Skip to content

SINTRAN III Version L — Install Notes

Status: SCAFFOLD. Verified facts only. Source: SINTRAN-L distribution archive.

Verified facts

  • Golden disk dump: FILE-INFO\BIGDISK0-L.TXT (PACK-ONE, 38400 pages).
  • Three VSX distribution floppies: VSXL1.TXT, VSXL2.TXT, VSXL3.TXT.
  • Has a dedicated startup script: Admin\START-SINTRAN-MULTIUSER.TXT.
  • L07 symbol set includes XMSG-SYMBOL-LIST.SYMB.TXT.

Source files

  • Admin\System initialisation.txt, Admin\START-SINTRAN-MULTIUSER.TXT, Admin\SINTRAN-COMMANDS-L.TXT
  • FILE-INFO\BIGDISK0-L.TXT, VSXL1/2/3.TXT

Installing the ND-500/5000 Monitor and ND-5000 Microcode

Applies only to ND-500/ND-5000 systems. Skip entirely for ND-100/ND-110-only installs.

1. Install the ND-500/5000 System Package (product ND-211305)

Source: ../../../SINTRAN/Release-Documentation/ND-860230-6-EN Sintran III - Release Information - L-Version.md (lines 710-746). Done after @ENTER-DIRECTORY on the main directory, logged in as user SYSTEM.

  1. Insert the diskette for the ND-500/5000 System Package for version L (ND-211305) into FLOPPY-DISC-1 unit 0.
  2. @ENTER-DIRECTORY,,FLOPPY-DISC-1,0
  3. Replace the ND-500 Background Monitor: 
    @DELETE-FILE ND-500-MON:PROG
@COPY-FILE "ND-500-MON-J:PROG" (211305:FL)ND-500-MON-J:PROG
  4. Replace the ND-500 Swapper: 
    @DELETE-FILE SWAPPER:PSEG
@DELETE-FILE SWAPPER:DSEG
@COPY-FILE "SWAPPER-K:PSEG" (211305:F-U)SWAPPER-K:PSEG
@COPY-FILE "SWAPPER-K:DSEG" (211305:F-U)SWAPPER-K:DSEG
  5. @RELEASE-DIRECTORY 211305, then remove the floppy.

Version constraints (same doc, lines 3547-3551): only ND-500/5000 Background Monitor version J or later (ND-210333) and ND-500/5000 Swapper version K (ND-211034) may run under SINTRAN III version L. As of L, both ship combined as one product: the ND-500/5000 System Package (ND-211305), which also includes the ND-500 Place-Library.

2. ND-5000 systems only: install the microcode

Same source doc, lines 748-779.

  1. Insert the ND-5000 microprogram diskette matching your model (ND-5200/5400/5500/5700/5800) into FLOPPY-DISC-1 unit 0.
  2. @ENTER-DIRECTORY,,FLOPPY-DISC-1,0
  3. Copy the microcode to disk:
    • ND-5200/5400/5500/5700/5800: @COPY-FILE CONTROL-STORE:DATA (211:)MIC-5xxx-2-500:DATA (substitute xxx = 200/400/500/700/800)
    • ND-5900: @COPY-FILE CONTROL-1-STORE:DATA (211:)MIC-5800-2-500:DATA, repeated to CONTROL-2-STORE:DATA etc. depending on the ND-5900 model.
  4. @RELEASE-DIRECTORY 211, then remove the floppy.

3. Run the cold-start mode file

Whether or not step 2 applied: 

@MODE HENT-MODE:MODE.,

4. Starting the Monitor — RESOLVED by live test (2026-07-09)

CORRECTION (superseded by verified fact below): an earlier version of this doc guessed the Monitor was loaded via RT-LOADER READ-BINARY ND500-MONITOR 62. That filename came from ../research/HDD-IMAGE-FINDINGS.md section 1's disk-image table (WD0-M.IMG, HD0.IMG), both explicitly marked M-version images — not L. It was wrongly presented here as applying to L. Confirmed wrong by live test on an L system: NO SUCH FILE NAME.

Verified fact: ND-500-MON:PROG needs NO RT-LOADER/HENT-MODE load step at all. It is an ordinary, self-contained SINTRAN background program. It is started by simply typing its name at the terminal, exactly like a compiler subsystem — confirmed both by the manual and by a live run:

  • Manual: ../../../Reference-Manuals/ND-60.136.04A ND-500 Loader Monitor.md lines 729-822 — @ND-500 or the full name @ND-500-MONITOR starts the Monitor; most installations abbreviate it to ND-500.
  • Live test (2026-07-09): copied ND-500-MON-J04:PROG from floppy 211305B02-XX-01D to disk and ran @ND-500. Output: 
    ND-500/5000 MONITOR  Version J04 88. 6.16 /     6. 6
ND-500(0) error:      No ND-500(0) CPU found
    This confirms the Monitor loads and runs correctly as a plain :PROG — no RT-LOADER step, no HENT-MODE entry, no segment load needed for the :PROG itself.

"No ND-500(0) CPU found" is a separate, later-stage issue — not a load/install problem. It is SINTRAN's boot-time hardware-presence flag never getting set. Per ../../../SINTRAN/ND500/ND500-BUS-INTERFACE-REFERENCE.md section 8.1: at SINTRAN startup, the routine CH5CPUPRESENT (PH-P2-OPPSTART.NPL:3895-3945) probes each generated ND-500 CPU datafield — an IOX read of RSTA5 via the 3022/5015 interface (or, on fault, an Octobus/Samson probe) — and only if that succeeds does it flag CPUAVAILABLE/set 5ALIVE. The Background Monitor's own runtime check reads that flag and refuses to proceed if it was never set. Fix path (UNVERIFIED which applies to your setup): either the physical 3022/5015 interface card isn't answering, or — if running under an emulator — the ND-500 CPU device isn't configured/attached (the RE task handoff, ../../../SINTRAN/ND500/ND500-L-RELEASE-RE-TASK-HANDOFF.md section 6.2, notes RetroCore needs "the 3022 device configured (thumbwheel 0, IOX base 60)" for this presence test to pass).

RESOLVED (2026-07-09): confirmed by live test with a bus-interface emulation that answers the CH5CPUPRESENT probe. With enough of the 3022/5015 bus interface emulated to make SINTRAN's presence probe succeed, @ND-500 now reaches the interactive Monitor prompt cleanly:

@ND-500
ND-500/5000 MONITOR  Version J04 88. 6.16 / 88. 8.17
N500: help
Command:
[... full command list, see below ...]
N500:

Note the two version-date fields changed from the earlier failed run (88. 6.16 / 6. 6, CPU not found) to (88. 6.16 / 88. 8.17, CPU found) — the first date is the Monitor build date (fixed), the second is read from the ND-500 CPU/microcode side and is only populated once a CPU is actually detected. UNVERIFIED: exact field semantics (e.g. whether the second date is a microcode build date) — not confirmed from any manual read so far, only inferred from this before/after comparison.

Live-verified N500: command reference (Version J04, HELP output, 2026-07-09)

The full interactive command set the Monitor advertises via HELP — this is a direct transcript, useful as a live cross-check against ../../../Reference-Manuals/ND-60.136.04A ND-500 Loader Monitor.md's alphabetic command list (manual lines 8177-8210+). Spot-check: RESTART-PROCESS is present, consistent with the release doc's claim (line 2764-2766) that it was reintroduced in L after being removed in the J-version Monitor — the version string here is J04, so this confirms the release-doc claim against a real J04 binary's own advertised command set.

GO  <Address: >
CONTINUE
RUN
HELP  <Command: >
OUTPUT-FILE  <File name: >
CC
EXIT
RECOVER-DOMAIN  <Domain name: >
LOOK-AT-PROGRAM  <Address: >,[<Domain name>]
LOOK-AT-DATA  <Address: >,[<Domain name>]
LOOK-AT-STACK
LOOK-AT-RELATIVE  <Relative to: >
LOOK-AT-REGISTER  [<Register name>]
LOOK-AT-FILE  <Address: >,<File name: >
INSPECT-DUMP  <File name: >
RESET-INSPECT-DUMP
MAIN-FORMAT  <Format: >
EXTRA-FORMAT  <Format: >
TRACE  <Address: >,[<Datatype: >]
RESET-TRACE
GUARD  <Address: >,<Datatype: >,[<Lower limit>],[<Upper limit>]
RESET-GUARD
BRANCH-TRACE  <Start address: >,<Min trace: >,<Max trace: >
RESET-BRANCH-TRACE
CALL-TRACE  <Start address: >,<Min trace: >,<Max trace: >
RESET-CALL-TRACE
BREAK  <Address: >,[<Count>],[<Command>]
TEMPORARY-BREAK  <Address: >,[<Count>],[<Command>]
RESET-LAST-BREAK
EXHIBIT-ADDRESS  <Address in program: >,<Address of variable: >,<Datatype: >
RESET-BREAKS  [<Break number>]
DEBUG-STATUS
DEBUGGER  [<Domain name>]
SPECIAL-DEBUGGER  <File name: >,<Segment number: >,[<Domain name>]
STEP  [<Step start address, [Execution start address], [Count]>]
PLACE-DOMAIN  <Domain name: >
DEBUG-PLACE  <Domain name: >
RESIDENT-PLACE  <Domain name: >
LOCAL-TRAP-ENABLE  <Label: >,<Trap condition: >
SYSTEM-TRAP-ENABLE  <Trap condition: >
LOCAL-TRAP-DISABLE  <Trap condition: >
SYSTEM-TRAP-DISABLE  <Trap condition: >
RESET-DEBUG
STATUS
ENABLED-TRAPS
RESTART-PROCESS  <Process name: >
OPEN-FILE  <File name: >,<Connect file number: >,<Access: >
CLOSE-FILE  <Connect number: >
SET-BLOCK-SIZE  <Connect number: >,<Size(in bytes): >
LIST-OPEN-FILES
SET-HISTOGRAM  <Start address: >,<Max address: >,<Number of intervals: >
PRINT-HISTOGRAM
STOP-HISTOGRAM
START-HISTOGRAM
RELEASE-HISTOGRAM
TIME-USED
WHO-IS-ON
DEFINE-MACRO  <Macro name: >
EXECUTE-MACRO  <Macro name: >,[<Parameters: >]
ERASE-MACRO  <Macro names: >
LIST-MACRO  <Macro names: >
DUMP-MACRO  <Macro name: >
RESUME-MACRO
ABORT-BATCH-ON-ERROR  <On/Off: >
AUTOMATIC-ERROR-MESSAGE
RESET-AUTOMATIC-ERROR-MESSAGE
SET-MEMORY-CONTENTS  <From address: >,<Up to address: >,<Value: >,<Datatype: >
SET-FLAG  <Process number: >,<Value: >
GET-FLAG  <Process number: >
DEFINE-MEMORY-CONFIGURATION  <ND-100 page number for ND-500 physical address 0: >
MEMORY-CONFIGURATION
VERSION  [<>],[<>]
MASTER-CLEAR
RUN-SELFTEST
RESET-CPU
INIT-TRACER  <Cycle: >,<Mode: >,<Trigger: >,<CSA: >,<Clear trace memory(yes/no): >
ARM-TRACER
DISARM-TRACER
CLEAR-TRACE-ADDRESS
CLEAR-TRACE-MEMORY
DUMP-TRACE-MEMORY
EXAMINE-TRACE
READ-TRACE-FILE  <File name: >
WRITE-TRACE-FILE  <File name: >
LOAD-CONTROL-STORE  <File name: >,<Start address: >,<Number of words: >
COMPARE-CONTROL-STORE  <File name: >,<Start address: >,<Number of words: >,<Max number of faults: >
LOOK-AT-CONTROL-STORE  <Address: >
LOOK-AT-RESIDENT-MEMORY  <Address: >
LOOK-AT-HARDWARE  <Register name: >
MICRO-START  <Start address: >
MICRO-STOP
SET-PRIORITY  <ND-500 process number: >,<ND-500 priority: >
START-PROCESS-LOG-ALL
START-PROCESS-LOG-ONE  <Process number: >
PROCESS-LOG-ALL  <First process: >,<Interval: >
PROCESS-LOG-ONE  <Process number: >,<Interval: >
PRINT-PROCESS-LOG  <First process: >
SWAPPING-LOG  <Interval: >
RELEASE-LOG-BUFFER
ATTACH-PROCESS  <Process number: >
LOOK-AT-PHYSICAL-SEGMENT  <Address: >,<Segment number: >
LOOK-AT-SRF  <SRF-address: >
SET-SEGMENT-LIMITS  <SEGMENT NUMBER: >,<Type(P or D): >,<Min number of pages: >,<Max number of pages:>,[<Process number.>]
FIX-SEGMENT-SCATTERED  <Segment number: >,<Type(P or D): >,<Low address: >,<High address: >
FIX-SEGMENT-CONTIGUOUS  <Segment number: >,<Type(P or D): >,<Low address: >,<High address: >
FIX-SEGMENT-ABSOLUTE  <Segment number: >,<Type(P or D): >,<Low address: >,<High address: >,<Physical page: >
UNFIX-SEGMENT  <Segment number: >,<Type (P or D): >
LIST-SYSTEM-PARAMETERS
SET-SYSTEM-PARAMETERS
VALUE-ENTRIES  <Entries: >
START-MONCALL-LOG  <Own/All: >
PRINT-MONCALL-LOG
STOP-MONCALL-LOG
DEFINE-STANDARD-DOMAIN  <Standard domain name: >,<Domain name: >
DELETE-STANDARD-DOMAIN  <Standard domain name: >
LIST-STANDARD-DOMAINS
LIST-EXECUTION-QUEUE  < Interval: >
LIST-TIME-QUEUE  < Interval: >
DEFINE-SWAP-FILE  <File name: >
DELETE-SWAP-FILE  <File name: >
SET-ND-500-AVAILABLE
SET-ND-500-UNAVAILABLE  [<>]
STOP-ND-500
STOP-ND-500
LOGOUT-PROCESS  <Process number: >
ABORT-PROCESS  <Process number: >
LIST-ACTIVE-PROCESSES
LIST-DOMAIN  <Domain name: >
DOMAIN-STATUS  <Domain name: >
LIST-STATUS  <Domain name: >
SET-PROCESS-NAME  <Process name: >
LIST-PROCESS-TABLE-ENTRY  <Process number: >
LIST-ACTIVE-SEGMENTS  <Process number: >
PROCESS-STATUS
LIST-SEGMENT-TABLE-ENTRY  <Physical segment number: >
LIST-SWAP-FILE-INFO  <Swap file number: >
DEBUG-SWAPPER  <On/Off: >
DUMP-SWAPPER  <File name: >
DUMP-PHYSICAL-SEGMENT  <File name: >,<Ph segno: >
LIST-TABLE  <Table name: >,[<Index: >]
LOAD-SWAPPER  <File name: >
START-SWAPPER
GIVE-N500-PAGES  <Number of pages: >
TAKE-N500-PAGES  <Number of pages: >
SET-CPU-STATUS  <CPU Number: >,<Image: >,<Save: >,<Status: >
CHANGE-CPU  <CPU Number: >
SET-PHYSICAL-SEGMENT-ADDRESS  <Segment number: >,<Physical page number: >
CACHE-MODE  <Program cache mode: >,<Data cache mode: >
REMOVE-FROM-TIME-SLICE  <Process number: >
INSERT-IN-TIME-SLICE  <Process number: >,<Time slice class: >

4b. Normal boot sequence — verified: mostly AUTOMATIC, not manual

Corrects the earlier "likely next steps" guess in this section (LOAD-CONTROL-STORE -> LOAD-SWAPPER -> START-SWAPPER -> SET-ND-500-AVAILABLE as a manual sequence). That guess is WRONG per the manual text below — do not follow it.

Verified from ../../../Reference-Manuals/ND-60.136.04A ND-500 Loader Monitor.md:

  • STOP-ND-500 (section 8.10.3, manual line 5359-5361): "When a user attempts to start an ND-500 process after this command has been executed, the microcode will automatically be reloaded, the swapper process placed in memory and started ('warm start' of ND-500)." — i.e. control-store load + swapper load/start are automatic, triggered by the first process start, not operator-issued commands.
  • LOAD-SWAPPER (section 8.10.10.4, manual line 5759): "Normally, this is done automatically when the first ND-500 process is initiated by the monitor, but this command may be useful to load a new copy if there are reasons to believe that the existing one is corrupted, or to load a non-standard version of the swapper process." — confirms LOAD-SWAPPER/START-SWAPPER are manual overrides for the abnormal case, not normal-boot commands. Default source file if invoked manually: (SYSTEM)SWAPPER (:PSEG/:DSEG — file type may not be specified).
  • LOAD-CONTROL-STORE (section 8.10.6.3, manual line 5471-5477): default file CONTROL-STORE:DATA (the file you copy the MIC-5xxx-2-500:DATA microcode to per section 2 above), default start address 0, default word count = entire control store (20000B words). Same pattern: this is the command SINTRAN issues automatically; you'd only run it by hand to reload/replace microcode.
  • DEFINE-MEMORY-CONFIGURATION (section 8.10.4/8.10.4.1, manual line 5373-5385) — the one piece that genuinely IS a required manual step: "The ND-500 system has itself limited capability to investigate its own memory configuration. Therefore the memory configuration must be defined by the command DEFINE-MEMORY-CONFIGURATION." Parameter: <ND-100 page number for ND-500 physical address 0> — i.e. the offset between ND-500 and ND-100 physical addressing for the same shared-memory cell. Persists across a warm start; "local ND-500 memory is not legal in the ND-500 multiuser Monitor" (line 5375). This is the command most likely relevant to your emulator setup — if your bus emulation doesn't already answer the presence probe with the right memory offsets, the Monitor will not know where shared/multiport memory lives without this.
  • MASTER-CLEAR (section 8.10.13, manual line 5828-5832): hardware reset only — "Brings the ND-500 out of any hang-up state... equivalent to pressing the MCL button on the ND-500 front panel... used before a complete restart... contents of registers are unpredictable." Not a routine boot step.
  • RUN-SELFTEST: present in your live J04 HELP output but not documented anywhere in ND-60.136.04A — no manual entry found despite a full-text search. Note it as UNVERIFIED/undocumented; I don't know its parameters or effect.
  • SET-ND-500-AVAILABLE/SET-ND-500-UNAVAILABLE (section 8.10.1-8.10.2): control whether other users may log in — access control, not initialization.

Practical implication for your emulator: since you already got past CH5CPUPRESENT (section 4 above) and reached the interactive N500: prompt, the "normal boot stuff" for a real system — microcode load and swapper start — should already trigger automatically the moment you start (or place/recover) a domain. If that fails, the manual points at DEFINE-MEMORY-CONFIGURATION as the thing your emulator setup most likely still needs to supply correctly, since the ND-500 side cannot self-discover it.

4d. DEFINE-MEMORY-CONFIGURATION — what it does, and what value to use

What it does (verified, manual section 8.10.4/8.10.4.1, lines 5377-5409, and kernel source):

  1. You give one parameter: <ND-100 page number for ND-500 physical address 0> — the offset between ND-500 physical addressing and ND-100 page addressing for the same shared-memory cell.
  2. The Monitor then asks a set of subcommands per memory part (line 5395-5407): size in pages, whether ND-100 has access, whether ND-500 has access as Program, whether ND-500 has access as Data, and whether this is the last part. Default answer is "access for both CPUs, both P and D for ND-500."
  3. Kernel-side, this is N500M function 40B (DEFM) — confirmed in the manual's N500M function table (ND-60.136.04A, chapter 10, line 6470: "Define memory config. "). Per ../../../SINTRAN/OS/20-MPM-VS-LOCAL-MEMORY-DETECTION.md section 8.2.3 (NPL-source traced, 5P-P2-MON60.NPL:587), the value you give is written straight into the kernel variable ADRZERO — the ND-100 page number of the first page of MPM as seen from the ND-100 side. This overrides whatever boot-time auto-detection found.
  4. Persistence: "saved and will survive a normal restart (warm start)" but "lost" on a MACM )HENT / 22! cold start (line 5385, 5409) — the manual recommends saving it as a permanent macro so it can be reissued after a cold start.

Confirms your assumption: yes, this is specifically about MPM (multiport) memory. ADRZERO is exactly the ND-100 page number where the shared multiport-memory window begins. Per the OS/20 doc's NPL trace (PH-P2-OPPSTART.NPL:2498, section 8.2.4): SINTRAN also has an automatic fallback — at boot, if ND-500 is configured but ADRZERO is still -1 ("not configured"), it scans physical memory for the first page that answers as an MPM page type and sets ADRZERO to that. DEFINE-MEMORY-CONFIGURATION is the explicit, authoritative override of that guess — priority order (section 8.2.4): (1) boot-time auto-detection, then (2) MEMDEF/DEFINE-MEMORY-CONFIGURATION overrides it if given.

What value the manual suggests: none — there is no manual-stated default or example number. ND-60.136.04A only describes the semantics (an ND-100 page number, site- and hardware-dependent); it never states a recommended figure, because on real hardware this depends entirely on where your MPM5/multiport card is physically strapped into the ND-100 address space.

UNVERIFIED / this repo's own derived guess, NOT an official ND value: a separate repo analysis doc, ../../../SINTRAN/OS/20-MPM-VS-LOCAL-MEMORY-DETECTION.md section 10.5, self-labels a "recommended" ADRZERO of 2000B octal (1024 pages, i.e. ND-100 word address 20000000B) as "typical," with 1000B as a safe minimum (above kernel space) — but that section is explicitly flagged in the source doc as based on generic SINTRAN layout patterns, not a cited manual value, and section 9.2 of the same doc carries its own warning: "The following offsets are ASSUMPTIONS based on symbol analysis. They have NOT been verified against authoritative documentation." Treat this number as a plausible starting guess only, not a fact.

What you actually need for your emulator: since ADRZERO must equal wherever your bus/MPM emulation exposes the shared-memory window in ND-100 page-number terms, the correct value is a property of your bus-interface emulation, not a constant from the manual — there is no single right answer without knowing what page range your emulated 3022/5015-or-MPM device already claims/responds to. If your emulation already answers CH5CPUPRESENT at a specific IOX base/page, that's the reference point to derive this from, not the 2000B guess above.

CAUTION — fabrication found in a companion doc: while researching this, ../../../SINTRAN/ND500/ND500-INITIALIZATION-AND-EXECUTION-GUIDE.md section 9 was found to use commands LIST-DOMAINS, EXAMINE-DOMAIN, and LINK-LIBRARY in its example transcripts. None of these three appear anywhere in ND-60.136.04A (full-text searched), and none appear in the live J04 HELP output captured above. The real equivalents per the verified command list are LIST-DOMAIN <domain name> (singular), DOMAIN-STATUS <domain name>, and LIST-STATUS <domain name>; there is no LINK-LIBRARY NLL command — real segment-linking commands are LINK-SEGMENT, OPEN-SEGMENT/CLOSE-SEGMENT, LOAD-SEGMENT (all confirmed present in the manual). That guide's section 9 example transcripts should not be trusted as-is; this has not yet been reported/fixed in that file.

Note on the resident kernel-side monitor (distinct from the :PROG you ran): the release doc states "the ND-500/5000 System Monitor is installed as part of SINTRAN" (line 485) — this piece needs no separate install step at all. Verified segment location: segment 30, name S3SMS5, page range 40000:177777, description "ND-500 System Monitor segment" (release doc line 2523); segment 62 holds a save copy of it (line 2549); fixed-segment table: ND-500 Monitor | 1057B | 60B | 40000B (line 2372).

4c. RUN-SELFTEST -> FATAL "Illegal function code in MON 60" (2026-07-09)

Live test result: 

N500: run-selftest

 *** FATAL SYSTEM ERROR ***
ND-500(0) error:      Illegal function code in MON 60

Verified error identity. This is the named kernel error EILFUNC, value 2011B (octal), defined at ../../../SINTRAN/NPL-SOURCE/NPL/5P-P2-MON60.NPL:49: SYMBOL EILFUNC= 2011 % ILLEGAL FUNCTION CODE IN MON 60. The manual documents the same error verbatim: ../../../Reference-Manuals/ND-60.136.04A ND-500 Loader Monitor.md lines 7383-7385 — "Illegal Function Code in MON 60 — Check the list of valid function codes in chapter 10 against the location pointed to by the ND-100 A register. If this monitor call was not used by the application program, the error message indicates an error in the Monitor that should be reported to Norsk Data."

Root cause traced to the resident kernel's MON-60 dispatch table. The :PROG Monitor calls MON 60B with a function code in the A register (mechanism confirmed independently in ../../../SINTRAN/ND500/ND500-MON-RE-FINDINGS.md section 2). The resident side dispatches through the 5IFUNC array in 5P-P2-MON60.NPL:1319-1335 ("TABLE OF ENTRY POINTS TO GOTO BEFORE CALLING THE SYSTEM MONITOR / THE FUNCTION CODE IS INDEX IN THE TABLE"). Reading that array by octal index, this SINTRAN build (s3vs-4 source, matching the L07/M06 symbol-table generation in this repo) explicitly hard-wires these function codes to ILLFUNC (i.e. deliberately rejects them, regardless of what the calling :PROG sends): 146B, 147B, 151B, 152B, 153B, 174B, 175B, 176B.

Not yet determined (needs disassembly, not guessed): which of these codes — or another one — the Monitor's RUN-SELFTEST command actually sends. ND500-MON-RE- FINDINGS.md section 2.3-2.4 already decoded the :PROG's MON-60 stub-dispatch array (ram:ccc8-ce6d in ND-500-MON-J04:PROG) but only named codes up to roughly 106B/143B — RUN-SELFTEST's specific code was not among those named. Finding it requires tracing the RUN-SELFTEST command-string handler in the same binary to its SAA <code> stub, the same technique already used for the other commands in that findings doc.

A plausible but UNVERIFIED lead: the driver source has a block of function-code constants documented only in comments (i.e. not compiled as active SYMBOLs in this NPL source revision — each line prefixed %SYMBOL, 5P-P2-MON60.NPL:270-279), including STSELFTST=155 "START SAMSON SELFTEST". That code name strongly suggests a selftest function exists but is SAMSON/ND-5000-specific — and per CH5CPUPRESENT(../../../SINTRAN/ND500/ND500-BUS-INTERFACE-REFERENCE.md section 8.1), your emulated CPU is flagged as type OLD500 or SAMSON depending on which probe path answered (IOX RSTA5 = OLD500; Octobus response = SAMSON). If your bus emulation answers as OLD500, a Samson-only selftest function could plausibly be rejected — but this is not confirmed: the dispatch-table index for 155B is 5NOPAR, a shared pre-dispatch entry routine (5P-P2-MON60.NPL:1889, "COMMON POINT TO GO THROUGH BEFORE CALLING THE SYSTEM MONITOR"), not ILLFUNC — so 155B is not itself in the confirmed illegal set, weakening this lead. Do not treat the STSELFTST connection as established; it is a hypothesis to test via disassembly, not a diagnosis.

Practical takeaway: RUN-SELFTEST is not part of the normal boot/bring-up path (see section 4b) — it is an optional hardware diagnostic. Given it hits a kernel-level ILLFUNC rejection rather than a bus/detection failure, this looks like a genuine function-not-implemented-in-this-kernel-build (or CPU-type-mismatch) condition rather than an emulator bus-interface gap. Recommend avoiding RUN-SELFTEST for now and proceeding with normal domain bring-up instead; treat this as a separate open item, not a blocker for the CPU-detection/monitor-startup work in sections 4-4b.

  • UNVERIFIED: no extracted RT-LOADER/HENT-MODE command for loading the ND-500 Swapper (SWAPPER-K:PSEG/:DSEG) was found in any source read. It is a :PSEG/:DSEG pair, not a :BPUN, so it likely does not load via the same READ-BINARY idiom — do not assume the syntax.

4a. The ND-500 Place Library — six :BRF variants, install step UNVERIFIED

Live floppy listing of 211305B02-XX-01D (2026-07-09) shows the package contains more than the Monitor and Swapper — full inventory and analysis now recorded in ../../../SINTRAN/ND500/ND500-L-PACKAGE-CONTENTS.md. Six Place Library files, all type :BRF, all under user FLOPPY-USER: PLACE-1BANK-C01, PLACE-2BANK-C01, PLACE-BIG-1B-C01, PLACE-BIG-2B-C01, PLACE-SML-1B-C01, PLACE-SML-2B-C01.

  • Verified: PLACE-BIG-2B-C01.BRF is ND-100 object code (confirmed by Ghidra disassembly, language ND-100:BE:16 — see ../../../SINTRAN/ND500/ND500-PLACE-LIBRARY-C9-FINDINGS.md), implementing the Monitor's PLACE-DOMAIN/DEBUG-PLACE functionality and MMS/hardware-fault decoding (strings: "Domain information table", "Contents of Physical segment Table", "SAMSON HARDWARE FAULT", "500 HARDWARE FAULT").
  • UNVERIFIED / no install command found: the release doc's copy commands (section 1 above) only cover the Monitor and Swapper — it gives no @COPY-FILE or NRL/link command for the Place Library files. Working hypothesis (NOT confirmed by any source read, do not act on it as fact): these :BRF files may already be linked into ND-500-MON-J04:PROG at Norsk Data's build time (analogous to how PASCAL-LIB/ PASCAL-2LIB get NRL-linked into PASCAL:PROG per ../../Software/INSTALL-METHODOLOGY.md section 5), which would explain why the release doc never mentions copying them separately. This is plausible but unverified.

5. HENT-MODE / LOAD-MODE changes required for version L (general, not ND-500-specific)

Source: same release doc, section 1.6 (lines 499-519).

  • HENT-MODE (cold start): remove any commands/mode files that load XMSG; replace Cosmos Basic Module loading with the version-F loader; remove ERS/SINTRAN III Watchdog load commands (if previously installed); remove @INITIALIZE-ERROR-LOG (K and earlier only).
  • LOAD-MODE (warm start): remove ERS/Watchdog start commands; if using DOMINO devices, add @START-SERVERS to start all system-included servers before starting the ND-500/5000 part of the system; ensure the XMSG-START file references the correct (M-version) XMSG-Command program.

TODO (validate via doc / command / manual exercise)

  • Transcribe START-SINTRAN-MULTIUSER.TXT (not yet read).
  • L bootstrap transcript and floppy file-system IDs.
  • Confirm the ND-500 Swapper's actual load command (PSEG/DSEG pair) — not found in sources read.
  • RESOLVED (2026-07-09): the Monitor :PROG needs no HENT-MODE/RT-LOADER load step — it is started directly with @ND-500. See section 4.
  • Diagnose "No ND-500(0) CPU found" — 3022/5015 interface or Octobus/Samson probe not answering CH5CPUPRESENT at boot (hardware/emulator config issue, not install issue). See section 4.
  • Find the actual install/link command for the six Place Library :BRF files — not in the release doc's copy-command list. See section 4a.