term: 40-column text terminal on the CGB LCD (4x8 dynamic tiles)

Add a background-tile text terminal that packs TWO 4px-wide characters
into each 8x8 tile, giving a 40x18 grid instead of the 20x18 you'd get
from an 8x8 font. The tilemap is static (one dedicated VRAM tile per
screen position); we rebuild a tile's 16 bytes from two glyphs whenever
a character changes. 360 tiles exceed the 256 a single tilemap can
address, so positions 256-359 live in VRAM bank 1 via the CGB tilemap
attribute bank-bit -- hence CGB-only.

- Switch the ROM to CGB (rgbfix -C). Safe for the FS: every process has
  PROC_WRAMB=1, so SVBK stays on bank 1 and the WRAMX FS caches don't move.
- CGB BG palette 0 = white bg / black text via BCPS/BCPD.
- src/term.asm: term_init (palette + static tilemap + clear buffer),
  build_tile (combine two 4px glyphs -> one 8x8 tile, correct VRAM bank),
  term_redraw (rebuild all tiles), term_puts, term_show, term_test.
- src/font.asm: 96-glyph 3x5-in-4x8 font, generated by tools/genfont.py.
- 40x18 text buffer at $D600 (WRAMX, above the FS caches).
- tools/ppmview.py: render a --shot PPM as ASCII so the LCD is inspectable
  from the shell during development.

Verified via emulator screenshot: "GBOS TERMINAL", the alphabet, and a
bank-1 row all render legibly at 40 columns. Serial shell + filesystem
still work unchanged.

Note: a full term_redraw builds 360 tiles and takes ~8 frames; fine for
incremental single-char updates, but scrolling needs a smarter path
(next milestone). Input (no keyboard) is also still TODO.

3 files changed, 482 insertions, 0 deletions

diff --git a/src/boot.asm b/src/boot.asm
index 9d9571e..0a5f2e3 100644
--- a/src/boot.asm
+++ b/src/boot.asm
@@ -56,6 +56,8 @@ KernelInit:
     ld a, HIGH(wKernelPCB)
     ld [wCurProc+1], a
     call fs_init                ; mount the disk (format on first boot; persists)
+    call term_init              ; set up the LCD text terminal
+    call term_test              ; render a test pattern
 
     ; --- spawn the init task (pid 1); it fork()s the rest ---
     ld hl, TaskInit
diff --git a/src/font.asm b/src/font.asm
new file mode 100644
index 0000000..a6ff7d9
--- /dev/null
+++ b/src/font.asm
@@ -0,0 +1,99 @@
+; auto-generated by tools/genfont.py - 4x8 font (3x5 glyphs), 96 glyphs
+SECTION "font", ROM0
+FontData::
+    db $00,$00,$00,$00,$00,$00,$00,$00    ; ' '
+    db $04,$04,$04,$00,$04,$00,$00,$00    ; '!'
+    db $0A,$0A,$00,$00,$00,$00,$00,$00    ; '"'
+    db $0A,$0E,$0A,$0E,$0A,$00,$00,$00    ; '#'
+    db $04,$06,$0C,$06,$0C,$00,$00,$00    ; '$'
+    db $0A,$02,$04,$08,$0A,$00,$00,$00    ; '%'
+    db $04,$0A,$04,$0A,$06,$00,$00,$00    ; '&'
+    db $04,$04,$00,$00,$00,$00,$00,$00    ; '''
+    db $02,$04,$04,$04,$02,$00,$00,$00    ; '('
+    db $08,$04,$04,$04,$08,$00,$00,$00    ; ')'
+    db $0A,$04,$0E,$04,$0A,$00,$00,$00    ; '*'
+    db $00,$04,$0E,$04,$00,$00,$00,$00    ; '+'
+    db $00,$00,$00,$04,$08,$00,$00,$00    ; ','
+    db $00,$00,$0E,$00,$00,$00,$00,$00    ; '-'
+    db $00,$00,$00,$00,$04,$00,$00,$00    ; '.'
+    db $02,$02,$04,$08,$08,$00,$00,$00    ; '/'
+    db $04,$0A,$0A,$0A,$04,$00,$00,$00    ; '0'
+    db $04,$0C,$04,$04,$0E,$00,$00,$00    ; '1'
+    db $0C,$02,$04,$08,$0E,$00,$00,$00    ; '2'
+    db $0C,$02,$04,$02,$0C,$00,$00,$00    ; '3'
+    db $0A,$0A,$0E,$02,$02,$00,$00,$00    ; '4'
+    db $0E,$08,$0C,$02,$0C,$00,$00,$00    ; '5'
+    db $06,$08,$0C,$0A,$04,$00,$00,$00    ; '6'
+    db $0E,$02,$04,$04,$04,$00,$00,$00    ; '7'
+    db $04,$0A,$04,$0A,$04,$00,$00,$00    ; '8'
+    db $04,$0A,$06,$02,$0C,$00,$00,$00    ; '9'
+    db $00,$04,$00,$04,$00,$00,$00,$00    ; ':'
+    db $00,$04,$00,$04,$08,$00,$00,$00    ; ';'
+    db $02,$04,$08,$04,$02,$00,$00,$00    ; '<'
+    db $00,$0E,$00,$0E,$00,$00,$00,$00    ; '='
+    db $08,$04,$02,$04,$08,$00,$00,$00    ; '>'
+    db $0C,$02,$04,$00,$04,$00,$00,$00    ; '?'
+    db $04,$0A,$0E,$08,$06,$00,$00,$00    ; '@'
+    db $0C,$0A,$0E,$0A,$0A,$00,$00,$00    ; 'A'
+    db $0C,$0A,$0C,$0A,$0C,$00,$00,$00    ; 'B'
+    db $06,$08,$08,$08,$06,$00,$00,$00    ; 'C'
+    db $0C,$0A,$0A,$0A,$0C,$00,$00,$00    ; 'D'
+    db $0E,$08,$0C,$08,$0E,$00,$00,$00    ; 'E'
+    db $0E,$08,$0C,$08,$08,$00,$00,$00    ; 'F'
+    db $06,$08,$0A,$0A,$06,$00,$00,$00    ; 'G'
+    db $0A,$0A,$0E,$0A,$0A,$00,$00,$00    ; 'H'
+    db $0E,$04,$04,$04,$0E,$00,$00,$00    ; 'I'
+    db $02,$02,$02,$0A,$04,$00,$00,$00    ; 'J'
+    db $0A,$0A,$0C,$0A,$0A,$00,$00,$00    ; 'K'
+    db $08,$08,$08,$08,$0E,$00,$00,$00    ; 'L'
+    db $0A,$0E,$0E,$0A,$0A,$00,$00,$00    ; 'M'
+    db $0A,$0E,$0E,$0E,$0A,$00,$00,$00    ; 'N'
+    db $04,$0A,$0A,$0A,$04,$00,$00,$00    ; 'O'
+    db $0C,$0A,$0C,$08,$08,$00,$00,$00    ; 'P'
+    db $04,$0A,$0A,$04,$02,$00,$00,$00    ; 'Q'
+    db $0C,$0A,$0C,$0A,$0A,$00,$00,$00    ; 'R'
+    db $06,$08,$04,$02,$0C,$00,$00,$00    ; 'S'
+    db $0E,$04,$04,$04,$04,$00,$00,$00    ; 'T'
+    db $0A,$0A,$0A,$0A,$06,$00,$00,$00    ; 'U'
+    db $0A,$0A,$0A,$04,$04,$00,$00,$00    ; 'V'
+    db $0A,$0A,$0E,$0E,$0A,$00,$00,$00    ; 'W'
+    db $0A,$0A,$04,$0A,$0A,$00,$00,$00    ; 'X'
+    db $0A,$0A,$04,$04,$04,$00,$00,$00    ; 'Y'
+    db $0E,$02,$04,$08,$0E,$00,$00,$00    ; 'Z'
+    db $0C,$08,$08,$08,$0C,$00,$00,$00    ; '['
+    db $08,$08,$04,$02,$02,$00,$00,$00    ; '?'
+    db $06,$02,$02,$02,$06,$00,$00,$00    ; ']'
+    db $04,$0A,$00,$00,$00,$00,$00,$00    ; '^'
+    db $00,$00,$00,$00,$0E,$00,$00,$00    ; '_'
+    db $08,$04,$00,$00,$00,$00,$00,$00    ; '`'
+    db $0C,$0A,$0E,$0A,$0A,$00,$00,$00    ; 'a'
+    db $0C,$0A,$0C,$0A,$0C,$00,$00,$00    ; 'b'
+    db $06,$08,$08,$08,$06,$00,$00,$00    ; 'c'
+    db $0C,$0A,$0A,$0A,$0C,$00,$00,$00    ; 'd'
+    db $0E,$08,$0C,$08,$0E,$00,$00,$00    ; 'e'
+    db $0E,$08,$0C,$08,$08,$00,$00,$00    ; 'f'
+    db $06,$08,$0A,$0A,$06,$00,$00,$00    ; 'g'
+    db $0A,$0A,$0E,$0A,$0A,$00,$00,$00    ; 'h'
+    db $0E,$04,$04,$04,$0E,$00,$00,$00    ; 'i'
+    db $02,$02,$02,$0A,$04,$00,$00,$00    ; 'j'
+    db $0A,$0A,$0C,$0A,$0A,$00,$00,$00    ; 'k'
+    db $08,$08,$08,$08,$0E,$00,$00,$00    ; 'l'
+    db $0A,$0E,$0E,$0A,$0A,$00,$00,$00    ; 'm'
+    db $0A,$0E,$0E,$0E,$0A,$00,$00,$00    ; 'n'
+    db $04,$0A,$0A,$0A,$04,$00,$00,$00    ; 'o'
+    db $0C,$0A,$0C,$08,$08,$00,$00,$00    ; 'p'
+    db $04,$0A,$0A,$04,$02,$00,$00,$00    ; 'q'
+    db $0C,$0A,$0C,$0A,$0A,$00,$00,$00    ; 'r'
+    db $06,$08,$04,$02,$0C,$00,$00,$00    ; 's'
+    db $0E,$04,$04,$04,$04,$00,$00,$00    ; 't'
+    db $0A,$0A,$0A,$0A,$06,$00,$00,$00    ; 'u'
+    db $0A,$0A,$0A,$04,$04,$00,$00,$00    ; 'v'
+    db $0A,$0A,$0E,$0E,$0A,$00,$00,$00    ; 'w'
+    db $0A,$0A,$04,$0A,$0A,$00,$00,$00    ; 'x'
+    db $0A,$0A,$04,$04,$04,$00,$00,$00    ; 'y'
+    db $0E,$02,$04,$08,$0E,$00,$00,$00    ; 'z'
+    db $02,$04,$0C,$04,$02,$00,$00,$00    ; '{'
+    db $04,$04,$04,$04,$04,$00,$00,$00    ; '|'
+    db $08,$04,$06,$04,$08,$00,$00,$00    ; '}'
+    db $00,$06,$0C,$00,$00,$00,$00,$00    ; '~'
+    db $0E,$0A,$0A,$0A,$0E,$00,$00,$00    ; '?'
diff --git a/src/term.asm b/src/term.asm
new file mode 100644
index 0000000..9d13b80
--- /dev/null
+++ b/src/term.asm
@@ -0,0 +1,381 @@
+; =============================================================================
+; term.asm - a 40x18 text terminal on the CGB LCD.
+;
+; 4x8 font: each 8x8 background tile holds TWO 4-px characters (left+right).
+; The tilemap is static - one dedicated VRAM tile per screen tile-position -
+; and we (re)build the 16 tile bytes when a character changes. 360 tiles:
+; positions 0-255 in VRAM bank 0, 256-359 in bank 1 (via the tilemap attribute
+; bank bit), which is why this is CGB-only.
+; =============================================================================
+INCLUDE "include/gbos.inc"
+
+DEF rSCY    EQU $FF42
+DEF rSCX    EQU $FF43
+DEF rVBK    EQU $FF4F   ; CGB VRAM bank select
+DEF rBCPS   EQU $FF68   ; CGB BG palette index
+DEF rBCPD   EQU $FF69   ; CGB BG palette data
+
+DEF TCOLS   EQU 40      ; text columns
+DEF TROWS   EQU 18      ; text rows
+DEF TSTRIDE EQU 64      ; bytes per row in the buffer (power of 2 for addressing)
+DEF TERM_BUF EQU $D600  ; 18*64 = 1152 B (WRAMX, above the FS caches)
+
+; terminal scratch (WRAMX gap $D500-$D5FF)
+DEF wGlyphL EQU $D580
+DEF wGlyphR EQU $D582
+DEF wVram   EQU $D584
+
+SECTION "term", ROM0
+
+; -----------------------------------------------------------------------------
+; row_addr - A = row -> HL = TERM_BUF + row*64
+; -----------------------------------------------------------------------------
+row_addr:
+    ld h, 0
+    ld l, a
+    add hl, hl
+    add hl, hl
+    add hl, hl
+    add hl, hl
+    add hl, hl
+    add hl, hl                  ; *64
+    ld de, TERM_BUF
+    add hl, de
+    ret
+
+; -----------------------------------------------------------------------------
+; term_init - CGB palette, static tilemap, clear buffer, LCD on.
+; -----------------------------------------------------------------------------
+term_init::
+    ; LCD off (so VRAM is freely writable)
+    xor a
+    ld [rLCDC], a
+    ld [rSCX], a
+    ld [rSCY], a
+
+    ; BG palette 0: color0 = white (bg), color1 = black (text)
+    ld a, $80
+    ld [rBCPS], a               ; index 0, auto-increment
+    ld a, $FF
+    ld [rBCPD], a               ; c0 lo  (white $7FFF)
+    ld a, $7F
+    ld [rBCPD], a               ; c0 hi
+    xor a
+    ld [rBCPD], a               ; c1 lo  (black)
+    ld [rBCPD], a               ; c1 hi
+    ld [rBCPD], a               ; c2..c3 = black
+    ld [rBCPD], a
+    ld [rBCPD], a
+    ld [rBCPD], a
+
+    ; --- tilemap indices (VRAM bank 0): [pos] = LOW(pos) ---
+    xor a
+    ld [rVBK], a
+    ld de, 0                    ; pos
+    ld c, 0                     ; row
+.mi_row
+    ; compute tilemap row start = $9800 + row*32
+    ld a, c
+    ld l, a
+    ld h, 0
+    add hl, hl
+    add hl, hl
+    add hl, hl
+    add hl, hl
+    add hl, hl                  ; row*32
+    ld a, h
+    add $98
+    ld h, a                     ; HL = $9800 + row*32
+    ld b, TCOLS/2               ; 20 tile columns
+.mi_col
+    ld a, e
+    ld [hl+], a                 ; index = LOW(pos)
+    inc de
+    dec b
+    jr nz, .mi_col
+    inc c
+    ld a, c
+    cp TROWS
+    jr c, .mi_row
+
+    ; --- tilemap attributes (VRAM bank 1): bank bit for pos>=256 ---
+    ld a, 1
+    ld [rVBK], a
+    ld de, 0
+    ld c, 0
+.ma_row
+    ld a, c
+    ld l, a
+    ld h, 0
+    add hl, hl
+    add hl, hl
+    add hl, hl
+    add hl, hl
+    add hl, hl
+    ld a, h
+    add $98
+    ld h, a
+    ld b, TCOLS/2
+.ma_col
+    ld a, d                     ; HIGH(pos): 0 for 0-255, 1 for 256+
+    or a
+    jr z, .ma_bank0
+    ld a, $08                   ; VRAM bank 1 for this tile
+    jr .ma_put
+.ma_bank0
+    xor a
+.ma_put
+    ld [hl+], a
+    inc de
+    dec b
+    jr nz, .ma_col
+    inc c
+    ld a, c
+    cp TROWS
+    jr c, .ma_row
+    xor a
+    ld [rVBK], a
+
+    ; clear the text buffer to spaces
+    ld hl, TERM_BUF
+    ld de, TROWS * TSTRIDE
+.clr
+    ld a, $20                   ; ' '
+    ld [hl+], a
+    dec de
+    ld a, d
+    or e
+    jr nz, .clr
+
+    ret
+
+; term_show - rebuild all tiles then turn the LCD on.
+term_show::
+    call term_redraw
+    ld a, %10010001             ; LCD on, BG on, tile data $8000, map $9800
+    ld [rLCDC], a
+    ret
+
+; -----------------------------------------------------------------------------
+; build_tile - build the 8x8 tile for screen tile-position DE (E=index within
+; bank, D=bank 0/1) from characters B (left) and C (right).
+; -----------------------------------------------------------------------------
+build_tile:
+    ld a, d
+    ld [rVBK], a                ; select VRAM bank for this tile
+    ; VRAM dest = $8000 + E*16
+    ld h, 0
+    ld l, e
+    add hl, hl
+    add hl, hl
+    add hl, hl
+    add hl, hl                  ; E*16
+    ld a, h
+    add $80
+    ld h, a                     ; HL = $8000 + E*16
+    ld a, l
+    ld [wVram], a
+    ld a, h
+    ld [wVram+1], a
+    ; glyphL = FontData + (clamp(B)-32)*8
+    ld a, b
+    call glyph_ptr
+    ld a, l
+    ld [wGlyphL], a
+    ld a, h
+    ld [wGlyphL+1], a
+    ld a, c
+    call glyph_ptr
+    ld a, l
+    ld [wGlyphR], a
+    ld a, h
+    ld [wGlyphR+1], a
+    ; 8 rows: dest[2r] = (L[r]<<4)|R[r] ; dest[2r+1] = 0
+    ld b, 8                     ; row counter
+    ld a, [wVram]
+    ld l, a
+    ld a, [wVram+1]
+    ld h, a                     ; HL = VRAM dest
+.brow
+    push hl                     ; save VRAM ptr
+    ; left glyph byte
+    ld a, [wGlyphL]
+    ld e, a
+    ld a, [wGlyphL+1]
+    ld d, a
+    ld a, [de]
+    swap a
+    and $F0
+    ld c, a                     ; left nibble -> high
+    ; right glyph byte
+    ld a, [wGlyphR]
+    ld e, a
+    ld a, [wGlyphR+1]
+    ld d, a
+    ld a, [de]
+    and $0F
+    or c                        ; A = combined row
+    pop hl
+    ld [hl+], a                 ; plane 0
+    xor a
+    ld [hl+], a                 ; plane 1 = 0
+    ; advance both glyph pointers
+    ld a, [wGlyphL]
+    inc a
+    ld [wGlyphL], a
+    jr nz, .nl
+    ld a, [wGlyphL+1]
+    inc a
+    ld [wGlyphL+1], a
+.nl
+    ld a, [wGlyphR]
+    inc a
+    ld [wGlyphR], a
+    jr nz, .nr
+    ld a, [wGlyphR+1]
+    inc a
+    ld [wGlyphR+1], a
+.nr
+    dec b
+    jr nz, .brow
+    ret
+
+; glyph_ptr - A = char -> HL = FontData + (clamp(A,32,127)-32)*8
+glyph_ptr:
+    cp 32
+    jr nc, .ok
+    ld a, 32
+.ok
+    cp 128
+    jr c, .ok2
+    ld a, 32
+.ok2
+    sub 32
+    ld l, a
+    ld h, 0
+    add hl, hl
+    add hl, hl
+    add hl, hl                  ; *8
+    ld de, FontData
+    add hl, de
+    ret
+
+; -----------------------------------------------------------------------------
+; term_redraw - rebuild every tile from the text buffer. (LCD should be off, or
+; call during VBlank for a full frame.)
+; -----------------------------------------------------------------------------
+term_redraw::
+    ld c, 0                     ; row
+.row
+    push bc
+    ld a, c
+    call row_addr               ; HL = &buffer[row][0]
+    ; iterate 20 tile columns
+    ld b, 0                     ; tile column
+.col
+    push bc
+    push hl
+    ; DE = pos = row*20 + tcol : maintain via a running counter is simpler, but
+    ; recompute here: pos = C(row)*20 + B(tcol)
+    ld a, c
+    ld l, a
+    ld h, 0
+    add hl, hl
+    add hl, hl                  ; row*4
+    ld d, h
+    ld e, l
+    add hl, hl
+    add hl, hl                  ; row*16
+    add hl, de                  ; row*20
+    ld a, b
+    add l
+    ld e, a
+    ld a, h
+    adc 0
+    ld d, a                     ; DE = pos
+    pop hl                      ; &buffer[row][0]
+    push hl
+    ; charL = buffer[row][tcol*2], charR = [+1]
+    ld a, b
+    add a                       ; tcol*2
+    add l
+    ld l, a
+    ld a, h
+    adc 0
+    ld h, a
+    ld a, [hl+]
+    ld b, a                     ; charL
+    ld a, [hl]
+    ld c, a                     ; charR
+    push de
+    call build_tile             ; DE=pos, B=charL, C=charR
+    pop de
+    pop hl
+    pop bc
+    inc b
+    ld a, b
+    cp TCOLS/2
+    jr c, .col
+    xor a
+    ld [rVBK], a
+    pop bc
+    inc c
+    ld a, c
+    cp TROWS
+    jr c, .row
+    ret
+
+; -----------------------------------------------------------------------------
+; term_puts - HL = NUL-terminated string, B = row, C = col : write into buffer.
+; -----------------------------------------------------------------------------
+term_puts::
+    push hl
+    ld a, b
+    call row_addr               ; HL = &buffer[row]
+    ld a, c
+    add l
+    ld l, a
+    ld a, h
+    adc 0
+    ld h, a                     ; HL = &buffer[row][col]
+    ld d, h
+    ld e, l                     ; DE = dest
+    pop hl                      ; HL = string
+.l
+    ld a, [hl+]
+    or a
+    ret z
+    ld [de], a
+    inc de
+    jr .l
+
+; -----------------------------------------------------------------------------
+; term_test - fill the screen with a test pattern and show it.
+; -----------------------------------------------------------------------------
+term_test::
+    ld hl, .s0
+    ld bc, $0000
+    call term_puts
+    ld hl, .s1
+    ld bc, $0100
+    call term_puts
+    ld hl, .s2
+    ld bc, $0200
+    call term_puts
+    ld hl, .s3
+    ld bc, $0500
+    call term_puts
+    ld hl, .s4
+    ld bc, $0A00
+    call term_puts
+    ld hl, .s5
+    ld bc, $1100
+    call term_puts
+    call term_show
+    ret
+.s0 db "GBOS TERMINAL - 40 COLUMNS x 18 ROWS", 0
+.s1 db "ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789", 0
+.s2 db "the quick brown fox jumps over the dog.", 0
+.s3 db "0123456789012345678901234567890123456789", 0
+.s4 db "row 10: some text in the middle here...", 0
+.s5 db "LAST ROW 17 - exercises bank-1 tiles!!!", 0