Optimize 24-bit and 32-bit implementations of and/or/xor

Author: calc84maniac

src/crt/iand.src

@@ -10,20 +10,23 @@ __iand := 000134h
 else
 
 __iand:
+; CC: 24*r(PC)+11*r(SPL)+10*w(SPL)+1
 	push	af
-	push	bc
-	inc	sp
-	inc	sp
+	public	__iand.hijack_land
+.hijack_land:
 	push	hl
-	ld	hl, 2
+	scf
+	sbc	hl, hl
 	add	hl, sp
+	push	bc
 	ld	a, (hl)
 	inc	hl
+	ld	sp, hl
+	inc	hl
+	inc	hl
 	and	a, (hl)
-	dec	hl
 	ld	(hl), a
 	pop	hl
-	inc	sp
 	ld	a, h
 	and	a, b
 	ld	h, a

src/crt/iand_fast.src

@@ -3,19 +3,20 @@
 	section	.text
 	public	__iand_fast
 __iand_fast:
-	push	bc
-	inc	sp
-	inc	sp
+; CC: 22*r(PC)+8*r(SPL)+7*w(SPL)+1
 	push	hl
-	ld	hl, 2
+	scf
+	sbc	hl, hl
 	add	hl, sp
+	push	bc
 	ld	a, (hl)
 	inc	hl
+	ld	sp, hl
+	inc	hl
+	inc	hl
 	and	a, (hl)
-	dec	hl
 	ld	(hl), a
 	pop	hl
-	inc	sp
 	ld	a, h
 	and	a, b
 	ld	h, a

src/crt/ior.src

@@ -10,20 +10,23 @@ __ior := 000168h
 else
 
 __ior:
+; CC: 24*r(PC)+11*r(SPL)+10*w(SPL)+1
 	push	af
-	push	bc
-	inc	sp
-	inc	sp
+	public	__ior.hijack_lor
+.hijack_lor:
 	push	hl
-	ld	hl, 2
+	scf
+	sbc	hl, hl
 	add	hl, sp
+	push	bc
 	ld	a, (hl)
 	inc	hl
+	ld	sp, hl
+	inc	hl
+	inc	hl
 	or	a, (hl)
-	dec	hl
 	ld	(hl), a
 	pop	hl
-	inc	sp
 	ld	a, h
 	or	a, b
 	ld	h, a

src/crt/ior_fast.src

@@ -3,19 +3,20 @@
 	section	.text
 	public	__ior_fast
 __ior_fast:
-	push	bc
-	inc	sp
-	inc	sp
+; CC: 22*r(PC)+8*r(SPL)+7*w(SPL)+1
 	push	hl
-	ld	hl, 2
+	scf
+	sbc	hl, hl
 	add	hl, sp
+	push	bc
 	ld	a, (hl)
 	inc	hl
+	ld	sp, hl
+	inc	hl
+	inc	hl
 	or	a, (hl)
-	dec	hl
 	ld	(hl), a
 	pop	hl
-	inc	sp
 	ld	a, h
 	or	a, b
 	ld	h, a

src/crt/ixor.src

@@ -10,20 +10,23 @@ __ixor := 000198h
 else
 
 __ixor:
+; CC: 24*r(PC)+11*r(SPL)+10*w(SPL)+1
 	push	af
-	push	bc
-	inc	sp
-	inc	sp
+	public	__ixor.hijack_lxor
+.hijack_lxor:
 	push	hl
-	ld	hl, 2
+	scf
+	sbc	hl, hl
 	add	hl, sp
+	push	bc
 	ld	a, (hl)
 	inc	hl
+	ld	sp, hl
+	inc	hl
+	inc	hl
 	xor	a, (hl)
-	dec	hl
 	ld	(hl), a
 	pop	hl
-	inc	sp
 	ld	a, h
 	xor	a, b
 	ld	h, a

src/crt/ixor_fast.src

@@ -3,19 +3,20 @@
 	section	.text
 	public	__ixor_fast
 __ixor_fast:
-	push	bc
-	inc	sp
-	inc	sp
+; CC: 22*r(PC)+8*r(SPL)+7*w(SPL)+1
 	push	hl
-	ld	hl, 2
+	scf
+	sbc	hl, hl
 	add	hl, sp
+	push	bc
 	ld	a, (hl)
 	inc	hl
+	ld	sp, hl
+	inc	hl
+	inc	hl
 	xor	a, (hl)
-	dec	hl
 	ld	(hl), a
 	pop	hl
-	inc	sp
 	ld	a, h
 	xor	a, b
 	ld	h, a

src/crt/land.src

@@ -10,29 +10,12 @@ __land := 0001A4h
 else
 
 __land:
+; CC: 30*r(PC)+11*r(SPL)+10*w(SPL)+2
 	push	af
 	and	a, e
 	ld	e, a
-	push	bc
-	inc	sp
-	inc	sp
-	push	hl
-	ld	hl, 2
-	add	hl, sp
-	ld	a, (hl)
-	inc	hl
-	and	a, (hl)
-	dec	hl
-	ld	(hl), a
-	pop	hl
-	inc	sp
-	ld	a, h
-	and	a, b
-	ld	h, a
-	ld	a, l
-	and	a, c
-	ld	l, a
-	pop	af
-	ret
+	jp	__iand.hijack_land
+
+	extern	__iand.hijack_land
 
 end if

src/crt/land_fast.src

@@ -3,21 +3,22 @@
 	section	.text
 	public	__land_fast
 __land_fast:
+; CC: 24*r(PC)+8*r(SPL)+7*w(SPL)+1
 	and	a, e
 	ld	e, a
-	push	bc
-	inc	sp
-	inc	sp
 	push	hl
-	ld	hl, 2
+	scf
+	sbc	hl, hl
 	add	hl, sp
+	push	bc
 	ld	a, (hl)
 	inc	hl
+	ld	sp, hl
+	inc	hl
+	inc	hl
 	and	a, (hl)
-	dec	hl
 	ld	(hl), a
 	pop	hl
-	inc	sp
 	ld	a, h
 	and	a, b
 	ld	h, a

src/crt/lor.src

@@ -10,29 +10,12 @@ __lor := 0001D8h
 else
 
 __lor:
+; CC: 30*r(PC)+11*r(SPL)+10*w(SPL)+2
 	push	af
 	or	a, e
 	ld	e, a
-	push	bc
-	inc	sp
-	inc	sp
-	push	hl
-	ld	hl, 2
-	add	hl, sp
-	ld	a, (hl)
-	inc	hl
-	or	a, (hl)
-	dec	hl
-	ld	(hl), a
-	pop	hl
-	inc	sp
-	ld	a, h
-	or	a, b
-	ld	h, a
-	ld	a, l
-	or	a, c
-	ld	l, a
-	pop	af
-	ret
+	jp	__ior.hijack_lor
+
+	extern	__ior.hijack_lor
 
 end if

src/crt/lor_fast.src

@@ -3,21 +3,22 @@
 	section	.text
 	public	__lor_fast
 __lor_fast:
+; CC: 24*r(PC)+8*r(SPL)+7*w(SPL)+1
 	or	a, e
 	ld	e, a
-	push	bc
-	inc	sp
-	inc	sp
 	push	hl
-	ld	hl, 2
+	scf
+	sbc	hl, hl
 	add	hl, sp
+	push	bc
 	ld	a, (hl)
 	inc	hl
+	ld	sp, hl
+	inc	hl
+	inc	hl
 	or	a, (hl)
-	dec	hl
 	ld	(hl), a
 	pop	hl
-	inc	sp
 	ld	a, h
 	or	a, b
 	ld	h, a