added __*equzero/__*cmpequ as a 'potentially' faster/smaller version of __*cmpzero/__*cmpu that only outputs the Zero flag

Author: ZERICO2005

src/crt/dtof.src

@@ -173,7 +173,7 @@ __dtof:
 	; BC:UDE:UHL = 1 << shift
 	; (SP) = X
 	call	__lland
-	call	__llcmpzero
+	call	__llequzero
 	pop	hl
 	; DE and BC are swapped here
 	pop	bc
@@ -215,6 +215,6 @@ __dtof:
 	jr	.L.ret_copysign
 
 	.extern	__lland
-	.extern	__llcmpzero
+	.extern	__llequzero
 	.extern	__llshru
 	.extern	__lshru

src/crt/i48cmpu.src

@@ -10,13 +10,21 @@
 	.assume	adl=1
 
 	.section	.text
+
+	.global	__i48cmpequ
+	.type	__i48cmpequ, @function
+
 	.global	__i48cmpu
 	.type	__i48cmpu, @function
 
+__i48cmpequ:
+	; Zero = (UDE:UHL == UIY:UBC)
 __i48cmpu:
 ; CC: 15 bytes
 ; Minimum: 11F + 6R + 3W + 2
 ; Maximum: 16F + 6R + 3W + 2
+	; Zero = (UDE:UHL == UIY:UBC)
+	; Carry = (UDE:UHL unsigned_less_than UIY:UBC)
 	push	hl
 	or	a, a
 	lea	hl, iy + 0

src/crt/i48cmpzero.src

@@ -6,10 +6,18 @@
 	.assume	adl=1
 
 	.section	.text
+
+	.global	__i48equzero
+	.type	__i48equzero, @function
+
 	.global	__i48cmpzero
 	.type	__i48cmpzero, @function
 
+__i48equzero:
+	; Zero flag = (UDE:UHL == 0)
 __i48cmpzero:
+	; Zero flag = (UDE:UHL == 0)
+	; Sign flag = (UDE:UHL < 0)
 ; CC: 13 bytes
 ; Minimum:  8F + 3R + 2
 ; Maximum: 14F + 3R + 3

src/crt/lcmpu.src

@@ -1,16 +1,25 @@
 	.assume	adl=1
 
 	.section	.text
+
+	.global	__lcmpequ
+	.type	__lcmpequ, @function
+
 	.global	__lcmpu
 	.type	__lcmpu, @function
 
 .ifdef PREFER_OS_CRT
 
 	.set	__lcmpu, 0x0001AC
+	.set	__lcmpequ, __lcmpu
 
 .else
 
+__lcmpequ:
+	; Zero = (E:UHL == A:UBC)
 __lcmpu:
+	; Zero = (E:UHL == A:UBC)
+	; Carry = (E:UHL unsigned_less_than A:UBC)
 	cp	a, e
 	ccf
 	ret	nz

src/crt/lcmpzero.src

@@ -1,10 +1,18 @@
 	.assume	adl=1
 
 	.section	.text
+
+	.global	__lequzero
+	.type	__lequzero, @function
+
 	.global	__lcmpzero
 	.type	__lcmpzero, @function
 
+__lequzero:
+	; Zero flag = (E:UHL == 0)
 __lcmpzero:
+	; Zero flag = (E:UHL == 0)
+	; Sign flag = (E:UHL < 0)
 	inc	e
 	dec	e
 	ret	nz

src/crt/llcmpu.src

@@ -1,10 +1,18 @@
 	.assume	adl=1
 
 	.section	.text
+
+	.global	__llcmpequ
+	.type	__llcmpequ, @function
+
 	.global	__llcmpu
 	.type	__llcmpu, @function
 
+__llcmpequ:
+	; Zero = (BC:UDE:UHL == (SP64))
 __llcmpu:
+	; Zero = (BC:UDE:UHL == (SP64))
+	; Carry = (BC:UDE:UHL unsigned_less_than (SP64))
 	push	iy
 	ld	iy, 0
 	add	iy, sp

src/crt/llcmpzero.src

@@ -1,10 +1,18 @@
 	.assume	adl=1
 
 	.section	.text
+
+	.global	__llequzero
+	.type	__llequzero, @function
+
 	.global	__llcmpzero
 	.type	__llcmpzero, @function
 
+__llequzero:
+	; Zero flag = (BC:UDE:UHL == 0)
 __llcmpzero:
+	; Zero flag = (BC:UDE:UHL == 0)
+	; Sign flag = (BC:UDE:UHL < 0)
 	inc	b
 	dec	b
 	ret	nz

src/libc/strtoll.src

@@ -26,7 +26,7 @@ _strtoll.maybe_out_of_range:
 	; negative
 	; check that the result is not an exact INT_MIN
 	ld	b, a		; B = (B << 1)
-	call	__llcmpzero
+	call	__llequzero
 	set	7, b
 	ret	z		; exact INT_MIN
 _strtoll.underflow:
@@ -344,4 +344,4 @@ __strtoll_common.set_overflow_bit:
 
 	.extern	_errno
 	.extern	__llneg
-	.extern	__llcmpzero
+	.extern	__llequzero

src/softfloat/s_shiftRightJam64.src

@@ -41,7 +41,7 @@ _softfloat_shiftRightJam64:
 	ld	hl, (iy + 4)
 	call	__llshl
 	xor	a, a
-	call	__llcmpzero
+	call	__llequzero
 	jr	z, .L.no_round
 	inc	a	; ld a, 1
 .L.no_round:
@@ -58,7 +58,7 @@ _softfloat_shiftRightJam64:
 .L.overflow_shift:
 	; A is zero here
 	ld	hl, (iy + 4)
-	call	__llcmpzero
+	call	__llequzero
 	ret	z
 	xor	a, a
 	sbc	hl, hl
@@ -69,6 +69,6 @@ _softfloat_shiftRightJam64:
 	inc	hl
 	ret
 
-	.extern	__llcmpzero
+	.extern	__llequzero
 	.extern	__llshl
 	.extern	__llshru