6502 unwound unsigned 8x8->16 multiply

6502umul_8x8_16.s

mul1 * mul2 + A/2 -> A:mul1 (mul2 is unchanged)

; inner loop credit Supercat

MUL:
     dec mul2	;5  ; decrement mul2 because we will be adding with carry set for speed (an extra one)
     ror mul1	;5      \
     bcc b1	  ;2/3     \  Best case 8 Worst case 10
     adc mul2	;3       /
b1: ror		    ;2       \
     ror mul1	;5         \
     bcc b2	  ;2/3      /  Best case 10 Worst case 12
     adc mul2	;3       /   
b2:  ror
     ror mul1
     bcc b3
     adc mul2  ; 10 or 12
b3:  ror
     ror mul1
     bcc b4
     adc mul2  ; 10 or 12
b4:  ror
     ror mul1
     bcc b5
     adc mul2   ; 10 or 12 
b5:  ror
     ror mul1
     bcc b6
     adc mul2   ; 10 or 12 
b6:  ror
     ror mul1
     bcc b7
     adc mul2   ; 10 or 12 
b7:  ror
     ror mul1
     bcc b8
     adc mul2  ; 10 or 12
b8:  ror		   ; 2
     ror mul1	 ; 5
     inc mul2  ; 5

	 ; total cycles worst case = 7 + 10 + (7 * 12)  + 12   = 113 
	 ; total cycles best case = 7 + 8 + (7*10) + 12 = 97 
	 ; avg = 105