Skip to content

Bugs report from Willi: INTERRUP.C #111

Open
Open
Bugs report from Willi: INTERRUP.C#111
@shidel

Description

@shidel
shidel
opened on May 17, 2026

LINES 1 - 1410:

Exhaustive Master Audit List – INTERRUP.C (Part 01)

[TECH]

Table 00415 (DESQview Register Save Record): Offset 0Ah is a WORD (SI). The next listed offset is 0Eh (DS).

    Logic Error: A WORD is 2 bytes. 0Ah + 02h = 0Ch. The offset 0Ch (which would normally be BP) is missing from the table.

Table 00416 (DESQview mapping context record): Offset 11h (BYTE flags).

    Error: Bit 0: Swapped out??? and Bit 2: Swapped out???. This is a duplicate description for two different bits.

Table 00419 (DESQview process record v2.52-v2.60): Offset 161h (WORD segment of ???). The next listed offset is 162h.

    Logic Error: If 161h is a WORD, it occupies 161h and 162h. The next offset must be 163h.

Table 00427 (.PIF/.DVP file): Offset 19Ah (WORD). Next offset is 19Ch. This is correct (2 bytes), but check typo in description (see below).

[TYPO]
5. Table 00427 (.PIF/.DVP file): Offset 19Ah description: length of data follownig XDVP signature -> Misspelling. Should be "following".
6. Table 00440 (WINDOW STREAM opcodes):

  • Opcode BAh: [contents of window unpredict after] -> Should be "unpredictable".
  • Opcode BBh: [contents of window unpredictbl after] -> Should be "unpredictable".
  1. Table 00423 (DESQview Common Memory Header): Offset 02h: bytes of commom memory -> Misspelling. Should be "common".
  2. Table 00440: Opcode E5h (panel file): create kyboard object -> Misspelling. Should be "keyboard".
  3. Table 00440: Opcode EEh: color correspondnces (Note: re-check from similar context in Part 9/10, ensure consistent "correspondences").

[HEX]
10. Table 00414 (DESQview object): Offset 04h: signature FEDCh. Standardize. Protocol 4.2 requires lowercase 'h'. Should be fedch.
11. Table 00440: Opcode E6h Note: physical size of 0x0. Standardize. Should be 00h x 00h or 0h x 0h.

[FORMAT]
12. Section INT 15 - TopView - "SOUND" (AX=1019h): Return description: tone continues to completion. Spacing/Alignment check: ensure the return block is clearly separated from notes.
13. Section INT 15 - TopView - "POSWIN" (AX=1023h): Bitfields for Table 00426: 0,1 horizontal position, 2,3 vertical position. For clarity, use binary notation: 00b = current, etc.
14. Table 00444 (TopView field table entry): Offset 00h-03h: The use of backslashes and slashes (, /) to group rows is a legacy RBIL visual style. Ensure UTF-8 rendering does not break the alignment.

LINES 1411 - 2780:

Exhaustive Master Audit List – INTERRUP.C (Part 02)

[TYPO]

Table 00446 (QUERY STREAM opcodes): Opcode A6h: height of physcial window -> Misspelling. Should be "physical".

Table 00449 (Async notification routine): Opcode 45h: object hande of window -> Misspelling. Should be "handle".

Section INT 15 - TopView - "LEN" (AH=12h/BH=09h): Description for 0Ch: owning handle on top of stk -> Abbreviation. Should be "stack".

Section INT 15 - TopView - "OPEN" (AH=12h/BH=0Ch): Description for 0Ch: handle on top of stck -> Misspelling. Should be "stack".

Section INT 15 - TopView - "EOF" (AH=12h/BH=10h): Description for 0Ch: check log crsr of window -> Abbreviation. Should be "cursor".

Section INT 15 - TopView - "SETSCALE" (AH=12h/BX=1100h): STACK description: number of colums -> Misspelling. Should be "columns". (This error also appears in the description for BX=1200h).

Section INT 15 - VMiX - "sys_qetqnext" (AH=0Ah): Method name: "sys_qetqnext" -> Misspelling. Consistent with other "get" functions in VMiX, it should be "sys_getqnext".

Section INT 15 - TopView - "SIZEOF" (AH=12h/BH=08h): Return description: handle on stack has been popped -> Redundant phrase. "Popped" is enough, but "has been popped" is acceptable.

[TECH]
9. Table 00446 (QUERY STREAM): Opcodes D0h-DFh: overwritten with D0h if... -> These descriptions imply that the query itself modifies the opcode byte in the stream, which is a critical technical detail for programmers. Ensure the wording "overwritten" is used consistently.
10. Table 00456 (MultiDOS Plus TCB): Offset D2h: 414 BYTEs. Standardize. Following the hex standard of the project, this should be 019Eh BYTEs.
11. Section INT 15 - VMiX - "sys_sysreq" (AH=0Bh): Note: reportedly disables the keyboard until the delay completes. This is a critical side-effect warning. Ensure it is preserved.
12. Section INT 15 - VMiX - "sys_clrwindow" (AH=14h): Note: clears window to color set with "sys_setcolors" (AH=18h). Correct cross-reference.

[HEX]
13. Table 00440 (WINDOW STREAM) (referenced in SeeAlso): Protocol 4.2 check. Ensure any mention of hex values like 0x0 in descriptions are changed to 00h or 0h.
14. Section INT 15 - MultiDOS Plus - CHANGE TIME SLICE (AH=0Ah): Values 00h, 80h, 40h, etc. All follow the lowercase 'h' suffix correctly.

[FORMAT]
15. Table 00447 (MANAGER STREAM): Opcode 60h-71h: opposite of 40h-51h. The extra space before opposite should be removed for consistent column alignment.
16. Table 00449 (Async notification routine): Sub-entry for 4Dh (Scissors/cUt): field number ul corner is in. For technical clarity, "UL" or "upper-left" is preferred.

[QUARANTINE]
17. Table 00446: Opcodes EC/ED/EEh: bit 7 unused???. The question marks indicate unconfirmed data from the original RBIL. Flagged for verification in Release 62.

LINES 2781 - 4248

Exhaustive Master Audit List – INTERRUP.C (Part 03)

[TECH]

Section INT 15 - SYSTEM - later PS/2s - EXPANSION UNIT... (AX=5101h): Return CF: CF set if successful ... CF clear on error.

    Technical Analysis: As the note correctly states, this is the reverse of the BIOS standard (where CF=1 usually indicates an error). This is a critical technical anomaly. The note Perhaps a typo in the IBM BIOS Tech Ref? should remain, but the return logic must be strictly verified against actual hardware behavior.

Section INT 15 - VMiX v2.???+ - "sys_sysreq" (AH=0Bh): Subfunction "open": 011 = 115200.

    Error: Missing binary suffix. Should be 011b = 115200.

Section INT 15 - Olivetti Quaderno - GET ??? (AX=44C3h): Table 00465, Bit 1-0: "tres complique".

    Error: This is French for "very complicated". According to the language rule (End-Korrektor only speaks English), this must be translated to "very complicated".

Table 00462 (A20 gate support status): Bit 15: location not yet defined. This placeholder should be flagged for Release 62 if the location is now known.

[TYPO]
5. Section INT 15 - Advanced Power Management v1.0+ - CPU IDLE (AX=5305h): Note 2: avoid reentrance problems.

  • Correction: In technical documentation, the term "reentrancy" is standard.
  1. Section INT 15 - HP 95LX/100LX/200LX - INSTALLATION CHECK (AX=4DD4h): Return DH/DL: german.
  • Correction: Should be capitalized: "German".
  1. Section INT 15 - APM SL Enhanced v1.0 - GET CALENDAR EVENT TIME (AX=5380h/BH=08h): Desc: CMOS ram.
  • Correction: Acronym should be uppercase: "RAM".
  1. Section INT 15 - VMiX v2.???+ - "sys_vm_func" (AH=51h): Note 1: privilege level 0. Standardize to "Privilege Level 0" or "Ring 0".

[HEX]
9. Section INT 15 - SYSTEM - POWER-ON SELF-TEST ERROR LOG (AH=21h): Note 4: actual device number minus 255.

  • Correction: For consistency with the rest of the document, 255 should be shown as 255 (FFh).
  1. Section INT 15 - IBM BIOS - SMART ENERGY SYSTEM - ??? (AX=2303h): DI = ??? (0352h). Correct lowercase suffix.
  2. Section INT 15 - SYSTEM - later PS/2s - GET A20 GATE STATUS (AX=2402h): Return CX: within C000h read attempts. Correct.

[FORMAT]
12. Section INT 15 - VMiX v2.???+ - "sys_scrollwin" (AH=1Fh): Note: high byte = row, low = col.

  • Correction: "col" is an informal abbreviation. Should be "column".
  1. Section INT 15 - Advanced Power Management v1.1+ - DRIVER VERSION (AX=530Eh): AL description: (02h for APM v1.2). Ensure consistent spacing: 02h is correct.

[QUARANTINE]
14. Section INT 15 - VMiX v2.???+: Multiple headers (AH=1Fh, AH=50h, AH=51h, AH=52h) contain v2.???+. This version numbering is incomplete and requires research for the final Release 62.

LINES 4249 - 5615

Exhaustive Master Audit List – INTERRUP.C (Part 04)

[TYPO]

Label #155380BH09 & #155380BH0A: Desc: CMOS ram and Extended CMOS ram.

    Correction: Acronym should be uppercase: "RAM".

Table 00488 (Smart Battery command codes): Cmd 07h: another 10sec? -> Missing space. Should be "10 sec".

Label #1561 (HP 100LX/200LX): Notes: The announciators are the indicator symbols.

    Correction: Misspelling. Should be "annunciators".

Label #1565 (HUNTER 16): Header: AUTHORISE CHARGING.

    Correction: British English. Standardize to US English: "AUTHORIZE".

Label #1583 (HUNTER 16): Header: CONTROL SCREEN SYNCHRONISATION.

    Correction: British English. Should be "SYNCHRONIZATION".

Table 00192 & Table 00199 (HUNTER 16): co-ordinate.

    Correction: Should be "coordinate" (no hyphen, consistent with the rest of RBIL).

Table 00497 (HUNTER sound pitch): MID C.

    Correction: Usually written as "Middle C" or "Mid-C".

Section INT 15 - BIOS - JOYSTICK SUPPORT (AH=84h): 250kOhm.

    Correction: Missing space. Should be "250 kOhm".

[TECH]
9. Table 00181 (C&T display mode): Bit(s) 1-0: 00 = 200, 01 = 350, 10 = 400.

  • Error: Missing binary suffix. Should be 00b, 01b, 10b.
  1. Table 04098 (MULTIJOY device classification): Bit(s) 3-0: 0000 keyboard..., 0001 gamepad....
  • Error: Missing binary suffix. Should be 0000b, 0001b, etc.
  1. Label #1583 (BIOS - SET EVENT WAIT INTERVAL): Note 2: IBM AT 1984/1/10 BIOS ignores AL. This is a critical technical note for historical accuracy. Ensure consistency with similar notes in INTERRUP.B Part 11.
  2. Section INT 15 - VMiX v2+ - "sys_sint" (AH=1Ah): Note: AX = returned flags. Clarification needed if this refers to the full EFLAGS or just the low 16 bits.

[HEX]
13. Label #157002 / #157003 (Tandy 1000): mapped at 0E0000h and mapped at 0F0000h.

  • Verification: Lowercase 'h' suffix is correct.
  1. Label #1586 (HUNTER 16): BX = 1234h. Correct.
  2. Table 00486 (SMBus addresses): 80h-86h, 90h-96h. Correct.
  3. Table 00499 (GDT): Offset 12h, 1Ah: 24-bit linear ... address. Correct.

[FORMAT]
17. Section INT 15 - VMiX v2+ - "sys_bitblt" (AH=1Ch): STACK description: 00h = OR, 01h = AND. Correct.
18. Table 00135 (EDID Text Identification Strings): Offset 04h-09h vs. 04h-14h.

  • Format Error: The overlap in offsets for different string types needs clearer visual separation (e.g., using "---" lines as seen in other tables).

[QUARANTINE]
19. Label #157002 / #157003: Tandy 1000 Model ???. The model identifier is still missing in this version.
20. Table 00184 (C&T status): Bit(s) 15-0: reserved. (Checked again: still illogical for a WORD status if bits 6-1 are used later. Flagged for correction to 15-7).

LINES 5617 - 6966:

Exhaustive Master Audit List – INTERRUP.C (Part 05)

[TYPO]

Table 00504 (Application segment descriptor type): Bit 3 description: 0 date -> Technical Typo. Should be "0 data".

Section INT 15 - HUNTER 16 - GET/SET BOOST CHARGE (AH=8Dh): Desc: 200ma rather than 70ma.

    Correction: Metric units should use standard capitalization: "200 mA" and "70 mA".

Section INT 15 - HP OmniShare - STANDBY.COM - GET STANDBY DURATION (AX=BA22h): Return BL: standy duration.

    Correction: Misspelling. Should be "standby duration".

Section INT 15 - DESQview/X ... DVDOS4GX.DVR (AX=BFDEh): Offset 19Ah (referenced from Part 4 logic): follownig XDVP signature.

    Correction: Misspelling. Should be "following".

Section INT 15 - SYSTEM - GET CONFIGURATION (AH=C0h): Note 7: contains two occurences of code.

    Correction: Misspelling. Should be "occurrences".

Section INT 15 - SYSTEM - GET CONFIGURATION (AH=C0h): Note 7: signature for an applyable patch.

    Correction: Misspelling. Should be "applicable".

Table 00515 (Model/submodel/revision):

    Entry FCh/04h/01h: 10 Mhz 286.

    Entry FCh/04h/02h: 10 Mhz 286.

    Entry FCh/23h/02h: 20Mhz386SX.

    Correction: Standardize to "MHz" (uppercase M and H, lowercase z).

Section INT 15 - SYSTEM - POINTING DEVICE BIOS INTERFACE (PS) - RESET (AX=C201h): Note 2: erratically on some systems after the boot. "the boot" is slightly awkward; "after booting" or "after a reboot" is preferred.

[TECH]
9. Table 00521 (Toshiba product ID): Entry FCh/34h: 02/04494.

  • Error: The year likely has a typo. Should probably be 02/04/94.
  1. Table 00176 (Chips&Technologies chip type): Note: F65535/F65545 ???. These models are unconfirmed. (Marked as Quarantine).
  2. Table 00509 (ROM configuration table): Offset 2Dh: offset to relocatable portion. Offset 2Eh: offset to unused section.
  • Error: As noted in Part 9 of Interrupt.A, if Offset 2Dh is a DWORD, it occupies 4 bytes. Offset 2Eh causes a 3-byte overlap. This structure logic is flawed.
  1. Section INT 15 - VHRBIOS.SYS - GET VENDOR ID (AH=F5h): Return: AXBX = vendor ID.
  • Correction: Standard register pair notation is AX:BX.

[HEX]
13. Table 00176 (Chips&Technologies chip type): Bit(s) 7-4: 0000:, 0001:, etc.

  • Correction: Standardize with binary suffix: 0000b:, 0001b:.
  1. Table 00515 (Model/submodel/revision): BIOS dates like 04/24/81. This is the correct US date format for RBIL. No correction needed.

[FORMAT]
15. Table 00460 (PS/1 ROM startup DAC register table): Offset 00h: WORD number of DAC registers. Standardize description capitalization if necessary (usually lowercase "number").
16. Table 00521 (Toshiba product ID): The "date" column contains markers like ../..*.., ../..+.., ../..i... These represent 8-bit characters above 80h. The note correctly explains this, but the alignment is very messy in UTF-8.

[QUARANTINE]
17. Section INT 15 - VMiX v2.???+: Still contains the v2.???+ placeholder in headers (AH=1Fh).
18. Table 00515: Entry F8h/0Bh/02h: PS/2 Model P70 ??. The double question marks indicate missing model confirmation.

LINES 6967 - 8382:

Exhaustive Master Audit List – INTERRUP.C (Part 06)

[TECH]

Table 00535 (Device Descriptor Table): Offset OCh and OEh.

    Error: The letter "O" is used instead of the number "0" for the hex offsets. Should be 0Ch and 0Eh.

Table 00184 (C&T F65520/525/530 status): Bit(s) 15-0: reserved.

    Error: As noted in previous parts, this covers the entire word. Since bits 6-1 are defined below, this range should be corrected to 15-7.

Table 00542 (AMI CMOS bitfield identifier): Entry 0Ah: power management enabled???.

    Quarantine: The triple question marks indicate unconfirmed technical data.

Section INT 15 - APM SL Enhanced v1.0 - GET CPU SPEED MODE (AX=5380h/BH=0Ch): Return CL: (1,2,4 or 8).

    Consistency: Should have a space after the comma: "1, 2, 4, or 8".

Table 00508 (DOS/16M XBRK structure): Offset 04h: last available byte + 1 ???.

    Quarantine: Technical ambiguity regarding the end-of-memory pointer.

[TYPO]
6. Label #155380BH09: Desc: CMOS ram.

  • Correction: Should be uppercase: "RAM". (This occurs multiple times in the block, e.g., labels #155380BH0A, #155380BH0B).
  1. Label #155F33: Desc: curent video mode.
  • Correction: Misspelling. Should be "current".
  1. Label #1561 (HP 100LX/200LX): Note 1: announciators.
  • Correction: Misspelling. Should be "annunciators".
  1. Table 00544 (AMI BIOS parameter block): Offset 28h: cooresponds.
  • Correction: Misspelling. Should be "corresponds".
  1. Label #15D102: Header: RETURN ... BY I/O ADDRSS.
  • Correction: Misspelling. Should be "ADDRESS".
  1. Section INT 15 - TelAPI - "telstatus" (AH=EAh): Table 00401, Offset 1Dh: negotiation.
  • Correction: Misspelling. Should be "negotiation".
  1. Section INT 15 - TopView - "GETSCALE" (AX=1200h): Return description: colums.
  • Correction: Misspelling. Should be "columns".

[HEX]
13. Label #158A (HUNTER 16): SI = 1234h, DI = 5678h. Correct.
14. Label #15B001CX5354: CX = 5354h, DX = 4143h. Correct.
15. Section INT 15 - DESQview - QEXT.SYS (AX=11DEh): Correct lowercase suffix.

[FORMAT]
16. Table 00515 (Model/submodel/revision): Entry FCh/34h: 02/04494.

  • Format Error: Likely a typo for a date. Should be 02/04/94.
  1. Table 00449 (Async notification routine): Opcode 45h: object hande.
  • Correction: Misspelling. Should be "handle".
  1. Section INT 15 - VMiX v2+ - "sys_sleep" (AH=12h): SeeAlso: AH=03h"MultiDOS".
  • Correction: Missing a comma or space before the label quote in some RBIL formats, though this one is mostly consistent.

[QUARANTINE]
19. Table 00492 (Smart Battery device chemistry): Zinc-Air for ZnAr. Technical check: ZnAr is often shorthand for Zinc-Air, but verify if ZnAir is the proper VESA/Smart Battery acronym.
20. Section INT 15 - VMiX v2.???+: Multiple instances of the version placeholder v2.???+.

LINES 8383 - 9651:

Exhaustive Master Audit List – INTERRUP.C (Part 07)

[TECH]

Table 00535 (Device Descriptor Table): Offset OCh and OEh.

    Error: The character 'O' (capital letter O) was used instead of '0' (number zero) for the offsets.

    Correction: Should be 0Ch and 0Eh.

Section INT 15 - AMD64 - DECLARE OPERATING MODE (AX=EC00h): BL=03h description: All Other values are reserved.

    Correction: "Other" should be lowercase to match the style of the previous lines: "all other values".

Section INT 15 - Tandon memory mapper (AX=F200h): Return description: initialised.

    Correction: British English. Standardize to US English: "initialized".

Table 00558 (DESQview/X socket record): Function 0019h return: servent strctures???.

    Analysis: This is likely a reference to Unix servent structures. The misspelling "strctures" must be corrected to "structures".

[TYPO]
5. Label #151D (MultiDOS Plus): Note 1: equivalent to AH=1Ah followed by AH=10h.

  • Context Check: Ensure this logic is correct, as AH=1Ah is the name-to-number mapping. (Logic seems correct).
  1. Label #151F (VMiX): Header: VMiX v2.???+.
  • MISSING: Version number is incomplete.
  1. Label #15D102: Header: RETURN ... BY I/O ADDRSS.
  • Correction: Misspelling. Should be "ADDRESS".
  1. Table 00488 (Smart Battery command codes): Cmd 06h: entirely discharge (min.). Cmd 05h: completely charge (in min.).
  • Consistency: Standardize the unit notation (either "min" or "minutes").
  1. Table 00544 (AMI BIOS Flash ROM parameter block): Offset 28h: cooresponds.
  • Correction: Misspelling. Should be "corresponds".
  1. Table 00529 (Stepping level): Code 09h under 80386SX: 386SXstatic.
  • Correction: Missing space. Should be "386SX static".
  1. Section INT 15 - later PS/2s - ALLOCATE DMA ARBITRATION LEVEL (AH=CEh): Return AL description: arbiration level.
  • Correction: Misspelling. Should be "arbitration".

[HEX]
12. Section INT 15 - PhysTechSoft PTS ROM-DOS (AX=E901h): codes 000h-1FFh.

  • Correction: According to hex standard, the first value should be 0000h or 0h, and 1FFh is correct.
  1. Table 00484 (SMBus error codes): Error code 0Ah, 86h. Correct.
  2. Table 00542 (AMI CMOS bitfield identifier): Address 1Ah, 1Dh, etc. Correct.

[FORMAT]
15. Table 04091, 04092, 04093 (PTS ROM-DOS flags/codes): Bit(s) / Values column: ???.

  • MISSING: These tables are placeholders and lack actual bit numbers or code values.
  1. Table 00521 (Toshiba product ID): Entry FCh/34h: 02/04494.
  • Correction: Date format error. Should be 02/04/94.

[QUARANTINE]
17. Section INT 15 - OS HOOK - SysReq KEY ACTIVITY (AH=85h): Note 3: SysReq is often labeled SysRq. Confirm if this redundant note (already in INT 09) is required in the Release 62 master.
18. Table 00514 (Feature byte 5): IBM "private" (multiple entries). These are undocumented IBM bits.

Status: Audit für INTERRUP.C vollständig abgeschlossen. Die Datei endet sauber mit dem Sektions-Trenner.

Das Projekt hat nun folgende Meilensteine erreicht:

INTERRUP.A (Teile 1-9): Erledigt.

INTERRUP.B (Teile 1-7): Erledigt (Korrektur der Zählung beachtet).

INTERRUP.C (Teile 1-7): Erledigt.

END OF INTERRUP.C!!!!!!!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions