Skip to content

MON 313B (octal) - InBufferState (IBRSIZ)

Returns terminal input-buffer state: the number of characters currently held in the input ring, and the number of characters before the next break (BRKMAX-limited). Terminals walk the input ring; TAD and internal-buffer devices take other paths. Files are not valid here.

Status: GOTAB[313] level-14 dispatch is byte-proven and live-confirmed; the IBRSIZ worker body is real SINTRAN L bytes; the exact MON 313 -> IBRSIZ link crosses an uncarved runtime CALLPROC/second-level dispatch (see Honest caveats). All addresses/values are octal.

Dispatch path

flowchart LR
    A["User program<br/>MON 313B"] --> B["ENT14 level-14<br/>T = MON number"]
    B --> C["GOTAB[313B] = 112164B<br/>(byte + live proven)"]
    C --> D["112164B entry stub<br/>025-S3IRPIT<br/>JMP I 10 -> 112247B"]
    D -.runtime CALLPROC.-> E["IBRSIZ worker body<br/>025-S3IRPIT :110543B"]
    E --> F["walk input ring<br/>held / before-break via LBYT"]
    class A blue
    class B,C,D teal
    class E,F green
    classDef blue fill:#E3F2FD,stroke:#0D47A1,color:#0D47A1
    classDef teal fill:#E0F7FA,stroke:#00838F,color:#00838F
    classDef green fill:#E8F5E9,stroke:#2E7D32,color:#2E7D32

The dashed hop (D ⇢ E) is the resident second-level / CALLPROC-style computed jump - the stub at 112164B byte-forwards to 112247B (a monitor/interrupt-entry routine), and the final worker is register-loaded at runtime. That MON# -> worker link is not present as a static pointer in any carved segment, so it cannot be followed from bytes.

Code location (dispatch path)

Every row is a real region you can open. Byte offset = (addr - loadbase) in octal words × 2. Load bases: SINTRAN-DATA_commoncode = 0, 025-S3IRPIT = 32000B.

Role Segment (full disasm) Addr range (octal) Byte offset Symbol Verdict
GOTAB[313] dispatch word commoncode.asm · .hex 071546B (= 071233B+313, 1 word) 59084 GOTAB+313 = 112164B VERIFIED
Entry stub (JMP I 10 -> ptr 112174B = 112247B) 025-S3IRPIT.asm · .hex 112164B..112174B 49384 SYMBOL-2-LIST mislabels as DIA3 VERIFIED
Reached routine (monitor/interrupt entry) 025-S3IRPIT.asm · .hex 112247B..112264B+ 49486 unnamed (list misaligned here) VERIFIED bytes
runtime CALLPROC / 2nd-level dispatch — (register-loaded target) — — CALLPROC UNVERIFIED
IBRSIZ worker body (carved window, 149 words) 025-S3IRPIT.asm · .hex 110543B..110770B 47814 IBRSIZ / T2P06 (L07) real bytes; link MISATTRIBUTED

Verify by hand (GOTAB entry): grep -E '^112164 ' ../../segments-ref/025-S3IRPIT/025-S3IRPIT.hex -> byte offset 49384; then dd if=../../../segments/025-S3IRPIT.bin bs=1 skip=49384 count=8 2>/dev/null | od -An -tx1 -> aa 08 cc 4d ba 07 ab 17 (= octal 125010 146115 135007 ... = the JMP I 10 stub). The GOTAB word itself: dd if=../../../resident/SINTRAN-DATA_commoncode.bin bs=1 skip=59084 count=2 2>/dev/null | od -An -tx1 -> 94 74 (big-endian 0x9474 = 112164B).

Verify by hand (worker body): grep -E '^110543 ' ../../segments-ref/025-S3IRPIT/025-S3IRPIT.hex -> byte offset 47814; then dd if=../../../segments/025-S3IRPIT.bin bs=1 skip=47814 count=8 2>/dev/null | od -An -tx1 -> ba 52 ba 52 aa 52 c4 05 (= octal 135122 135122 125122 ... = the IBRSIZ: CALL GET0 / CALL GZTREG prologue).

Instruction walkthrough

Full listing: 313B-InBufferState.ASM. Region A is the level-14 entry; Region B is the functional IBRSIZ body.

GOTAB entry (112164B) - the level-14 dispatch word GOTAB[313] = 112164B (live-confirmed against the running system). The word at 112164B is 125010 = JMP I 10, an indirect jump whose pointer word is at 112164B + 10B = 112174B. That pointer reads 112247B, so the stub forwards to 112247B (a different routine - a monitor/interrupt-entry sequence, SAA 2 / ORA 12 / critical-section setup), which itself continues via 112254B JMP I 10 -> pointer 112264B = 112632B and onward through register-indexed computed jumps.

IBRSIZ prologue (110543B..110547B) - 110543 JPL I 122 dereferences pointer word 110665B = 000374B (GET0, the NPL IBRSIZ: CALL GET0 prologue), immediately followed by 110544 JPL I 122 -> 110666B = 044276B (GZTREG, fetch the logical device number). A SKP IF DA UEQ 0 guard (110546) bails for non-eligible devices.

Ring scan (110555B..110662B) - an IOF critical section (110555) walks the input ring uninterrupted. The inner loop (110606..110630) uses LBYT (110614) to fetch each input byte and AAX 1 to advance, counting characters held and characters-before-break, honoring BRKMAX via the compare/skip chain around 110573..110576 and 110645..110662.

Result (110663B onward) - 110663 LDA ,B 16 loads the held-count; the before-break count is staged through 110723..110762, and 110762 JMP I ,B 34 returns to the caller.

Parameter / register contract

Reg / field Dir Meaning Verdict
ZTREG (T) in logical device number, fetched via GZTREG (110544) VERIFIED (bytes: CALL GZTREG)
ZAREG (A) out BHOLD = characters held in input buffer (110663 LDA ,B 16) inferred (NPL IBRSIZ)
ZXREG (X) out BCOUNT = characters before break (BRKMAX-limited) inferred (NPL IBRSIZ)
error path out ZAREG = 240B on error inferred (NPL, not proven in these bytes)
IOF/ION internal ring walked in a critical section (110555) VERIFIED (bytes)

The exact user-visible A/X/T assignment lives in the caller-side MON 313 wrapper and the uncarved second-level dispatch frame, so the precise register convention is inferred, not byte-proven here.

Pseudo-code (for an emulator)

See 313B-InBufferState.pseudo.c - a pseudo-C model of the handler for emulator authors. The worker-body control flow (GET0/GZTREG prologue, the IOF ring scan, held / before-break accounting) is byte-verified; the semantic labels (which register holds which count, the exact break arithmetic) are inferred from the NPL IBRSIZ structure.

Every instruction in the .pseudo.c is translated against the canonical ND100-INSTRUCTION-SEMANTICS.md (IOF opens the ring-scan critical section but NO matching ION is in the carved window - the re-enable is out of carve; LBYT fetches one input byte; SKP/BSKP skip polarity, do not invert).

Honest caveats

What is byte-proven: GOTAB[313] = 112164B (level-14 dispatch, matches a live read of the running system); the 112164B stub is real code (JMP I 10) whose pointer word 112174B reads 112247B; the routine at 112247B is real bytes (a monitor/interrupt-entry sequence); and the IBRSIZ worker body at 110543B..110770B is real SINTRAN L bytes, structurally IBRSIZ (CALL GET0 / CALL GZTREG prologue, IOF ring scan, LBYT byte walk).

What is NOT proven: the link from the GOTAB dispatch to the 110543B IBRSIZ body. On the static evidence they do not connect: - GOTAB[313] resolves (via the 112164B stub) to 112247B, not 110543B. - Scanning all GOTAB entries 0..377B: none point to 110543B, and no single JMP I stub forwards there. - The value 110543 appears as a pointer word nowhere in 025-S3IRPIT.bin, so no JMP I/JPL I in the segment can dereference to it. - 110543B is ~1400B words below 112247B, far outside the ±127-word reach of a direct jump.

So the MON 313 -> IBRSIZ attribution rests on the symbol name + the matching InBufferState behaviour (GET0/GZTREG/ring-scan), reachable only via a runtime computed jump (CALLPROC second-level dispatch) - hence MISATTRIBUTED in the strict sense. The earlier ANALYSIS claim of a direct GOTAB[313] -> 110543B path and the DISPATCH byte-trace disagree; the byte-trace WINS - the true static target is 112247B, and 110543B is the structurally-correct worker whose runtime reachability is unproven. Confirming it needs a live trace: break at 112164B on a real MON 313, single-step 112164 -> 112247 (via 112174) and through the 112247/112632 computed jumps, and capture whether the final indexed JMP/JPL lands on 110543B or on a different worker.

Address note: the NPL source is a different revision (NPL IBRSIZ=110642B vs L07 110543B). Match behaviour/structure, not raw bytes.

Method: ../../../../../EXTRACTING-RESIDENT-CODE.md · dispatch reality: ../../TASK-05-mismatches.md · master map: ../../MON-CALL-INDEX.md.