initial

boot.asm

@@ -0,0 +1,120 @@
+STACK_TOP = 0xFFFF
+; At 500 kHz the CPU has about 500 T-states per millisecond.
+; The inner delay loop below costs about 26 T-states per taken iteration:
+;   dec hl  =  6T
+;   ld a,h  =  4T
+;   or l    =  4T
+;   jr nz   = 12T
+; 500 / 26 = 19.2, so 20 iterations is the nearest simple whole-number choice.
+DELAY_LOOPS_PER_MS = 20
+
+        .globl _main
+        .globl _delay_ms
+        .globl _io_out
+        .globl _io_in
+        .globl s__DATA
+        .globl l__DATA
+        .globl s__INITIALIZER
+        .globl s__INITIALIZED
+        .globl l__INITIALIZER
+
+.area _ITABLE (ABS)
+
+.org 0x00
+        jp init
+.org 0x08
+        reti
+.org 0x10
+        reti
+.org 0x18
+        reti
+.org 0x20
+        reti
+.org 0x28
+        reti
+.org 0x30
+        reti
+.org 0x38
+        reti
+
+.area _HOME
+init:
+        di
+        ld sp, #STACK_TOP
+
+        call gsinit
+        call _main
+
+hang:
+        jr hang
+
+.area _GSINIT
+gsinit:
+        ld bc, #l__INITIALIZER
+        ld a, b
+        or c
+        jr z, zero_data
+
+        ld de, #s__INITIALIZED
+        ld hl, #s__INITIALIZER
+        ldir
+
+zero_data:
+        ld bc, #l__DATA
+        ld a, b
+        or c
+        jr z, gsinit_done
+
+        ld hl, #s__DATA
+        xor a
+
+clear_loop:
+        ld (hl), a
+        inc hl
+        dec bc
+        ld a, b
+        or c
+        jr nz, clear_loop
+
+gsinit_done:
+.area _GSFINAL
+        ret
+
+.area _CODE
+; Write a byte to an I/O port.
+; Input: A = port address, L = data byte.
+_io_out::
+        ld      c, a
+        ld      a, l
+        out     (c), a
+        ret
+
+; Read a byte from an I/O port.
+; Input: A = port address.
+; Return: A = data byte.
+_io_in::
+        ld      c, a
+        in      a, (c)
+        ret
+
+; Approximate millisecond delay.
+; Input: HL = delay in milliseconds.
+; For CLK_SPEED = 500000 Hz this uses 20 inner iterations per ms.
+_delay_ms::
+        ld      a, h
+        or      l
+        ret     z
+delay_ms_outer$:
+        push    hl
+        ld      hl, #DELAY_LOOPS_PER_MS
+delay_ms_inner$:
+        dec     hl
+        ld      a, h
+        or      l
+        jr      nz, delay_ms_inner$
+        pop     hl
+        dec     hl
+        ld      a, h
+        or      l
+        jr      nz, delay_ms_outer$
+        ret