Implement enough junk for my long long test program to link

src/std/linked/linked.src

@@ -86,6 +86,10 @@ __ixor		:= 000198h
 __ladd		:= 00019Ch
 	public	__land
 __land		:= 0001A4h
+	public	__lcmps
+__lcmps		:= 0001A8h
+	public	__lcmpu
+__lcmpu		:= 0001ACh
 	public	__lcmpzero
 __lcmpzero	:= 0001B0h
 	public	__lmuls

src/std/shared/lcmps_fast.src

@@ -0,0 +1,12 @@
+	public	__lcmps_fast
+__lcmps_fast:
+	ld	d, a
+	ld	a, e
+	sub 	a, d
+	ret	nz
+	sbc	hl, bc
+	ret	z
+	sbc	a, a
+	ret	nz
+	inc	a
+	ret

src/std/shared/lcmpu_fast.src

@@ -0,0 +1,8 @@
+	public	__lcmpu_fast
+__lcmpu_fast:
+	sub	a, e
+	ccf
+	ret	nz
+	or	a, a
+	sbc	hl, bc
+	ret

src/std/shared/llcmpu.src

@@ -0,0 +1,53 @@
+	public	__llcmpu
+__llcmpu:
+if 0
+	ld	iy, 0
+	add	iy, sp
+	ld	a, b
+	sub	a, (iy + 10)
+	ret	nz
+	ld	a, c
+	sub	a, (iy + 9)
+	ret	nz
+	ex	de, hl
+	ld	bc, (iy + 6)
+	sbc	hl, bc
+	ret	nz
+	ex	de, hl
+	ld	bc, (iy + 3)
+	sbc	hl, bc
+	ret
+else
+	ld 	a, c
+	ld	iyh, b
+	pop	bc
+	or	a, a
+	sbc	hl, bc
+	ex	de, hl
+	pop	bc
+	pop	de
+	jr	nz, .ne24
+	sbc	hl, bc
+	jr	nz, .ne48
+	sbc	a, e
+	jr	nz, .ne56
+	ld	a, iyh
+	sbc	a, d
+	ld	hl, 9
+	add	hl, sp
+	ld	sp, hl
+	ret
+.ne24:
+	sbc	hl, bc
+.ne48:
+	sbc	a, e
+.ne56:
+	ld	a, iyh
+	sbc	a, d
+	ld	hl, 9
+	add	hl, sp
+	ld	sp, hl
+	ret	nz
+	inc	a
+	ret
+end if

src/std/shared/llcmpu_fast.src

@@ -0,0 +1,53 @@
+	public	__llcmpu_fast
+__llcmpu_fast:
+if 1
+	ld	iy, 0
+	add	iy, sp
+	ld	a, b
+	sub	a, (iy + 10)
+	ret	nz
+	ld	a, c
+	sub	a, (iy + 9)
+	ret	nz
+	ex	de, hl
+	ld	bc, (iy + 6)
+	sbc	hl, bc
+	ret	nz
+	ex	de, hl
+	ld	bc, (iy + 3)
+	sbc	hl, bc
+	ret
+else
+	ld 	a, c
+	ld	iyh, b
+	pop	bc
+	or	a, a
+	sbc	hl, bc
+	ex	de, hl
+	pop	bc
+	pop	de
+	jr	nz, .ne24
+	sbc	hl, bc
+	jr	nz, .ne48
+	sbc	a, e
+	jr	nz, .ne56
+	ld	a, iyh
+	sbc	a, d
+	ld	hl, 9
+	add	hl, sp
+	ld	sp, hl
+	ret
+.ne24:
+	sbc	hl, bc
+.ne48:
+	sbc	a, e
+.ne56:
+	ld	a, iyh
+	sbc	a, d
+	ld	hl, 9
+	add	hl, sp
+	ld	sp, hl
+	ret	nz
+	inc	a
+	ret
+end if

src/std/shared/lldivu_b_fast.src

@@ -0,0 +1,27 @@
+	public	__lldivu
+__lldivu:
+	public	__lldivu_b_fast
+__lldivu_b_fast:
+; Not well-optimized
+	ld	iy, 0
+	add	iy, sp
+	ld	iy, (iy + 3)
+	ld	iyh, 64
+.loop:
+	adc	hl, hl
+	ex	de, hl
+	adc	hl, hl
+	ex	de, hl
+	rl	c
+	rl	b
+	rla
+	jr	c, .1
+	cp	a, iyl
+	jr	nc, .0
+.1:
+	sub	a, iyl
+	inc	l
+.0:
+	dec	iyh
+	jr	nz, .loop
+	ret

src/std/shared/llmulu_b_fast.src

@@ -0,0 +1,63 @@
+	public	__llmulu
+__llmulu:
+	public	__llmulu_b_fast
+__llmulu_b_fast:
+	ld	iy, 0
+	add	iy, sp
+	push	de
+	push	hl
+	ld	a, (iy + 3)
+	ld	h, a
+	mlt	hl
+	ld	(iy - 6), l
+	ld	d, a
+	ld	e, (iy - 5)
+	mlt	de
+	ld	l, h
+	ld	h, 0
+	add	hl, de
+	ld	(iy - 5), l
+	ld	d, a
+	ld	e, (iy - 4)
+	mlt	de
+	ld	l, h
+	ld	h, 0
+	add	hl, de
+	ld	(iy - 4), l
+	ld	d, a
+	ld	e, (iy - 3)
+	mlt	de
+	ld	l, h
+	ld	h, 0
+	add	hl, de
+	ld	(iy - 3), l
+	ld	d, a
+	ld	e, (iy - 2)
+	mlt	de
+	ld	l, h
+	ld	h, 0
+	add	hl, de
+	ld	(iy - 2), l
+	ld	d, a
+	ld	e, (iy - 1)
+	mlt	de
+	ld	l, h
+	ld	h, 0
+	add	hl, de
+	ld	(iy - 1), l
+	ld	d, a
+	ld	e, c
+	mlt	de
+	ld	l, h
+	ld	h, 0
+	add	hl, de
+	ld	c, l
+	ld	d, a
+	ld	e, b
+	mlt	de
+	ld	a, h
+	add	a, e
+	ld	b, a
+	pop	hl
+	pop	de
+	ret

src/std/shared/llshrs.src

@@ -0,0 +1,31 @@
+	public	__llshrs
+__llshrs:
+; Not well-optimized, but it probably works
+	push	iy
+	ld	iy, 0
+	add	iy, sp
+	push	af
+	ld	a, c
+	ld	c, b
+	ld	b, (iy + 6)
+	inc	b
+	dec	b
+	jr	z, .finish
+	push	de
+	push	hl
+.loop:
+	sra	c
+	rra
+	rr	(iy - 1)
+	rr	d
+	rr	e
+	rr	(iy - 4)
+	rr	h
+	rr	l
+	djnz	.loop
+	ld	b, c
+	ld	c, a
+.finish:
+	pop	af
+	pop	iy
+	ret