Erann Gat <gat@jpl.nasa.gov> wrote:
+---------------
| P.S.  Here's what MCL does.
| ? (defun m-f (x y) (declare (fixnum x) (fixnum y)) (the fixnum (* x y)))
| M-F
| ? (disassemble 'm-f)
|   (TWNEI NARGS 8)
|   (MFLR LOC-PC)
|   (BLA .SPSAVECONTEXTVSP)
|   (VPUSH ARG_Y)
|   (VPUSH ARG_Z)
|   (LWZ NARGS 331 RNIL)
|   (TWGTI NARGS 0)
|   (LWZ ARG_Y 4 VSP)
|   (LWZ ARG_Z 0 VSP)
|   (SRAWI IMM0 ARG_Z 2)
|   (MULLW ARG_Z ARG_Y IMM0)
|   (BA .SPPOPJ)
| ?
+---------------

For CMUCL, you have to crank SPEED up and SAFETY down, but then you get
something even faster:

   > (defun m-f (x y)
       (declare (fixnum x y) (optimize (speed 3) (safety 0)))
       (the fixnum (* x y)))

   M-F
   > (disassemble 'm-f)
   ; Compiling LAMBDA (X Y): 
   ; In: LAMBDA (X Y)
   ;   (* X Y)
   ; Note: Unable to recode as shift and add due to type uncertainty:
   ;     The first argument is a FIXNUM, not a (UNSIGNED-BYTE 32).
   ;     The second argument is a FIXNUM, not a (UNSIGNED-BYTE 32).
   ;     The result is a FIXNUM, not a (UNSIGNED-BYTE 32).
   ; Compilation unit finished.
   ;   1 note
   483F4468:   .ENTRY "LAMBDA (X Y)"(x y)   ; (FUNCTION (FIXNUM FIXNUM) FIXNUM)
	 80:   POP   DWORD PTR [EBP-8]
	 83:   LEA   ESP, [EBP-32]
	 86:   SAR   EDX, 2                 ; No-arg-parsing entry point
	 89:   IMUL  EDX, EDI
	 8C:   MOV   ECX, [EBP-8]
	 8F:   MOV   EAX, [EBP-4]
	 92:   ADD   ECX, 2
	 95:   MOV   ESP, EBP
	 97:   MOV   EBP, EAX
	 99:   JMP   ECX
	 9B:   NOP
	 9C:   NOP
	 9D:   NOP
	 9E:   NOP
	 9F:   NOP
   > 


-Rob

-----
Rob Warnock, PP-ASEL-IA		<rpw3@rpw3.org>
627 26th Avenue			<URL:http://rpw3.org/>
San Mateo, CA 94403		(650)572-2607