/*============================================================*/
/*============================================================*/
/*==														==*/
/*==			Assembler Rendering Routines				==*/
/*==														==*/
/*============================================================*/
/*============================================================*/


/*========================================================  Includes  */

#include "Globals.h"
#include "UnivUtilities.h"
#include "RenderAsm4.h"


/*========================================================  Functions  */
/*========================================================  DropDoorToWorkMapAsm4  */

void DropDoorToWorkMapAsm4 (void)

{
	long			srcAddress, destAddress;

	srcAddress = theDoor.srcAddrs[theDoor.doorOpen][theDoor.doorState];
	destAddress = theDoor.workDestAddr[theDoor.doorOpen];
	
	asm 68000
	{
		move.l		srcAddress, a0		; src bitmap address
		move.l		destAddress, a1		; dest screen address
										; 0 = source   1 = dest
		move.w		partsRowBytes, d0	; next row offset
		move.w		workRowBytes, d1	; next row offset
		subi.w		#12, d0				; compensate for (a0)+ increment
		subi.w		#12, d1				; compensate for (a1)+ increment
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0),(a1)			; move row 1
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0),(a1)			; move row 2
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0),(a1)			; move row 3
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0),(a1)			; move row 4
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0),(a1)			; move row 5
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0),(a1)			; move row 6
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0),(a1)			; move row 7
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0),(a1)			; move row 8
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0),(a1)			; move row 9
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0),(a1)			; move row 10
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0),(a1)			; move row 11
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0),(a1)			; move row 12
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0),(a1)			; move row 13
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0),(a1)			; move row 14
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0),(a1)			; move row 15
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0),(a1)			; move row 16
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0),(a1)			; move row 17
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0),(a1)			; move row 18
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0),(a1)			; move row 19
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0),(a1)			; move row 20
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0),(a1)			; move row 21
	}
	
	destAddress = theDoor.backDestAddr[theDoor.doorOpen];
	asm 68000
	{
		move.l		srcAddress, a0		; src bitmap address
		move.l		destAddress, a1		; dest screen address
										; 0 = source   1 = dest
		move.w		partsRowBytes, d0	; next row offset
		move.w		backRowBytes, d1	; next row offset
		subi.w		#12, d0				; compensate for (a0)+ increment
		subi.w		#12, d1				; compensate for (a1)+ increment
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0),(a1)			; move row 1
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0),(a1)			; move row 2
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0),(a1)			; move row 3
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0),(a1)			; move row 4
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0),(a1)			; move row 5
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0),(a1)			; move row 6
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0),(a1)			; move row 7
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0),(a1)			; move row 8
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0),(a1)			; move row 9
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0),(a1)			; move row 10
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0),(a1)			; move row 11
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0),(a1)			; move row 12
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0),(a1)			; move row 13
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0),(a1)			; move row 14
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0),(a1)			; move row 15
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0),(a1)			; move row 16
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0),(a1)			; move row 17
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0),(a1)			; move row 18
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0),(a1)			; move row 19
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0),(a1)			; move row 20
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0),(a1)			; move row 21
	}
}


/*========================================================  MaskBoardCursorAsm4  */

void MaskBoardCursorAsm4 (long dest, short shiftRight)

{
	asm 68000
	{
		move.l		a2, -(sp)			; save a2 on stack
		move.l		d3, -(sp)			; save d3 on stack
		move.l		d4, -(sp)			; save d4 on stack
		move.l		d5, -(sp)			; save d5 on stack
		move.l		d6, -(sp)			; save d4 on stack
		move.l		d7, -(sp)			; save d5 on stack
		
		move.l		boardCursor.srcAddr, a0		; src bitmap address
		move.l		dest, a1					; dest screen address
		move.l		boardCursor.maskAddr, a2	; mask bitmap address
		
		move.w		partsRowBytes, d0	; next row offset (parts)
		subq.l		#4, d0				; compensate for little adds
		move.w		workRowBytes, d1	; next row offset
		subq.l		#4, d1				; compensate for little adds
		move.w		shiftRight, d2		; d2 will hold # of bits to shift right
		move.w		#32, d3				; prepare to subtract from 32
		sub.w		d2, d3				; calculate value to shift left in d3
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask/parts row diff.
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask/parts row diff.
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask/parts row diff.
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask/parts row diff.
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask/parts row diff.
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask/parts row diff.
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask/parts row diff.
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask/parts row diff.
										;
		move.l		(sp)+, d7			; restore d5 from stack
		move.l		(sp)+, d6			; restore d4 from stack
		move.l		(sp)+, d5			; restore d5 from stack
		move.l		(sp)+, d4			; restore d4 from stack
		move.l		(sp)+, d3			; restore d3 from stack
		move.l		(sp)+, a2			; restore a2 from stack
	}
}


/*========================================================  MaskBallAsm4  */

void MaskBallAsm4 (long dest, short shiftRight)

{
	asm 68000
	{
		move.l		a2, -(sp)			; save a2 on stack
		move.l		d3, -(sp)			; save d3 on stack
		move.l		d4, -(sp)			; save d4 on stack
		move.l		d5, -(sp)			; save d5 on stack
		move.l		d6, -(sp)			; save d4 on stack
		move.l		d7, -(sp)			; save d5 on stack
		
		move.l		ballSrcAddr, a0		; src bitmap address
		move.l		dest, a1			; dest screen address
		move.l		ballMaskAddr, a2	; mask bitmap address
		
		move.w		partsRowBytes, d0	; next row offset (parts)
		subq.l		#4, d0				; compensate for little adds
		move.w		workRowBytes, d1	; next row offset
		subq.l		#4, d1				; compensate for little adds
		move.w		shiftRight, d2		; d2 will hold # of bits to shift right
		move.w		#32, d3				; prepare to subtract from 32
		sub.w		d2, d3				; calculate value to shift left in d3
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask row 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; === prepare for next word
		move.l		(a2)+, d6			; move mask row 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src row 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the left
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
		move.l		(sp)+, d7			; restore d5 from stack
		move.l		(sp)+, d6			; restore d4 from stack
		move.l		(sp)+, d5			; restore d5 from stack
		move.l		(sp)+, d4			; restore d4 from stack
		move.l		(sp)+, d3			; restore d3 from stack
		move.l		(sp)+, a2			; restore a2 from stack
	}
}


/*========================================================  MaskPlayerAsm4  */

void MaskPlayerAsm4 (long src, long dest, long mask, short shiftRight)

{
	asm 68000
	{
		move.l		a2, -(sp)			; save a2 on stack
		move.l		d3, -(sp)			; save d3 on stack
		move.l		d4, -(sp)			; save d4 on stack
		move.l		d5, -(sp)			; save d5 on stack
		move.l		d6, -(sp)			; save d4 on stack
		move.l		d7, -(sp)			; save d5 on stack
		
		move.l		src, a0				; src bitmap address
		move.l		dest, a1			; dest screen address
		move.l		mask, a2			; mask bitmap address
		
		move.w		partsRowBytes, d0	; next row offset (parts)
		sub.w		#12, d0				; compensate for little adds
		move.w		workRowBytes, d1	; next row offset
		sub.w		#12, d1				; compensate for little adds
		move.w		shiftRight, d2		; d2 will hold # of bits to shift right
		move.w		#32, d3				; prepare to subtract from 32
		sub.w		d2, d3				; calculate value to shift left in d3
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										; === increment pointers
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		adda.w		d0, a2				; add row offset to mask
		suba.w		#40, a2				; compensate for mask row bytes difference
										;
										; === BEGIN a ROW
		move.l		(a2), d4			; move mask seg 1
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src seg 1
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask bits
		and.l		d4, (a1)			; and mask to dest (create hole)
		or.l		d5, (a1)+			; or src to dest (fill hole)
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 1
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 1
		lsl.l		d3, d7				; shift d7 logically d3 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask seg 2
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 2
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 2
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 2
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 3
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 3
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)+			; or src to dest
										; <<< prepare for next word
		move.l		(a2)+, d6			; move mask seg 3
		lsl.l		d3, d6				; shift d6 logically d3 bits to the left
		move.l		(a0)+, d7			; move src seg 3
		lsl.l		d3, d7				; shift d7 logically d2 bits to the right
		and.l		d6, d7				; clear trash bits
										; === load next word in row
		move.l		(a2), d4			; move mask row 4
		lsr.l		d2, d4				; shift d4 logically d2 bits to the right
		move.l		(a0), d5			; move src row 4
		lsr.l		d2, d5				; shift d5 logically d2 bits to the right
										; === add overhang from last word in
		or.l		d6, d4				; or old mask to new mask
		or.l		d7, d5				; or old src to new src
		and.l		d4, d5				; clear trash bits
		not.l		d4					; invert mask
		and.l		d4, (a1)			; and mask to dest
		or.l		d5, (a1)			; or src to dest
										;
		move.l		(sp)+, d7			; restore d7 from stack
		move.l		(sp)+, d6			; restore d6 from stack
		move.l		(sp)+, d5			; restore d5 from stack
		move.l		(sp)+, d4			; restore d4 from stack
		move.l		(sp)+, d3			; restore d3 from stack
		move.l		(sp)+, a2			; restore a2 from stack
	}
}


/*========================================================  DumpWholePlayerAsm4  */

void DumpWholePlayerAsm4 (long src, long dest)

{
	asm 68000
	{
		move.l		src, a0				; src bitmap address
		move.l		dest, a1			; dest screen address
										; 0 = source   1 = dest
		move.w		workRowBytes, d0	; next row offset
		move.w		screenRowBytes, d1	; next row offset
		subi.w		#16, d0				; compensate for (a0)+ inc. (4 x +'s)
		subi.w		#16, d1				; compensate for (a1)+ increment
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0),(a1)			; move row 1
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0),(a1)			; move row 2
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0),(a1)			; move row 3
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0),(a1)			; move row 4
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0),(a1)			; move row 5
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0),(a1)			; move row 6
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0),(a1)			; move row 7
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0),(a1)			; move row 8
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0),(a1)			; move row 9
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0),(a1)			; move row 10
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0),(a1)			; move row 11
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0),(a1)			; move row 12
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0),(a1)			; move row 13
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0),(a1)			; move row 14
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0),(a1)			; move row 15
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0),(a1)			; move row 16
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0),(a1)			; move row 17
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0),(a1)			; move row 18
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0),(a1)			; move row 19
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0),(a1)			; move row 20
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0),(a1)			; move row 21
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 22
		move.l		(a0)+,(a1)+			; move row 22
		move.l		(a0)+,(a1)+			; move row 22
		move.l		(a0)+,(a1)+			; move row 22
		move.l		(a0),(a1)			; move row 22
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 23
		move.l		(a0)+,(a1)+			; move row 23
		move.l		(a0)+,(a1)+			; move row 23
		move.l		(a0)+,(a1)+			; move row 23
		move.l		(a0),(a1)			; move row 23
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 24
		move.l		(a0)+,(a1)+			; move row 24
		move.l		(a0)+,(a1)+			; move row 24
		move.l		(a0)+,(a1)+			; move row 24
		move.l		(a0),(a1)			; move row 24
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 25
		move.l		(a0)+,(a1)+			; move row 25
		move.l		(a0)+,(a1)+			; move row 25
		move.l		(a0)+,(a1)+			; move row 25
		move.l		(a0),(a1)			; move row 25
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 26
		move.l		(a0)+,(a1)+			; move row 26
		move.l		(a0)+,(a1)+			; move row 26
		move.l		(a0)+,(a1)+			; move row 26
		move.l		(a0),(a1)			; move row 26
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 27
		move.l		(a0)+,(a1)+			; move row 27
		move.l		(a0)+,(a1)+			; move row 27
		move.l		(a0)+,(a1)+			; move row 27
		move.l		(a0),(a1)			; move row 27
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 28
		move.l		(a0)+,(a1)+			; move row 28
		move.l		(a0)+,(a1)+			; move row 28
		move.l		(a0)+,(a1)+			; move row 28
		move.l		(a0),(a1)			; move row 28
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 29
		move.l		(a0)+,(a1)+			; move row 29
		move.l		(a0)+,(a1)+			; move row 29
		move.l		(a0)+,(a1)+			; move row 29
		move.l		(a0),(a1)			; move row 29
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 30
		move.l		(a0)+,(a1)+			; move row 30
		move.l		(a0)+,(a1)+			; move row 30
		move.l		(a0)+,(a1)+			; move row 30
		move.l		(a0),(a1)			; move row 30
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 31
		move.l		(a0)+,(a1)+			; move row 31
		move.l		(a0)+,(a1)+			; move row 31
		move.l		(a0)+,(a1)+			; move row 31
		move.l		(a0),(a1)			; move row 31
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 32
		move.l		(a0)+,(a1)+			; move row 32
		move.l		(a0)+,(a1)+			; move row 32
		move.l		(a0)+,(a1)+			; move row 32
		move.l		(a0),(a1)			; move row 32
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 33
		move.l		(a0)+,(a1)+			; move row 33
		move.l		(a0)+,(a1)+			; move row 33
		move.l		(a0)+,(a1)+			; move row 33
		move.l		(a0),(a1)			; move row 33
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 34
		move.l		(a0)+,(a1)+			; move row 34
		move.l		(a0)+,(a1)+			; move row 34
		move.l		(a0)+,(a1)+			; move row 34
		move.l		(a0),(a1)			; move row 34
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 35
		move.l		(a0)+,(a1)+			; move row 35
		move.l		(a0)+,(a1)+			; move row 35
		move.l		(a0)+,(a1)+			; move row 35
		move.l		(a0),(a1)			; move row 35
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 36
		move.l		(a0)+,(a1)+			; move row 36
		move.l		(a0)+,(a1)+			; move row 36
		move.l		(a0)+,(a1)+			; move row 36
		move.l		(a0),(a1)			; move row 36
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 37
		move.l		(a0)+,(a1)+			; move row 37
		move.l		(a0)+,(a1)+			; move row 37
		move.l		(a0)+,(a1)+			; move row 37
		move.l		(a0),(a1)			; move row 37
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 38
		move.l		(a0)+,(a1)+			; move row 38
		move.l		(a0)+,(a1)+			; move row 38
		move.l		(a0)+,(a1)+			; move row 38
		move.l		(a0),(a1)			; move row 38
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 39
		move.l		(a0)+,(a1)+			; move row 39
		move.l		(a0)+,(a1)+			; move row 39
		move.l		(a0)+,(a1)+			; move row 39
		move.l		(a0),(a1)			; move row 39
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 40
		move.l		(a0)+,(a1)+			; move row 40
		move.l		(a0)+,(a1)+			; move row 40
		move.l		(a0)+,(a1)+			; move row 40
		move.l		(a0),(a1)			; move row 40
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 41
		move.l		(a0)+,(a1)+			; move row 41
		move.l		(a0)+,(a1)+			; move row 41
		move.l		(a0)+,(a1)+			; move row 41
		move.l		(a0),(a1)			; move row 41
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 42
		move.l		(a0)+,(a1)+			; move row 42
		move.l		(a0)+,(a1)+			; move row 42
		move.l		(a0)+,(a1)+			; move row 42
		move.l		(a0),(a1)			; move row 42
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 43
		move.l		(a0)+,(a1)+			; move row 43
		move.l		(a0)+,(a1)+			; move row 43
		move.l		(a0)+,(a1)+			; move row 43
		move.l		(a0),(a1)			; move row 43
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 44
		move.l		(a0)+,(a1)+			; move row 44
		move.l		(a0)+,(a1)+			; move row 44
		move.l		(a0)+,(a1)+			; move row 44
		move.l		(a0),(a1)			; move row 44
	}
}


/*========================================================  DumpWholeBallAsm4  */

void DumpWholeBallAsm4 (long src, long dest)

{
	asm 68000
	{
		move.l		src, a0				; src bitmap address
		move.l		dest, a1			; dest screen address
										; 0 = source   1 = dest
		move.w		workRowBytes, d0	; next row offset
		move.w		screenRowBytes, d1	; next row offset
		subq.w		#8, d0				; compensate for (a0)+ inc. (4 x +'s)
		subq.w		#8, d1				; compensate for (a1)+ increment
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0),(a1)			; move row 1
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0),(a1)			; move row 2
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0),(a1)			; move row 3
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0),(a1)			; move row 4
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0),(a1)			; move row 5
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0),(a1)			; move row 6
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0),(a1)			; move row 7
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0),(a1)			; move row 8
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0),(a1)			; move row 9
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0),(a1)			; move row 10
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0),(a1)			; move row 11
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0),(a1)			; move row 12
	}
}


/*========================================================  DumpCursorAsm4  */

void DumpCursorAsm4 (long src, long dest)

{
	asm 68000
	{
		move.l		src, a0				; src bitmap address
		move.l		dest, a1			; dest screen address
										; 0 = source   1 = dest
		move.w		workRowBytes, d0	; next row offset
		move.w		screenRowBytes, d1	; next row offset
		subq.w		#8, d0				; compensate for (a0)+ inc. (4 x +'s)
		subq.w		#8, d1				; compensate for (a1)+ increment
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0),(a1)			; move row 1
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0),(a1)			; move row 2
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0),(a1)			; move row 3
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0),(a1)			; move row 4
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0),(a1)			; move row 5
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0),(a1)			; move row 6
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0),(a1)			; move row 7
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0),(a1)			; move row 8
	}
}


/*========================================================  DumpDoorToScreenAsm4  */

void DumpDoorToScreenAsm4 (void)

{
	long			srcAddress, destAddress;

	srcAddress = theDoor.workDestAddr[theDoor.doorOpen];
	destAddress = theDoor.screenDestAddr[theDoor.doorOpen];
	
	asm 68000
	{
		move.l		srcAddress, a0		; src bitmap address
		move.l		destAddress, a1		; dest screen address
										; 0 = source   1 = dest
		move.w		workRowBytes, d0	; next row offset
		move.w		screenRowBytes, d1	; next row offset
		subi.w		#12, d0				; compensate for (a0)+ increment
		subi.w		#12, d1				; compensate for (a1)+ increment
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0),(a1)			; move row 1
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0),(a1)			; move row 2
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0),(a1)			; move row 3
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0),(a1)			; move row 4
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0),(a1)			; move row 5
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0),(a1)			; move row 6
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0),(a1)			; move row 7
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0),(a1)			; move row 8
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0),(a1)			; move row 9
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0),(a1)			; move row 10
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0),(a1)			; move row 11
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0),(a1)			; move row 12
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0),(a1)			; move row 13
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0),(a1)			; move row 14
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0),(a1)			; move row 15
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0),(a1)			; move row 16
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0),(a1)			; move row 17
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0),(a1)			; move row 18
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0),(a1)			; move row 19
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0),(a1)			; move row 20
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0),(a1)			; move row 21
	}
}


/*========================================================  PatchPlayerAsm4  */

void PatchPlayerAsm4 (long src, long dest)

{
	asm 68000
	{
		move.l		src, a0				; src bitmap address
		move.l		dest, a1			; dest screen address
										; 0 = source   1 = dest
		move.w		backRowBytes, d0	; next row offset
		move.w		workRowBytes, d1	; next row offset
		subi.w		#16, d0				; compensate for (a0)+ inc. (4 x +'s)
		subi.w		#16, d1				; compensate for (a1)+ increment
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.w		(a0),(a1)			; move row 1
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.w		(a0),(a1)			; move row 2
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.w		(a0),(a1)			; move row 3
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.w		(a0),(a1)			; move row 4
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.w		(a0),(a1)			; move row 5
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.w		(a0),(a1)			; move row 6
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.w		(a0),(a1)			; move row 7
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.w		(a0),(a1)			; move row 8
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.w		(a0),(a1)			; move row 9
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.w		(a0),(a1)			; move row 10
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.w		(a0),(a1)			; move row 11
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.w		(a0),(a1)			; move row 12
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.l		(a0)+,(a1)+			; move row 13
		move.w		(a0),(a1)			; move row 13
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.l		(a0)+,(a1)+			; move row 14
		move.w		(a0),(a1)			; move row 14
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.l		(a0)+,(a1)+			; move row 15
		move.w		(a0),(a1)			; move row 15
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.l		(a0)+,(a1)+			; move row 16
		move.w		(a0),(a1)			; move row 16
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.l		(a0)+,(a1)+			; move row 17
		move.w		(a0),(a1)			; move row 17
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.l		(a0)+,(a1)+			; move row 18
		move.w		(a0),(a1)			; move row 18
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.l		(a0)+,(a1)+			; move row 19
		move.w		(a0),(a1)			; move row 19
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.l		(a0)+,(a1)+			; move row 20
		move.w		(a0),(a1)			; move row 20
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.l		(a0)+,(a1)+			; move row 21
		move.w		(a0),(a1)			; move row 21
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 22
		move.l		(a0)+,(a1)+			; move row 22
		move.l		(a0)+,(a1)+			; move row 22
		move.l		(a0)+,(a1)+			; move row 22
		move.w		(a0),(a1)			; move row 22
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 23
		move.l		(a0)+,(a1)+			; move row 23
		move.l		(a0)+,(a1)+			; move row 23
		move.l		(a0)+,(a1)+			; move row 23
		move.w		(a0),(a1)			; move row 23
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 24
		move.l		(a0)+,(a1)+			; move row 24
		move.l		(a0)+,(a1)+			; move row 24
		move.l		(a0)+,(a1)+			; move row 24
		move.w		(a0),(a1)			; move row 24
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 25
		move.l		(a0)+,(a1)+			; move row 25
		move.l		(a0)+,(a1)+			; move row 25
		move.l		(a0)+,(a1)+			; move row 25
		move.w		(a0),(a1)			; move row 25
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 26
		move.l		(a0)+,(a1)+			; move row 26
		move.l		(a0)+,(a1)+			; move row 26
		move.l		(a0)+,(a1)+			; move row 26
		move.w		(a0),(a1)			; move row 26
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 27
		move.l		(a0)+,(a1)+			; move row 27
		move.l		(a0)+,(a1)+			; move row 27
		move.l		(a0)+,(a1)+			; move row 27
		move.w		(a0),(a1)			; move row 27
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 28
		move.l		(a0)+,(a1)+			; move row 28
		move.l		(a0)+,(a1)+			; move row 28
		move.l		(a0)+,(a1)+			; move row 28
		move.w		(a0),(a1)			; move row 28
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 29
		move.l		(a0)+,(a1)+			; move row 29
		move.l		(a0)+,(a1)+			; move row 29
		move.l		(a0)+,(a1)+			; move row 29
		move.w		(a0),(a1)			; move row 29
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 30
		move.l		(a0)+,(a1)+			; move row 30
		move.l		(a0)+,(a1)+			; move row 30
		move.l		(a0)+,(a1)+			; move row 30
		move.w		(a0),(a1)			; move row 30
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 31
		move.l		(a0)+,(a1)+			; move row 31
		move.l		(a0)+,(a1)+			; move row 31
		move.l		(a0)+,(a1)+			; move row 31
		move.w		(a0),(a1)			; move row 31
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 32
		move.l		(a0)+,(a1)+			; move row 32
		move.l		(a0)+,(a1)+			; move row 32
		move.l		(a0)+,(a1)+			; move row 32
		move.w		(a0),(a1)			; move row 32
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 33
		move.l		(a0)+,(a1)+			; move row 33
		move.l		(a0)+,(a1)+			; move row 33
		move.l		(a0)+,(a1)+			; move row 33
		move.w		(a0),(a1)			; move row 33
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 34
		move.l		(a0)+,(a1)+			; move row 34
		move.l		(a0)+,(a1)+			; move row 34
		move.l		(a0)+,(a1)+			; move row 34
		move.w		(a0),(a1)			; move row 34
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 35
		move.l		(a0)+,(a1)+			; move row 35
		move.l		(a0)+,(a1)+			; move row 35
		move.l		(a0)+,(a1)+			; move row 35
		move.w		(a0),(a1)			; move row 35
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 36
		move.l		(a0)+,(a1)+			; move row 36
		move.l		(a0)+,(a1)+			; move row 36
		move.l		(a0)+,(a1)+			; move row 36
		move.w		(a0),(a1)			; move row 36
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 37
		move.l		(a0)+,(a1)+			; move row 37
		move.l		(a0)+,(a1)+			; move row 37
		move.l		(a0)+,(a1)+			; move row 37
		move.w		(a0),(a1)			; move row 37
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 38
		move.l		(a0)+,(a1)+			; move row 38
		move.l		(a0)+,(a1)+			; move row 38
		move.l		(a0)+,(a1)+			; move row 38
		move.w		(a0),(a1)			; move row 38
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 39
		move.l		(a0)+,(a1)+			; move row 39
		move.l		(a0)+,(a1)+			; move row 39
		move.l		(a0)+,(a1)+			; move row 39
		move.w		(a0),(a1)			; move row 39
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 40
		move.l		(a0)+,(a1)+			; move row 40
		move.l		(a0)+,(a1)+			; move row 40
		move.l		(a0)+,(a1)+			; move row 40
		move.w		(a0),(a1)			; move row 40
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 41
		move.l		(a0)+,(a1)+			; move row 41
		move.l		(a0)+,(a1)+			; move row 41
		move.l		(a0)+,(a1)+			; move row 41
		move.w		(a0),(a1)			; move row 41
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 42
		move.l		(a0)+,(a1)+			; move row 42
		move.l		(a0)+,(a1)+			; move row 42
		move.l		(a0)+,(a1)+			; move row 42
		move.w		(a0),(a1)			; move row 42
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 43
		move.l		(a0)+,(a1)+			; move row 43
		move.l		(a0)+,(a1)+			; move row 43
		move.l		(a0)+,(a1)+			; move row 43
		move.w		(a0),(a1)			; move row 43
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 44
		move.l		(a0)+,(a1)+			; move row 44
		move.l		(a0)+,(a1)+			; move row 44
		move.l		(a0)+,(a1)+			; move row 44
		move.w		(a0),(a1)			; move row 44
	}
}


/*========================================================  PatchBallAsm4  */

void PatchBallAsm4 (long src, long dest)

{
	asm 68000
	{
		move.l		src, a0				; src bitmap address
		move.l		dest, a1			; dest screen address
										; 0 = source   1 = dest
		move.w		backRowBytes, d0	; next row offset
		move.w		workRowBytes, d1	; next row offset
		subq.w		#8, d0				; compensate for (a0)+ inc. (4 x +'s)
		subq.w		#8, d1				; compensate for (a1)+ increment
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.w		(a0),(a1)			; move row 1
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.w		(a0),(a1)			; move row 2
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.w		(a0),(a1)			; move row 3
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.w		(a0),(a1)			; move row 4
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.w		(a0),(a1)			; move row 5
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.w		(a0),(a1)			; move row 6
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.w		(a0),(a1)			; move row 7
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.w		(a0),(a1)			; move row 8
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 9
		move.l		(a0)+,(a1)+			; move row 9
		move.w		(a0),(a1)			; move row 9
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 10
		move.l		(a0)+,(a1)+			; move row 10
		move.w		(a0),(a1)			; move row 10
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 11
		move.l		(a0)+,(a1)+			; move row 11
		move.w		(a0),(a1)			; move row 11
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 12
		move.l		(a0)+,(a1)+			; move row 12
		move.w		(a0),(a1)			; move row 12
	}
}


/*========================================================  PatchCursorAsm4  */

void PatchCursorAsm4 (long src, long dest)

{
	asm 68000
	{
		move.l		src, a0				; src bitmap address
		move.l		dest, a1			; dest screen address
										; 0 = source   1 = dest
		move.w		backRowBytes, d0	; next row offset
		move.w		workRowBytes, d1	; next row offset
		subq.w		#8, d0				; compensate for (a0)+ inc. (4 x +'s)
		subq.w		#8, d1				; compensate for (a1)+ increment
		move.l		(a0)+,(a1)+			; move row 1
		move.l		(a0)+,(a1)+			; move row 1
		move.w		(a0),(a1)			; move row 1
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 2
		move.l		(a0)+,(a1)+			; move row 2
		move.w		(a0),(a1)			; move row 2
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 3
		move.l		(a0)+,(a1)+			; move row 3
		move.w		(a0),(a1)			; move row 3
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 4
		move.l		(a0)+,(a1)+			; move row 4
		move.w		(a0),(a1)			; move row 4
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 5
		move.l		(a0)+,(a1)+			; move row 5
		move.w		(a0),(a1)			; move row 5
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 6
		move.l		(a0)+,(a1)+			; move row 6
		move.w		(a0),(a1)			; move row 6
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 7
		move.l		(a0)+,(a1)+			; move row 7
		move.w		(a0),(a1)			; move row 7
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		move.l		(a0)+,(a1)+			; move row 8
		move.l		(a0)+,(a1)+			; move row 8
		move.w		(a0),(a1)			; move row 8
	}
}


/*========================================================  RenderSceneAsm4  */

void RenderSceneAsm4 (void)

{
	long			srcAddress, destAddress;
	long			ballOffIs, playerOffIs, opponentOffIs, cursorOffIs;
	Point			offsetPt;
	char			theMode;
	
	ballOffIs = workRowOffsets[theBall.isRect.top] + 
			(long)(((short)theBall.isRect.left & 0xFFFC) >> 1);
	
	playerOffIs = workRowOffsets[thePlayer.isRect.top] + 
			(long)(((short)thePlayer.isRect.left & 0xFFFC) >> 1);
	
	opponentOffIs = workRowOffsets[theOpponent.isRect.top] + 
			(long)(((short)theOpponent.isRect.left & 0xFFFC) >> 1);;
	
	offsetPt.h = 0;
	offsetPt.v = 0;
	ShieldCursor(&screenRect, offsetPt);
	
	theMode = true32b;
	SwapMMUMode(&theMode);
	
	if (theDoor.stateChanged)
		DropDoorToWorkMapAsm4();
	
	if ((showBoardCursor) && (!disableBoardCursor) && (drawThisFrame))
	{
		cursorOffIs = workRowOffsets[boardCursor.isRect.top] + 
				(long)(((short)boardCursor.isRect.left & 0xFFFC) >> 1);
		destAddress = (long)offWorkPix + cursorOffIs;
		MaskBoardCursorAsm4 (destAddress, (boardCursor.isRect.left % 4) * 4);
	}
	
	if ((theBall.mode == kBallRolling) && (!theBall.dontDraw) && (drawThisFrame))
	{
		destAddress = (long)offWorkPix + ballOffIs;
		MaskBallAsm4 (destAddress, (theBall.isRect.left % 4) * 4);
	}

/* mask the bastards to the work map - back to front! */
	
	if (playerInBack)
	{
		if ((thePlayer.mode != kInStasis) && (drawThisFrame))
		{
			destAddress = (long)offWorkPix + playerOffIs;
			MaskPlayerAsm4(thePlayer.srcAddr, destAddress, thePlayer.maskAddr, 
					(thePlayer.isRect.left % 4) * 4);
		}
		if ((theOpponent.mode != kInStasis) && (drawThisFrame))
		{
			destAddress = (long)offWorkPix + opponentOffIs;
			MaskPlayerAsm4(theOpponent.srcAddr, destAddress, theOpponent.maskAddr, 
					(theOpponent.isRect.left % 4) * 4);
		}
	}
	else
	{
		if ((theOpponent.mode != kInStasis) && (drawThisFrame))
		{
			destAddress = (long)offWorkPix + opponentOffIs;
			MaskPlayerAsm4(theOpponent.srcAddr, destAddress, theOpponent.maskAddr, 
					(theOpponent.isRect.left % 4) * 4);
		}
		if ((thePlayer.mode != kInStasis) && (drawThisFrame))
		{
			destAddress = (long)offWorkPix + playerOffIs;
			MaskPlayerAsm4(thePlayer.srcAddr, destAddress, thePlayer.maskAddr, 
					(thePlayer.isRect.left % 4) * 4);
		}
	}

/* blast them to the screen */

	if ((thePlayer.mode != kInStasis) && (drawThisFrame))
	{
		srcAddress = (long)offWorkPix + playerOffIs;
		destAddress = screenRowAddrs[thePlayer.isRect.top] + 
				(long)(((short)thePlayer.isRect.left & 0xFFFC) >> 1);
		DumpWholePlayerAsm4(srcAddress, destAddress);
		
		srcAddress = (long)offWorkPix + workRowOffsets[thePlayer.wasRect.top] + 
				(long)(((short)thePlayer.wasRect.left & 0xFFFC) >> 1);
		destAddress = screenRowAddrs[thePlayer.wasRect.top] + 
				(long)(((short)thePlayer.wasRect.left & 0xFFFC) >> 1);
		DumpWholePlayerAsm4(srcAddress, destAddress);
	}
	
	if ((theOpponent.mode != kInStasis) && (drawThisFrame))
	{
		srcAddress = (long)offWorkPix + opponentOffIs;
		destAddress = screenRowAddrs[theOpponent.isRect.top] + 
				(long)(((short)theOpponent.isRect.left & 0xFFFC) >> 1);
		DumpWholePlayerAsm4(srcAddress, destAddress);
		
		srcAddress = (long)offWorkPix + workRowOffsets[theOpponent.wasRect.top] + 
				(long)(((short)theOpponent.wasRect.left & 0xFFFC) >> 1);
		destAddress = screenRowAddrs[theOpponent.wasRect.top] + 
				(long)(((short)theOpponent.wasRect.left & 0xFFFC) >> 1);
		DumpWholePlayerAsm4(srcAddress, destAddress);
	}
	if ((theBall.mode == kBallRolling) && (drawThisFrame))
	{
		srcAddress = (long)offWorkPix + ballOffIs;
		destAddress = screenRowAddrs[theBall.isRect.top] + 
				(long)(((short)theBall.isRect.left & 0xFFFC) >> 1);
		DumpWholeBallAsm4(srcAddress, destAddress);
		
		srcAddress = (long)offWorkPix + workRowOffsets[theBall.wasRect.top] + 
				(long)(((short)theBall.wasRect.left & 0xFFFC) >> 1);
		destAddress = screenRowAddrs[theBall.wasRect.top] + 
				(long)(((short)theBall.wasRect.left & 0xFFFC) >> 1);
		DumpWholeBallAsm4(srcAddress, destAddress);
	}
	else if (theBall.eraseTheBall)
	{
		srcAddress = (long)offWorkPix + workRowOffsets[theBall.eraser.top] + 
				(long)(((short)theBall.eraser.left & 0xFFFC) >> 1);
		destAddress = screenRowAddrs[theBall.eraser.top] + 
				(long)(((short)theBall.eraser.left & 0xFFFC) >> 1);
		DumpWholeBallAsm4(srcAddress, destAddress);
		
		theBall.eraseTheBall = FALSE;
	}
	
	if ((showBoardCursor) && (!disableBoardCursor) && (drawThisFrame))
	{
		srcAddress = (long)offWorkPix + cursorOffIs;
		destAddress = screenRowAddrs[boardCursor.isRect.top] + 
				(long)(((short)boardCursor.isRect.left & 0xFFFC) >> 1);
		DumpCursorAsm4(srcAddress, destAddress);
		
		srcAddress = (long)offWorkPix + workRowOffsets[boardCursor.wasRect.top] + 
				(long)(((short)boardCursor.wasRect.left & 0xFFFC) >> 1);
		destAddress = screenRowAddrs[boardCursor.wasRect.top] + 
				(long)(((short)boardCursor.wasRect.left & 0xFFFC) >> 1);
		DumpCursorAsm4(srcAddress, destAddress);
	}
	
	if (theDoor.stateChanged)
		DumpDoorToScreenAsm4();

/* plug up the holes on the work map */

	if ((thePlayer.mode != kInStasis) && (drawThisFrame))
	{
		srcAddress = (long)offBackPix + playerOffIs;
		destAddress = (long)offWorkPix + playerOffIs;
		PatchPlayerAsm4(srcAddress, destAddress);
	}
	
	if ((theOpponent.mode != kInStasis) && (drawThisFrame))
	{
		srcAddress = (long)offBackPix + opponentOffIs;
		destAddress = (long)offWorkPix + opponentOffIs;
		PatchPlayerAsm4(srcAddress, destAddress);
	}
	
	if ((theBall.mode == kBallRolling) && (drawThisFrame))
	{
		srcAddress = (long)offBackPix + ballOffIs;
		destAddress = (long)offWorkPix + ballOffIs;
		PatchBallAsm4(srcAddress, destAddress);
	}
	
	if ((showBoardCursor) && (!disableBoardCursor) && (drawThisFrame))
	{
		srcAddress = (long)offBackPix + cursorOffIs;
		destAddress = (long)offWorkPix + cursorOffIs;
		PatchCursorAsm4(srcAddress, destAddress);
	}
	
	SwapMMUMode(&theMode);
	ShowCursor();
}


/*========================================================  ReplayWorkToMainAsm4  */

void ReplayWorkToMainAsm4 (void)

{
	Point			offsetPt;
	char			theMode;
	
	offsetPt.h = 0;
	offsetPt.v = 0;
	ShieldCursor(&screenRect, offsetPt);
	
	theMode = true32b;
	SwapMMUMode(&theMode);
	
	asm 68000
	{
		move.l		replaySrc, a0		; src bitmap address
		move.l		replayDest, a1		; dest screen address
										; 0 = source   1 = dest
		move.w		workRowBytes, d0	; next row offset
		move.w		screenRowBytes, d1	; next row offset
		subi.w		#92, d0				; compensate for (a0)+ increment
		subi.w		#92, d1				; compensate for (a1)+ increment
		move.w		#128, d2			; d2 will count rows
	@1	move.l		(a0)+,(a1)+			; move long 1
		move.l		(a0)+,(a1)+			; move long 2
		move.l		(a0)+,(a1)+			; move long 3
		move.l		(a0)+,(a1)+			; move long 4
		move.l		(a0)+,(a1)+			; move long 5
		move.l		(a0)+,(a1)+			; move long 6
		move.l		(a0)+,(a1)+			; move long 7
		move.l		(a0)+,(a1)+			; move long 8
		move.l		(a0)+,(a1)+			; move long 9
		move.l		(a0)+,(a1)+			; move long 10
		move.l		(a0)+,(a1)+			; move long 11
		move.l		(a0)+,(a1)+			; move long 12
		move.l		(a0)+,(a1)+			; move long 13
		move.l		(a0)+,(a1)+			; move long 14
		move.l		(a0)+,(a1)+			; move long 15
		move.l		(a0)+,(a1)+			; move long 16
		move.l		(a0)+,(a1)+			; move long 17
		move.l		(a0)+,(a1)+			; move long 18
		move.l		(a0)+,(a1)+			; move long 19
		move.l		(a0)+,(a1)+			; move long 20
		move.l		(a0)+,(a1)+			; move long 21
		move.l		(a0)+,(a1)+			; move long 22
		move.l		(a0)+,(a1)+			; move long 23
		move.l		(a0),(a1)			; move long 24
		adda.w		d0, a0				; add row offset to src
		adda.w		d1, a1				; add row offset to dest
		subq.w		#1, d2				; decrement row counter
		bne			@1					; do another row until d2 = 0
	}
	
	SwapMMUMode(&theMode);
	ShowCursor();
}