incdir "include:"
    include "rtgmaster/rtgc2p.i"
    include "rtgmaster/rtgmaster.i"
    include "rtgmaster/rtgsublibs.i"
    include "rtgmaster/rtgAMI.i"
    include "intuition/screens.i"
    include "exec/memory.i"
    moveq #0,d0
    rts
.s  dc.l "c2p!"
    dc.l .s
    dc.l 1
    dc.l Structure

Structure
    dc.w CI_256+CI_128+CI_64+CI_EHB+CI_32+CI_16+CI_8+CI_4+CI_2
    dc.w CI_68040+CI_68030D+CI_68020D
    dc.w 0
    dc.b 0
    dc.b 0
    dc.l c2p_1x1
    dc.w 320
    dc.w 200
    dc.w CI_Fixed
    dc.b 'AMI '
    dc.l Name
    dc.l Init
    dc.l Expunge
    dc.l chunky2planar
    dc.l 0
    dc.l 0
    dc.l 0
    dc.b 0
    dc.b 0,0,0 ; Test
    even

Init:
    movem.l a0/a1/a6/d0/d1,-(sp)
    move.l #0,rs_c2pdata(a0)
    movem.l (sp)+,a0/a1/a6/d0/d1
    move.l #0,d0
    rts
Error:
Name: dc.b 'c2p for 040 with 320x200 *ONLY*',0
    even

Expunge:
    rts

width           equ     320             ; must be multiple of 32
height          equ     200
plsiz           equ     (width/8)*height


merge   MACRO in1,in2,tmp3,tmp4,mask,shift
        ;               \1 = abqr
        ;               \2 = ijyz
        move.l  \2,\4
        move.l  #\5,\3
        and.l   \3,\2   \2 = 0j0z
        and.l   \1,\3   \3 = 0b0r
        eor.l   \3,\1   \1 = a0q0
        eor.l   \2,\4   \4 = i0y0
        IFEQ    \6-1
        add.l   \3,\3
        ELSE
        lsl.l   #\6,\3  \3 = b0r0
        ENDC
        lsr.l   #\6,\4  \4 = 0i0y
        or.l    \3,\2   \2 = bjrz
        or.l    \4,\1   \1 = aiqy
        ENDM



chunky2planar:
                movem.l d0/d3/a3/d7,-(sp)
                move.l #0,d7
                cmp.l #c2p_1x1,d0
                beq .OK
                cmp.l #c2p_Best,d0
                beq .OK
                cmp.l #c2p_BestD,d0
                beq .OK
                cmp.l #c2p_Fastest,d0
                beq .OK
                cmp.l #c2p_FastestD,d0
                beq .OK
                cmp.l #c2p_Selected,d0
                beq .Selected
                cmp.l #c2p_SelectedD,d0
                beq .Selected
                move.l #c2p_warn_Wrong_Pixelmode,d7
                bra .OK
.Selected:
                move.l rs_c2pcurr(a3),d3
                cmp.l #c2p_1x1,d3
                beq .OK
                move.l #c2p_warn_Wrong_Pixelmode,d7
                bra .OK
.OK:
                cmp.l #1,.Merk2
                beq .Mistake
                cmp.l #1,.Merk
                beq .OK3
                move.l #1,.Merk
                move.l #0,.Merk2
                move.l a3,-(sp)
                move.l rsAMI_ScreenHandle(a3),a3
                cmp.w #320,sc_Width(a3)
                bne .Mistake
                cmp.w #200,sc_Height(a3)
                bne .Mistake
                move.l (sp)+,a3
                lea rsAMI_Bitmap1(a3),a3
                sub.l d3,d3
                bra .OK3
.Merk: dc.l 0
.Merk2: dc.l 0
.Mistake:
                move.l (sp)+,a3
                move.l #1,.Merk2
                movem.l (sp)+,d0/d3/a3/a7
                move.l d7,d0
                movem.l d2/a6,-(sp)
                move.l $4,a6
                move.l #0,a1
                jsr -294(a6)
                move.l d0,a1
                movem.l (sp)+,d0/a6
                move.l a6,-(sp)
                move.l $4,a6
                move.l d0,d1
                sub.l d0,d0
                bset d1,d0
                jsr -324(a6)
                move.l (sp)+,a6
                move.l #c2p_err_internal,d0
                rts
.OK3:
                movem.l d2-d7/a2-a6,-(sp)
                jsr _chunky2planar
                movem.l (sp)+,d2-d7/a2-a6
                bra Raus
_chunky2planar:
                jmp     next
next
        ; round down address of c2p
        lea     c2p(pc),a0
        move.l  a0,d0
        and.b   #%11110000,d0
        move.l  d0,a1

        ; patch jmp
        move.l  d0,_chunky2planar+2
        move.w  #(end-c2p)-1,d0
loop    move.b  (a0)+,(a1)+
        dbra    d0,loop

        ;tidy cache
        movem.l d2-d7/a2-a6,-(sp)
        move.l  $4.w,a6
        jsr     -636(a6)
        movem.l (sp)+,d2-d7/a2-a6
        rts

        cnop    0,16
c2p:

                movem.l d2-d7/a2-a6,-(sp)

                ; a0 = chunky buffer
                ; a1 = output area

                lea     4*plsiz(a1),a1  ; a1 -> plane4

                move.l  a0,d0
                add.l   #16,d0
                and.b   #%11110000,d0
                move.l  d0,a0

                move.l  a0,a2
                add.l   #8*plsiz,a2

                lea     p0(pc),a3
                bra.s   mainloop

        cnop    0,16
mainloop:
        move.l  0(a0),d0
        move.l  4(a0),d2
        move.l  8(a0),d1
        move.l  12(a0),d3
        move.l  2(a0),d4
        move.l  10(a0),d5
        move.l  6(a0),d6
        move.l  14(a0),d7

        move.w  16(a0),d0
        move.w  24(a0),d1
        move.w  20(a0),d2
        move.w  28(a0),d3
        move.w  18(a0),d4
        move.w  26(a0),d5
        move.w  22(a0),d6
        move.w  30(a0),d7

        adda.w  #32,a0
        move.l  d6,a5
        move.l  d7,a6

        merge   d0,d1,d6,d7,$00FF00FF,8
        merge   d2,d3,d6,d7,$00FF00FF,8

        merge   d0,d2,d6,d7,$0F0F0F0F,4
        merge   d1,d3,d6,d7,$0F0F0F0F,4

        exg.l   d0,a5
        exg.l   d1,a6

        merge   d4,d5,d6,d7,$00FF00FF,8
        merge   d0,d1,d6,d7,$00FF00FF,8

        merge   d4,d0,d6,d7,$0F0F0F0F,4
        merge   d5,d1,d6,d7,$0F0F0F0F,4

        merge   d2,d0,d6,d7,$33333333,2
        merge   d3,d1,d6,d7,$33333333,2

        merge   d2,d3,d6,d7,$55555555,1
        merge   d0,d1,d6,d7,$55555555,1
        move.l  d3,2*4(a3)      ;plane2
        move.l  d2,3*4(a3)      ;plane3
        move.l  d1,0*4(a3)      ;plane0
        move.l  d0,1*4(a3)      ;plane1

        move.l  a5,d2
        move.l  a6,d3

        merge   d2,d4,d6,d7,$33333333,2
        merge   d3,d5,d6,d7,$33333333,2

        merge   d2,d3,d6,d7,$55555555,1
        merge   d4,d5,d6,d7,$55555555,1
        move.l  d3,6*4(a3)              ;bitplane6
        move.l  d2,7*4(a3)              ;bitplane7
        move.l  d5,4*4(a3)              ;bitplane4
        move.l  d4,5*4(a3)              ;bitplane5


inner:
        move.l  0(a0),d0
        move.l  4(a0),d2
        move.l  8(a0),d1
        move.l  12(a0),d3
        move.l  2(a0),d4
        move.l  10(a0),d5
        move.l  6(a0),d6
        move.l  14(a0),d7

        move.w  16(a0),d0
        move.w  24(a0),d1
        move.w  20(a0),d2
        move.w  28(a0),d3
        move.w  18(a0),d4
        move.w  26(a0),d5
        move.w  22(a0),d6
        move.w  30(a0),d7

        adda.w  #32,a0
        move.l  d6,a5
        move.l  d7,a6

        ; write bitplane 7

        move.l  2*4(a3),-2*plsiz(a1)    ;plane2
        merge   d0,d1,d6,d7,$00FF00FF,8
        merge   d2,d3,d6,d7,$00FF00FF,8

        ; write
        move.l  3*4(a3),-plsiz(a1)      ;plane3
        merge   d0,d2,d6,d7,$0F0F0F0F,4
        merge   d1,d3,d6,d7,$0F0F0F0F,4

        exg.l   d0,a5
        exg.l   d1,a6

        ; write
        move.l  0*4(a3),-4*plsiz(a1)    ;plane0
        merge   d4,d5,d6,d7,$00FF00FF,8
        merge   d0,d1,d6,d7,$00FF00FF,8

        ; write
        move.l  1*4(a3),-3*plsiz(a1) ;plane1
        merge   d4,d0,d6,d7,$0F0F0F0F,4
        merge   d5,d1,d6,d7,$0F0F0F0F,4

        ; write
        move.l  6*4(a3),2*plsiz(a1)     ;bitplane6
        merge   d2,d0,d6,d7,$33333333,2
        merge   d3,d1,d6,d7,$33333333,2

        ; write
        move.l  7*4(a3),3*plsiz(a1)     ;bitplane7
        merge   d2,d3,d6,d7,$55555555,1
        merge   d0,d1,d6,d7,$55555555,1
        move.l  d3,2*4(a3)      ;plane2
        move.l  d2,3*4(a3)      ;plane3
        move.l  d1,0*4(a3)      ;plane0
        move.l  d0,1*4(a3)      ;plane1

        move.l  a5,d2
        move.l  a6,d3

        move.l  4*4(a3),(a1)+           ;bitplane4
        merge   d2,d4,d6,d7,$33333333,2
        merge   d3,d5,d6,d7,$33333333,2

        move.l  5*4(a3),-4+1*plsiz(a1)  ;bitplane5
        merge   d2,d3,d6,d7,$55555555,1
        merge   d4,d5,d6,d7,$55555555,1
        move.l  d3,6*4(a3)              ;bitplane6
        move.l  d2,7*4(a3)              ;bitplane7
        move.l  d5,4*4(a3)              ;bitplane4
        move.l  d4,5*4(a3)              ;bitplane5

        cmpa.l  a0,a2
        bne.w   inner

        move.l  2*4(a3),-2*plsiz(a1)    ;plane2
        move.l  3*4(a3),-plsiz(a1)      ;plane3
        move.l  0*4(a3),-4*plsiz(a1)    ;plane0
        move.l  1*4(a3),-3*plsiz(a1)    ;plane1
        move.l  6*4(a3),2*plsiz(a1)     ;bitplane6
        move.l  7*4(a3),3*plsiz(a1)     ;bitplane7
        move.l  4*4(a3),(a1)+           ;bitplane4
        move.l  5*4(a3),-4+1*plsiz(a1)  ;bitplane5

exit
        movem.l (sp)+,d2-d7/a2-a6
        rts

        cnop    0,4
end:
p0      dc.l    0
p1      dc.l    0
p2      dc.l    0
p3      dc.l    0
p4      dc.l    0
p5      dc.l    0
p6      dc.l    0
p7      dc.l    0
              

Raus:
                move.l d7,d0
                movem.l (sp)+,d0/d3/a3/d7
                movem.l d2/a6,-(sp)
                move.l $4,a6
                move.l #0,a1
                jsr -294(a6)
                move.l d0,a1
                movem.l (sp)+,d0/a6
                move.l a6,-(sp)
                move.l $4,a6
                move.l d0,d1
                sub.l d0,d0
                bset d1,d0
                jsr -324(a6)
                move.l (sp)+,a6
                move.l #0,d0
                rts

                cnop    0,4

    END