NEW
  AUTO 3,1
*--------------------------------------
M32.ADD			lda #$72				ADC (zp)
				.HS 2C					BIT ABS
M32.SUB			lda #$F2				SBC (zp)

				sta .3

				asl						CS if SBC, CC if ADC

				ldy #4
				ldx #4

.2				lda (pStack),y
.3				adc (pStack)			SELF MODIFIED
				sta (pStack),y

				inc pStack
				dex
				bne .2

				rts
*--------------------------------------
* Returns:
*  ff X < Y
*   0 X = Y
*   1 X > Y
*--------------------------------------
*M32.UCMP
*M32.ICMP
M32.CMP			jsr M32.SUB

				tay						A = BYTE 3
				bmi .2

.1				ora (pStack)			BYTE 0
				ldy #1
				ora (pStack),y			BYTE 1
				iny
				ora (pStack),y			BYTE 2
				beq .8

				lda #1
				.HS 2C					BIT ABS

.2				lda #$ff

.8				inc pStack
				inc pStack
				inc pStack
				inc pStack
				
				rts			
*--------------------------------------
M32.UMUL		clc
				.HS B0					BCS
M32.IMUL		sec

				php
				jsr M32.MULDIVMOD
				jsr M32.MUL

				plp
				bcc M32.PutTMP32

M32.ITMP		lda ACC32.Sign
				eor ARG32.Sign
				bpl M32.PutTMP32
*--------------------------------------
* M32.PutnTMP32
*--------------------------------------
				lda #TMP32
				.HS 2C					BIT ABS
*--------------------------------------
M32.PutnARG32	lda #ARG32
				sta .1+1
				ldy #0
				ldx #3
				sec

.1				lda $ff,y				SELF MODIFIED
				eor #$ff				two's complement of X bytes
				adc #0
				sta (pStack),y
				iny
				dex
				bpl .1

				rts
*--------------------------------------
M32.UDIV		clc
				.HS B0					BCS
M32.IDIV		sec
				clv
				bra M32.MOD
				
M32.UMOD		clc
				.HS B0					BCS
M32.IMOD		sec
				bit M32.RTS				$60

M32.MOD			php
				jsr M32.MULDIVMOD
				jsr M32.DIVMOD

				plp
				bcc .3					unsigned

				bvs M32.ITMP			return ITMP

				lda ACC32.Sign			IDIV
				eor ARG32.Sign
				bpl M32.PutARG32
				bra M32.PutnARG32

.3				bvc M32.PutARG32		DIV
*--------------------------------------
M32.PutTMP32	lda #TMP32				MOD
				.HS 2C					BIT ABS
*--------------------------------------
M32.PutARG32	lda #ARG32
				sta .1+1
				
				ldy #3
				
.1				lda $ff,y				SELF MODIFIED
				sta (pStack),y
				dey
				bpl .1

				rts
*--------------------------------------
M32.PopACC32	ldy #0

.1				lda (pStack)
				inc pStack
				sta ACC32,y
				iny
				cpy #4
				bne .1

				sta ACC32.Sign
				rts
*--------------------------------------
M32.GetARG32	ldy #0

.1				lda (pStack),y
				sta ARG32,y
				iny
				cpy #4
				bne .1

				sta ARG32.Sign

M32.RTS			rts
*--------------------------------------
M32.MULDIVMOD	php
				jsr M32.PopACC32
				jsr M32.GetARG32
				plp
				bcc M32.RTS
				jsr .1
*--------------------------------------
* M32.ARG32ABS	
*--------------------------------------
				lda ARG32.Sign
				bmi M32.nARG32
				rts
*--------------------------------------
* M32.ACC32ABS
*--------------------------------------
.1				lda ACC32.Sign
				bpl M32.RTS
*--------------------------------------
M32.nACC32		ldy #ACC32
				.HS 2C					BIT ABS
M32.nARG32		ldy #ARG32

				ldx #3
				sec

.1				lda $0,y				two's complement of X bytes
				eor #$ff
				adc #0
				sta $0,y
				iny
				dex
				bpl .1

				rts
*--------------------------------------
* http://6502.org/source/integers/32muldiv.htm
* http://nparker.llx.com/a2/mult.html
*--------------------------------------
* ARG32*ACC32->TMP32
*--------------------------------------
M32.MUL			jsr M32.TMP32Z
				ldx #32

.1				lsr ARG32+3
				ror ARG32+2
				ror ARG32+1
				ror ARG32
				bcc .3

				clc

				ldy #0

.2				lda TMP32,y
				adc ACC32,y
				sta TMP32,y

				iny
				tya
				eor #4
				bne .2

.3				asl ACC32
				rol ACC32+1
				rol ACC32+2
				rol ACC32+3
				
				dex
				bne  .1

				clc
				rts
*--------------------------------------
* ARG32 = ARG32 div ACC32
* TMP32 = ARG32 mod ACC32
*--------------------------------------
M32.DIVMOD		jsr M32.TMP32Z
				
				ldx #32
				
.1				asl ARG32
				rol ARG32+1
				rol ARG32+2
				rol ARG32+3
				rol TMP32
				rol TMP32+1
				rol TMP32+2
				rol TMP32+3
				sec 
				lda TMP32
				sbc ACC32
				pha
				lda TMP32+1
				sbc ACC32+1
				pha
				lda TMP32+2
				sbc ACC32+2
				pha
				lda TMP32+3
				sbc ACC32+3
				bcs .2
				
				pla
				pla
				pla
				dex
				bne .1

				rts
				
.2				sta TMP32+3
				pla
				sta TMP32+2
				pla
				sta TMP32+1
				pla
				sta TMP32
				inc ARG32				bit0 always 0 because of .1 asl

				dex
				bne .1

				rts
*--------------------------------------
*uint32_t lcg_parkmiller(uint32_t *state)
*{
*	// Precomputed parameters for Schrage's method
*	const uint32_t M = 0x7fffffff;
*	const uint32_t A = 48271;
*	const uint32_t Q = M / A;    // 44488
*	const uint32_t R = M % A;    //  3399
*
*	uint32_t div = *state / Q;	// max: M / Q = A = 48,271
*	uint32_t rem = *state % Q;	// max: Q - 1     = 44,487
*
*	int32_t s = rem * A;	// max: 44,487 * 48,271 = 2,147,431,977 = 0x7fff3629
*	int32_t t = div * R;	// max: 48,271 *  3,399 =   164,073,129
*	int32_t result = s - t;
*
*	if (result < 0)
*		result += M;
*
*	return *state = result;
*}
*--------------------------------------
* M32.RND		TODO
*--------------------------------------
M32.A2ACC32		sta ACC32
				.HS 2C					BIT ABS
M32.ACC32Z		stz ACC32
				stz ACC32+1
				stz ACC32+2
				stz ACC32+3
				rts
*--------------------------------------
M32.TMP32Z		stz TMP32
				stz TMP32+1
				stz TMP32+2
				stz TMP32+3
				rts
*--------------------------------------
MAN
SAVE usr/src/sys/kernel.s.math32
LOAD usr/src/sys/kernel.s
ASM