. = 0x8c010000

.include "regs.s"

DC_PCLOCK = 49900000 ; Hz
BAUDRATE = 57600

VIDBEG = 0xa520a000
VIDSTRIDE = 640 * 4

DOTWD = 16	; bytes (implicitly known in putdot)
DOTHT = 5	; lines

XDOT = VIDSTRIDE / DOTWD

	.entry	.

stacktop:
; r0 = scratch
; r1 = video ram base
; r2 = row stride, in bytes
; r3 = pointer to data for next dot to store
; r4 = number of dots remaining to store
; r5 = dot X
; r6 = dot Y
; r7 = dot V
; r8 = scratch
; r9 = scratch
; r10 = scratch
; r11 = scratch
; r12 = scratch
	mov.l	vidbeg,r1
	mov.l	vidstride,r2
	mov.l	dotptr,r3
	mov.l	ndots,r4
1:	mov.b	@r3+,r5
	mov.b	@r3+,r6
	bsr	putdot
	 mov.b	@r3+,r7
	dt	r4
	bf	1b

	; We want to be running in P2, to fiddle CCR.  (The hardware
	; PDF, page 77, says that "CCR modifications must only be made
	; by a program in the non-cached P2 area".)
	mova	enable_cache,r0
	mov.l	px_mask,r8
	mov.l	p2_bits,r9
	and	r8,r0
	or	r9,r0
	jmp	@r0
	 nop

	.align	4
vidbeg:
	.long	VIDBEG
vidstride:
	.long	VIDSTRIDE
dotptr:
	.long	dots
ndots:
	.long	[enddots-dots]/3
px_mask:			; mask off Px-selecting bits
	.long	0x1fffffff
p2_bits:			; bits for P2
	.long	0xa0000000

	.align	4
enable_cache:
	; Now running in P2, so we can turn on the cache.
	mov.l	ccr_addr,r0
	mov.l	ccr_bits,r9
	mov.l	r9,@r0
	; The hardware PDF, page 77, says that "After CCR is updated,
	; an instruction that performs data access to the P0, P1, P3,
	; or U0 area should be located at least four instructions after
	; the CCR update instruction.  Also, a branch instruction to
	; the P0, P1, P3, or U0 area should be located at least eight
	; instructions after the CCR update instruction."  It doesn't
	; say why this is "should" rather than "must", nor does it
	; describe the consequences if this is not done, nor does it
	; say whether this "beyond" refers to address space or
	; instruction execution order (eg, does a three-instruction
	; loop that's executed three times count as nine instructions
	; or three? does a branch seven instructions forward count?).
	; We treat it pessimistically, making sure we burn eight
	; instructions by any of these measures.
	;
	; Gotta love incomplete doc.
	;
	mova	cacheon,r0	; #1
	mov.l	p0_bits,r9	; #2
	and	r8,r0		; #3
	or	r9,r0		; #4
	mov.l	stackptr,r15	; #5
	nop			; #6
	nop			; #7
	nop			; #8
	jmp	@r0
	 nop

	.align	4
ccr_addr:
	.long	CCR
ccr_bits:			; Not set: IIX OIX ORA WT
	.long	CCR_ICI | CCR_ICE | CCR_OCI | CCR_CB | CCR_OCE
p0_bits:			; bits for P0
	.long	0x80000000
stackptr:
	.long	stacktop

	.align	2
progress:
	mov	r0,r5
	mov	#8,r6
	bra	putdot
	 mov	#1,r7

	.align	4
cacheon:
2:	mov	r10,r11
	mov	#32,r12
3:	shlr	r11
	rotcl	r7
	mov	r12,r5
	bsr	putdot
	 mov	#10,r6
	dt	r12
	bf	3b
	mov.l	delay,r0
1:	dt	r0
	bf	1b
	bra	2b
	 add	#1,r10
	.align	4
delay:
	.long	1<<16

putdot:
; Stores dot of all 0s or all 1s, depending on low bit of r7, at (r5,r6)
; Destroys r0, r8, r9
	add	#1,r5
	add	#1,r6
	mov	#DOTWD,r0
	mul.l	r0,r5
	sts	macl,r8		; X offset in bytes
	mov.l	9f,r0
	mul.l	r0,r6
	sts	macl,r0		; Y offset in bytes
	add	r8,r0
	add	r1,r0
	mov	r7,r9
	shlr	r9
	bt	1f
	bra	2f
	 mov	#0,r8
1:	mov	#-1,r8
2:	mov	#DOTHT,r9
3:
	.macro	foo	b, o
	.if	$(b) >= 4
	mov.l	r8,@($(o),r0)
	foo	$(b)-4,$(o)+4
	.endif
	.endm
	foo	DOTWD,0
	dt	r9
	bf/s	3b
	 add	r2,r0
	rts
	 nop
	.align	4
9:	.long	DOTHT * VIDSTRIDE

dots:
; y\x 0 1 2 3 4 5 6 7
;  0  . . . . . . . .
;  1  . * * * * * * .
;  2  . * . . . * . .
;  3  . . . . * . . .
;  4  . . . . * . . .
;  5  . . . . * . . .
;  6  . * * * * * * .
;  7  . . . . . . . .
	;	x, y, v
	.byte	0, 0, 0
	.byte	1, 0, 0
	.byte	2, 0, 0
	.byte	3, 0, 0
	.byte	4, 0, 0
	.byte	5, 0, 0
	.byte	6, 0, 0
	.byte	7, 0, 0
	.byte	0, 1, 0
	.byte	1, 1, 1
	.byte	2, 1, 1
	.byte	3, 1, 1
	.byte	4, 1, 1
	.byte	5, 1, 1
	.byte	6, 1, 1
	.byte	7, 1, 0
	.byte	0, 2, 0
	.byte	1, 2, 1
	.byte	2, 2, 0
	.byte	3, 2, 0
	.byte	4, 2, 0
	.byte	5, 2, 1
	.byte	6, 2, 0
	.byte	7, 2, 0
	.byte	0, 3, 0
	.byte	1, 3, 0
	.byte	2, 3, 0
	.byte	3, 3, 0
	.byte	4, 3, 1
	.byte	5, 3, 0
	.byte	6, 3, 0
	.byte	7, 3, 0
	.byte	0, 4, 0
	.byte	1, 4, 0
	.byte	2, 4, 0
	.byte	3, 4, 0
	.byte	4, 4, 1
	.byte	5, 4, 0
	.byte	6, 4, 0
	.byte	7, 4, 0
	.byte	0, 5, 0
	.byte	1, 5, 0
	.byte	2, 5, 0
	.byte	3, 5, 0
	.byte	4, 5, 1
	.byte	5, 5, 0
	.byte	6, 5, 0
	.byte	7, 5, 0
	.byte	0, 6, 0
	.byte	1, 6, 1
	.byte	2, 6, 1
	.byte	3, 6, 1
	.byte	4, 6, 1
	.byte	5, 6, 1
	.byte	6, 6, 1
	.byte	7, 6, 0
	.byte	0, 7, 0
	.byte	1, 7, 0
	.byte	2, 7, 0
	.byte	3, 7, 0
	.byte	4, 7, 0
	.byte	5, 7, 0
	.byte	6, 7, 0
	.byte	7, 7, 0
	.byte	8, 8, 0
	.byte	9, 8, 0
	.byte	10, 8, 0
	.byte	11, 8, 0
	.byte	12, 8, 0
	.byte	13, 8, 0
	.byte	14, 8, 0
	.byte	15, 8, 0
	.byte	16, 8, 0
	.byte	17, 8, 0
	.byte	18, 8, 0
	.byte	19, 8, 0
	.byte	20, 8, 0
	.byte	21, 8, 0
	.byte	22, 8, 0
	.byte	23, 8, 0
	.byte	24, 8, 0
	.byte	25, 8, 0
	.byte	26, 8, 0
	.byte	27, 8, 0
	.byte	28, 8, 0
	.byte	29, 8, 0
	.byte	30, 8, 0
	.byte	31, 8, 0
	.byte	32, 8, 0
	.byte	33, 8, 0
	.byte	34, 8, 0
enddots: