Repeated map values can take less room as a count and a value—but only if the output buffer is big enough.
Use it in a project
INPUT_COUNT selects 0–8 bytes B0–B7. Encode consecutive equal values as positive count/value byte pairs in R0–R15, up to CAPACITY 0–16. Preflight all runs before writing. Invalid bounds return VALID=0, ERROR=1; insufficient output returns VALID=1, ERROR=2. Both failures leave output bytes unchanged and WRITTEN=0. Success returns ERROR=0 and the even byte count; unused output bytes are preserved.
Include component.inc once (MOS callers include the shared 6502 implementation through it). Reserve the public fields in this exact order: WRITTEN, ERROR, VALID, INPUT_COUNT, CAPACITY, B0, B1, B2, B3, B4, B5, B6, B7, R0, R1, R2, R3, R4, R5, R6, R7, R8, R9, R10, R11, R12, R13, R14, R15. They occupy 29 contiguous bytes. Also reserve the private workspace: T0, T1, T2, T3, T4, EXPR_SAVE (6 bytes). Array names in the explanation refer to the numbered consecutive bindings, such as C0, C1 and so on.
Call bounded_rle_encoder_init once at the intended reset boundary. It clears inputs as well as outputs and workspace. Set your starting state and configuration afterwards, then call bounded_rle_encoder_step whenever the described event occurs. The helper reads no device, waits for no frame and installs no interrupt handler.
| Ownership | Fields |
|---|---|
| Retain between calls | R0–R15 as the owned destination |
| Supply before service | INPUT_COUNT, CAPACITY and B0–B7 |
| Read after service | WRITTEN, ERROR and VALID |
A first pass counts runs and proves that every pair will fit. A second pass writes the pairs. That extra scan gives the caller an all-or-nothing output contract.
Reusable initialisation and service
; SPDX-License-Identifier: MIT
; Encode a Small Run-length Block. See README.md for the byte interface and ownership.
; init clears the public fields and private workspace; set inputs afterwards.
bounded_rle_encoder_init:
move.b #0,d0
move.b d0,WRITTEN
move.b d0,ERROR
move.b d0,VALID
move.b d0,INPUT_COUNT
move.b d0,CAPACITY
move.b d0,B0
move.b d0,B1
move.b d0,B2
move.b d0,B3
move.b d0,B4
move.b d0,B5
move.b d0,B6
move.b d0,B7
move.b d0,R0
move.b d0,R1
move.b d0,R2
move.b d0,R3
move.b d0,R4
move.b d0,R5
move.b d0,R6
move.b d0,R7
move.b d0,R8
move.b d0,R9
move.b d0,R10
move.b d0,R11
move.b d0,R12
move.b d0,R13
move.b d0,R14
move.b d0,R15
move.b d0,T0
move.b d0,T1
move.b d0,T2
move.b d0,T3
move.b d0,T4
move.b d0,EXPR_SAVE
rts
bounded_rle_encoder_step:
; WRITTEN = 0
move.b #0,d0
move.b d0,WRITTEN
; ERROR = 1
move.b #1,d0
move.b d0,ERROR
; VALID = 0
move.b #0,d0
move.b d0,VALID
; if INPUT_COUNT > 8:
move.b INPUT_COUNT,d0
cmp.b #8,d0
bhi bounded_rle_encoder_L1
jmp bounded_rle_encoder_L2
bounded_rle_encoder_L1:
; return
rts
jmp bounded_rle_encoder_L3
bounded_rle_encoder_L2:
bounded_rle_encoder_L3:
; if CAPACITY > 16:
move.b CAPACITY,d0
cmp.b #16,d0
bhi bounded_rle_encoder_L4
jmp bounded_rle_encoder_L5
bounded_rle_encoder_L4:
; return
rts
jmp bounded_rle_encoder_L6
bounded_rle_encoder_L5:
bounded_rle_encoder_L6:
; VALID = 1
move.b #1,d0
move.b d0,VALID
; ERROR = 0
move.b #0,d0
move.b d0,ERROR
; if INPUT_COUNT == 0:
move.b INPUT_COUNT,d0
cmp.b #0,d0
beq bounded_rle_encoder_L7
jmp bounded_rle_encoder_L8
bounded_rle_encoder_L7:
; return
rts
jmp bounded_rle_encoder_L9
bounded_rle_encoder_L8:
bounded_rle_encoder_L9:
; T0 = 1
move.b #1,d0
move.b d0,T0
; T1 = B[0]
moveq #0,d1
move.b #0,d1
lea B0,a0
move.b 0(a0,d1.w),d0
move.b d0,T1
; T2 = 1
move.b #1,d0
move.b d0,T2
; while T0 < INPUT_COUNT:
bounded_rle_encoder_L10:
move.b T0,d0
move.b INPUT_COUNT,d1
cmp.b d1,d0
bcs bounded_rle_encoder_L11
jmp bounded_rle_encoder_L12
bounded_rle_encoder_L11:
; T3 = B[T0]
moveq #0,d1
move.b T0,d1
lea B0,a0
move.b 0(a0,d1.w),d0
move.b d0,T3
; if T3 != T1:
move.b T3,d0
move.b T1,d1
cmp.b d1,d0
bne bounded_rle_encoder_L13
jmp bounded_rle_encoder_L14
bounded_rle_encoder_L13:
; T2 = T2 + 1
move.b T2,d0
add.b #1,d0
move.b d0,T2
; T1 = T3
move.b T3,d0
move.b d0,T1
jmp bounded_rle_encoder_L15
bounded_rle_encoder_L14:
bounded_rle_encoder_L15:
; T0 = T0 + 1
move.b T0,d0
add.b #1,d0
move.b d0,T0
jmp bounded_rle_encoder_L10
bounded_rle_encoder_L12:
; T2 = T2 << 1
move.b T2,d0
lsl.b #1,d0
move.b d0,T2
; if T2 > CAPACITY:
move.b T2,d0
move.b CAPACITY,d1
cmp.b d1,d0
bhi bounded_rle_encoder_L16
jmp bounded_rle_encoder_L17
bounded_rle_encoder_L16:
; ERROR = 2
move.b #2,d0
move.b d0,ERROR
; return
rts
jmp bounded_rle_encoder_L18
bounded_rle_encoder_L17:
bounded_rle_encoder_L18:
; T0 = 0
move.b #0,d0
move.b d0,T0
; T1 = 0
move.b #0,d0
move.b d0,T1
; while T0 < INPUT_COUNT:
bounded_rle_encoder_L19:
move.b T0,d0
move.b INPUT_COUNT,d1
cmp.b d1,d0
bcs bounded_rle_encoder_L20
jmp bounded_rle_encoder_L21
bounded_rle_encoder_L20:
; T2 = B[T0]
moveq #0,d1
move.b T0,d1
lea B0,a0
move.b 0(a0,d1.w),d0
move.b d0,T2
; T3 = 1
move.b #1,d0
move.b d0,T3
; T0 = T0 + 1
move.b T0,d0
add.b #1,d0
move.b d0,T0
; while T0 < INPUT_COUNT:
bounded_rle_encoder_L22:
move.b T0,d0
move.b INPUT_COUNT,d1
cmp.b d1,d0
bcs bounded_rle_encoder_L23
jmp bounded_rle_encoder_L24
bounded_rle_encoder_L23:
; T4 = B[T0]
moveq #0,d1
move.b T0,d1
lea B0,a0
move.b 0(a0,d1.w),d0
move.b d0,T4
; if T4 != T2:
move.b T4,d0
move.b T2,d1
cmp.b d1,d0
bne bounded_rle_encoder_L25
jmp bounded_rle_encoder_L26
bounded_rle_encoder_L25:
; break
jmp bounded_rle_encoder_L24
jmp bounded_rle_encoder_L27
bounded_rle_encoder_L26:
bounded_rle_encoder_L27:
; T3 = T3 + 1
move.b T3,d0
add.b #1,d0
move.b d0,T3
; T0 = T0 + 1
move.b T0,d0
add.b #1,d0
move.b d0,T0
jmp bounded_rle_encoder_L22
bounded_rle_encoder_L24:
; R[T1] = T3
move.b T3,d0
move.b d0,d2
moveq #0,d1
move.b T1,d1
lea R0,a0
move.b d2,0(a0,d1.w)
; T1 = T1 + 1
move.b T1,d0
add.b #1,d0
move.b d0,T1
; R[T1] = T2
move.b T2,d0
move.b d0,d2
moveq #0,d1
move.b T1,d1
lea R0,a0
move.b d2,0(a0,d1.w)
; T1 = T1 + 1
move.b T1,d0
add.b #1,d0
move.b d0,T1
jmp bounded_rle_encoder_L19
bounded_rle_encoder_L21:
; WRITTEN = T1
move.b T1,d0
move.b d0,WRITTEN
rts
component_end:Machine contract
D0–D3/A0–A1/CCR may change; D4–D7/A2–A6 are preserved. A direct BSR/JSR needs four return-address stack bytes. The helper makes no nested calls and adds no other stack use. Its named scratch is private workspace, not a pointer, a ROM variable or shared zero-page space that the game may borrow concurrently. The samples use fixed symbol bindings. Scope or rename every label and binding, including the component_end measurement marker, and allocate a separate block for each independent instance. Different helpers can also use common field names, so combining them needs separate namespaces or explicitly renamed bindings. Give each block one mutation owner.
The A500 OCS diagnostic cold-boots an OFS disk, owns the Copper and bitplane and disables interrupt/DMA sources. The helper does not take over hardware or allocate Chip RAM.
Adapt the caller
demo.asm supplies the bindings and include order. exercise.inc first runs table cases, then a separate client retaining state between calls. A game can call the public step entry directly and remove the diagnostic packet and display. Do not copy the diagnostic’s hardware takeover into an existing game loop.
Build with make ASM198X=/path/to/asm198x and BUILD198X=/path/to/build198x. The Makefile produces normal, custom, deliberately wrong and repaired programs. Tools must already be installed; the sample downloads nothing.
Bindings and include order
section code,code_c
CUSTOM = 0
BROKEN = 0
COYOTE_STEPS = 4
BUFFER_STEPS = 4
TEXT_CAP = 8
PROPERTY_COUNT = 16
STREAM_BYTES = 13
CRC_POLY = 7
include "caller.inc"
include "exercise.inc"
component_start:
include "component.inc"
even
include "data.inc"
include "cases.inc"
even
include "display.inc"
include "../../framework/turn-deadline/font.inc"
code_end:
STATE:
WRITTEN: dc.b 0
ERROR: dc.b 0
VALID: dc.b 0
INPUT_COUNT: dc.b 0
CAPACITY: dc.b 0
B0: dc.b 0
B1: dc.b 0
B2: dc.b 0
B3: dc.b 0
B4: dc.b 0
B5: dc.b 0
B6: dc.b 0
B7: dc.b 0
R0: dc.b 0
R1: dc.b 0
R2: dc.b 0
R3: dc.b 0
R4: dc.b 0
R5: dc.b 0
R6: dc.b 0
R7: dc.b 0
R8: dc.b 0
R9: dc.b 0
R10: dc.b 0
R11: dc.b 0
R12: dc.b 0
R13: dc.b 0
R14: dc.b 0
R15: dc.b 0
T0: dc.b 0
T1: dc.b 0
T2: dc.b 0
T3: dc.b 0
T4: dc.b 0
EXPR_SAVE: dc.b 0
even
ready: dc.w 0
preview: dcb.b 8,0
OUTPUT: dcb.b 1300,0
even
signature: dc.b "198XWORLDKIT000",0
even
copperlist:
dc.w $08e,$2c81,$090,$2cc1,$092,$0038,$094,$00d0
dc.w $100,$1200,$102,0,$104,0,$108,0,$10a,0
dc.w $180,$012,$182,$fff
copbpl:
dc.w $0e0,0,$0e2,0
dc.w $ffff,$fffe
plane: dcb.b 10240,0A client retaining state between calls
preview_rules:
ifne CUSTOM
; INPUT_COUNT = 8
move.b #8,d0
move.b d0,INPUT_COUNT
; CAPACITY = 16
move.b #16,d0
move.b d0,CAPACITY
; B0 = 0
move.b #0,d0
move.b d0,B0
; B1 = 1
move.b #1,d0
move.b d0,B1
; B2 = 2
move.b #2,d0
move.b d0,B2
; B3 = 3
move.b #3,d0
move.b d0,B3
; B4 = 4
move.b #4,d0
move.b d0,B4
; B5 = 5
move.b #5,d0
move.b d0,B5
; B6 = 6
move.b #6,d0
move.b d0,B6
; B7 = 7
move.b #7,d0
move.b d0,B7
ifne BROKEN
; CAPACITY = 0
move.b #0,d0
move.b d0,CAPACITY
endif
bsr checked_step
bsr emit_state
rts
else
; INPUT_COUNT = 6
move.b #6,d0
move.b d0,INPUT_COUNT
; CAPACITY = 16
move.b #16,d0
move.b d0,CAPACITY
; B0 = 65
move.b #65,d0
move.b d0,B0
; B1 = 65
move.b #65,d0
move.b d0,B1
; B2 = 65
move.b #65,d0
move.b d0,B2
; B3 = 66
move.b #66,d0
move.b d0,B3
; B4 = 66
move.b #66,d0
move.b d0,B4
; B5 = 67
move.b #67,d0
move.b d0,B5
ifne BROKEN
; CAPACITY = 0
move.b #0,d0
move.b d0,CAPACITY
endif
bsr checked_step
bsr emit_state
endif
rtsCustomise it
Encode eight distinct values, the worst expansion for this small block. The maintained custom caller changes its inputs or client sequence rather than hiding a different implementation behind the same label. Compare every state of the two clients, not only their final display.
Custom client selection
section code,code_c
CUSTOM = 1
BROKEN = 0
COYOTE_STEPS = 2
BUFFER_STEPS = 2
TEXT_CAP = 4
PROPERTY_COUNT = 16
STREAM_BYTES = 13
CRC_POLY = 49
include "caller.inc"
include "exercise.inc"
component_start:
include "component.inc"
even
include "data.inc"
include "cases.inc"
even
include "display.inc"
include "../../framework/turn-deadline/font.inc"
code_end:
STATE:
WRITTEN: dc.b 0
ERROR: dc.b 0
VALID: dc.b 0
INPUT_COUNT: dc.b 0
CAPACITY: dc.b 0
B0: dc.b 0
B1: dc.b 0
B2: dc.b 0
B3: dc.b 0
B4: dc.b 0
B5: dc.b 0
B6: dc.b 0
B7: dc.b 0
R0: dc.b 0
R1: dc.b 0
R2: dc.b 0
R3: dc.b 0
R4: dc.b 0
R5: dc.b 0
R6: dc.b 0
R7: dc.b 0
R8: dc.b 0
R9: dc.b 0
R10: dc.b 0
R11: dc.b 0
R12: dc.b 0
R13: dc.b 0
R14: dc.b 0
R15: dc.b 0
T0: dc.b 0
T1: dc.b 0
T2: dc.b 0
T3: dc.b 0
T4: dc.b 0
EXPR_SAVE: dc.b 0
even
ready: dc.w 0
preview: dcb.b 8,0
OUTPUT: dcb.b 1300,0
even
signature: dc.b "198XWORLDKIT000",0
even
copperlist:
dc.w $08e,$2c81,$090,$2cc1,$092,$0038,$094,$00d0
dc.w $100,$1200,$102,0,$104,0,$108,0,$10a,0
dc.w $180,$012,$182,$fff
copbpl:
dc.w $0e0,0,$0e2,0
dc.w $ffff,$fffe
plane: dcb.b 10240,0The custom diagnostic row is 10 00 01 08 10 00 01 02. The full client trace in the source explains how it gets there.
Mistake and repair
The wrong caller passes no output capacity for a nonempty block. The encoder rejects the whole request and preserves its output buffer. Pass the actual reserved capacity, rather than a count that has not yet been filled.
The BROKEN branch lives in the sequential client, not the reusable helper. wrong.asm selects it; repaired.asm restores the normal call sequence. The repaired executable matches the normal program (for Spectrum, compare loaded CODE rather than the tape filename).
Select the deliberately wrong client
section code,code_c
CUSTOM = 0
BROKEN = 1
COYOTE_STEPS = 4
BUFFER_STEPS = 4
TEXT_CAP = 8
PROPERTY_COUNT = 16
STREAM_BYTES = 13
CRC_POLY = 7
include "caller.inc"
include "exercise.inc"
component_start:
include "component.inc"
even
include "data.inc"
include "cases.inc"
even
include "display.inc"
include "../../framework/turn-deadline/font.inc"
code_end:
STATE:
WRITTEN: dc.b 0
ERROR: dc.b 0
VALID: dc.b 0
INPUT_COUNT: dc.b 0
CAPACITY: dc.b 0
B0: dc.b 0
B1: dc.b 0
B2: dc.b 0
B3: dc.b 0
B4: dc.b 0
B5: dc.b 0
B6: dc.b 0
B7: dc.b 0
R0: dc.b 0
R1: dc.b 0
R2: dc.b 0
R3: dc.b 0
R4: dc.b 0
R5: dc.b 0
R6: dc.b 0
R7: dc.b 0
R8: dc.b 0
R9: dc.b 0
R10: dc.b 0
R11: dc.b 0
R12: dc.b 0
R13: dc.b 0
R14: dc.b 0
R15: dc.b 0
T0: dc.b 0
T1: dc.b 0
T2: dc.b 0
T3: dc.b 0
T4: dc.b 0
EXPR_SAVE: dc.b 0
even
ready: dc.w 0
preview: dcb.b 8,0
OUTPUT: dcb.b 1300,0
even
signature: dc.b "198XWORLDKIT000",0
even
copperlist:
dc.w $08e,$2c81,$090,$2cc1,$092,$0038,$094,$00d0
dc.w $100,$1200,$102,0,$104,0,$108,0,$10a,0
dc.w $180,$012,$182,$fff
copbpl:
dc.w $0e0,0,$0e2,0
dc.w $ffff,$fffe
plane: dcb.b 10240,0
Cost and alternatives
The measured helper extent is 748 code bytes; public state is 29 bytes and private scratch is 6 bytes. There are no immutable helper tables. Caller data, the test cases, return-address stack, alignment and display buffers cost additional space. These are assembled sizes, not cycle measurements.
Two bounded scans over at most eight bytes, plus sixteen destination writes in the worst case. A streaming encoder can avoid preflight but must report partial output and support continuation explicitly.
RLE can expand data: eight distinct bytes need sixteen encoded bytes. This uses the existing count/value idea, not a new project-wide asset container. A decoder must still validate its own lengths and output bounds.
Execution evidence
Four variants run on A500 OCS PAL, Kickstart1.3 OFS cold boot, using Asm198x 0.0.72, Emu198x 0.44.2 and Build198x 0.2.9. Each executes 34 boundary/varied/error table cases, dirty initialisation and the separate retained-state client. An independent mathematical/container model checks every public byte after every call. This is bounded example coverage, not exhaustive correctness for every possible state.
The records retain loaded immutable code, completion, register sentinels, full packets, source/tool/firmware hashes and captures. The public gate validates these retained observations and rejects altered evidence; it does not run restricted firmware. No firmware is distributed. The eight displayed bytes are the first eight public fields, zero-padded when shorter; the packet includes every field.
Where it fits
Bounded RLE decoding: Encoding and decoding share count/value semantics but need different buffer and failure contracts.