The PICList code generator: Multiplication and Division by any Constant through a simple html form
Binary Multiplication by a Constant
8 bits by any constant by Myke Predko
See Division Methods for divide by 10, etc...
Since rotating any binary value one place left or right will multiply or
divide by 2:
;Multiply 8 bit ACC by 2 (double) rlf ACC, f ;Multiply 16 bit ACC by 2 (double) clrc rlf ACC0, f rlf ACC1, f ;Divide 8 bit ACC by 2 (half) rrf ACC, f ;Divide 16 bit ACC by 2 (half) clrc rrf ACC0, f rrf ACC1, f ; 28/32-bit unsigned Multiply by 16 ; by Martin Sturm 2010 ; ; Left shift by 4 with zero fill on right ; NOTE: discards uppermost bits shifted out ; 2 instructions shorter than unrolled RLF ; 1 instruction longer than code generator but can tolerate ; non-zero bits in the upper nibble of high byte ; Can be easily modified for 24, 16 bit versions ; ; 17 instructions, 17 cycles MULT_U12_BY16 MACRO x4,x3,x2,x1 ; x4 x3 x2 x1 W ; a b c d e f g h MOVLW 0x0F ; 0x0F ANDWF x4,F ; 0 b SWAPF x4,F ; b 0 SWAPF x3,F ; d c ANDWF x3,W ; 0 c IORWF x4,F ; b c XORWF x3,F ; d 0 ;stop here for 16bit MOVLW 0x0F ; 0x0F SWAPF x2,F ; f e ANDWF x2,W ; 0 e IORWF x3,F ; d e XORWF x2,F ; f 0 ;stop here for 24bit MOVLW 0x0F ; 0x0F SWAPF x1,F ; h g ANDWF x1,W ; 0 g IORWF x2,F ; f g XORWF x2,F ; h 0 ;repeat the 5 inst. pattern above for larger datatypes ENDM ; b c d e f g h 0 ; 32-bit unsigned Divide by 16 ; by Martin Sturm 2010 ; ; Right shift 4byte value by 4 with zero fill on left. ; 2 instructions shorter than unrolled RLF implementation ; 1 instruction faster than code generator ! ; Can be easily modified for 24, 16 bit versions ; ; 17 instructions, 17 cycles DIV_U32_BY16 MACRO x4,x3,x2,x1 ; x4 x3 x2 x1 W ; a b c d e f g h MOVLW 0xF0 ; 0xF0 ANDWF x1,F ; g 0 SWAPF x1,F ; 0 g SWAPF x2,F ; f e ANDWF x2,W ; f 0 IORWF x1,F ; f g XORWF x2,F ; 0 e ;stop here for 16bit MOVLW 0xF0 ; 0xF0 SWAPF x3,F ; d c ANDWF x3,W ; d 0 IORWF x2,F ; d e XORWF x3,F ; 0 c ;stop here for 24bit MOVLW 0xF0 ; 0xF0 SWAPF x4,F ; b a ANDWF x4,W ; b 0 IORWF x3,F ; b c XORWF x4,F ; 0 a ;repeat the 5 inst. pattern above for larger datatypes ENDM ; 0 a b c d e f g
;****************************************************************************** ;Multiply 32 bit number by 10 ; for PIC18F by VegiePete gmail ;Inline code, in place result ;FSR0 points to 32 bit number (low byte in low memory) ;Returns overflow byte in WREG, Z set if no overflow Times10: movf POSTINC0,w ;grab 1st digit, point to 2nd mullw 10 movf POSTDEC0,w ;rescue 2nd digit, point to first movff PRODL,POSTINC0 ;save result low, point to 2nd movff PRODH,INDF0 ;save result high mullw 10 movf PRODL,w ;grab result low byte addwf POSTINC0,f ;add to highest previous byte, point tp 3rd movlw 0 addwfc PRODH,f ;propagate the carry movf INDF0,w ;rescue 3rd digit movff PRODH,INDF0 ;save result high mullw 10 movf PRODL,w ;grab result low byte addwfc POSTINC0,f ;add to highest previous byte, plus any carry movlw 0 addwfc PRODH,f ;propagate the carry movf INDF0,w ;rescue 4th digit movff PRODH,INDF0 ;save result high mullw 10 ;don't care about overflow byte movf PRODL,w ;grab result low byte addwfc INDF0,w ;add to highest previous byte, plus any carry movlw 0 addwfc PRODH,w ;propagate the carry, load overflow return
.