86/memory.lisp
	* src/compiler/x86/move.lisp, src/compiler/x86/nlx.lisp
	* src/compiler/x86/pred.lisp, src/compiler/x86/static-fn.lisp
	* src/compiler/x86/subprim.lisp, src/compiler/x86/values.lisp:
	Use fixnumize instead of fixnum.

o Added immediate SC to move-64bit-word-argument.  (I think this is
  needed because the 64-bit stuff doesn't do very good representation
  selection and can't implicitly coerce things from one rep to another.)

sparc V9 architecture using the V8plus ABI.  Still needs work, but the
basic functionality is there.

o Always allocate 2 words for bignums even if we don't need both
  words.  (GC will take care of the extra word.)

semantically they are always signed.  If for some reason we are moving a
negative fixnum to an unsigned reg, we presumably know what we are doing.

The fixnum cases are split off so that we have a better idea of the cost (for
efficiency notes.)  Added notes to these VOPs.

the new representation selsection stuff knows to only use these VOPs
when the operands are fixnums.  An ambiguity existed because 
CONTROL-STACK is used to store both fixnums and bignums.  Previously 
the first-defined VOP was always used, which was wrong.

so that cost info will be more accurate, which is useful now that
there are representation selection efficency notes.  It also makes
the code clearer.  Also, added notes for these VOPs for use in
these efficency notes.

and do an immediate load.  We still allocate the indirect constant,
but don't waste time indirecting twice.

any-reg/descriptor-reg distinctions consistent.

thing else.

any attention to the primitive type restrictions when picking load-tns.

to/from registers.  As it's going away, I'm not going to worry about it.