APPENDIX A. MONITOR SYSTEM ERROR AND OPERATIONAL MESSAGES

With the exception of the System Library Mainline Programs, this appendix lists all Monitor System WAITS and messages. SYSUP, the DCOM update subroutine, is also available in the System Library. The errors for the user callable version of SYSUP are listed in the System Library Utility Subroutines section of the manual. All messages for stand-alone utilities are included in the writeups of the individual programs.

System Loader and FORTRAN I/O errors cause the system to WAIT at $PRET. At the WAIT, bits 2 and 3 of the OPERATION REGISTER are on. FORTRAN I/O errors can be identified by the Fxxx code in the accumulator. A $PRET WAIT also occurs when a system I/O device is required but is not ready (see Table 18).

All error tables in this appendix are listed alphabetically by prefix letter. Unless otherwise noted, the operational and error messages are printed on the principal printer. All Monitor system control records are printed on the principal printer.

The error tables in order of appearance are as follows.

ASSEMBLER MESSAGES AND ERROR CODES

At the completion of an assembly, the following messages are printed on the principal printer.

    XXX OVERFLOW SECTORS SPECIFIED
    XXX OVERFLOW SECTORS REQUIRED
    XXX SYMBOLS DEFINED
    XX ERROR(S) FLAGGED IN ABOVE ASSEMBLY

If LIST DECK or LIST DECK E is specified, the error detection codes shown in Table 7 are punched in columns 18 and 19. For the first error detected in each statement the Assembler stores and then punches the cede in column 18; the cede for a second error is stored, overlaid by any subsequent errors, and punched in column 19. Thus, if more than two errors are detected in the same statement, only the first and last are indicated. These error detection codes will appear on the printout if the deck is listed.

Table 7. Assembler Error Detection Codes

At the end of the assembly, a message is printed indicating the number of assembly errors detected in the source program (see above). Since no more than two errors are flagged per statement, the error count may exceed the actual number of flags.

Assembler error messages are listed in Table 8.

FORTRAN MESSAGES AND ERROR CODES

Compilation Messages

Near the end of the compilation, core usage information and the features supported (control records used) are printed out as follows:

    FEATURES SUPPORTED
        EXTENDED PRECISION
        ONE WORD INTEGERS
        TRANSFER TRACE
        ARITHMETIC TRACE
        IOCS

    CORE REQUIREMENTS FOR XXXXX

    COMMON YYYYY VARIABLES YYYYY PROGRAM YYYYY

where XXXXX is the name of the program designated in the *NAME control record or in the SUBROUTINE or FUNCTION statement, and YYYYY is the number of words allocated for the specified parts of the program.

The following message is printed for a normal end of compilation (with or without errors):

    END OF COMPILATION

Table 8. Assembler Error Messages

Compilation Error Messages

During compilation, a check is made to determine if certain errors have occurred. If one or more of these errors have been detected the error indications are printed at the conclusion of compilation, and no object program is stored on the disk. Only one error is detected for each statement. In addition, due to the interaction of error conditions, the occurrence of some errors may prevent the detection of others until those which have been detected are corrected. With the exception of the COO error message and those noted below, the error message appears in the following format:

    CNN ERROR IN STATEMENT NUMBER XXXXX+YY

NN is the error code number listed in Table 9. With the exception of specification statement errors, XXXXX is the last valid statement number preceding the erroneous statement and YYY is the count of statements from XXXXX to the statement that is in error. If the erroneous statement has a valid statement number, XXXXX will be the statement in error and YYY will not be printed. For example:

  105 FORMAT (15, F8.4)
  110 IF (A-B) 10, 30, 20
      A=A+1.0
  ABC B = B-2.0     (error CO1)
  135 GO TO 105     (error C43)

This example will cause the following error messages to be printed.

    C01 ERROR IN STATEMENT NUMBER 110 + 002
    C43 ERROR IN STATEMENT NUMBER 135

For specification statements, XXXXX is always 00000 and YYY is the count of the number of specification statements in error. YYY is never 000, i. e., for the first error YYY is 001. Specification statements are not counted unless they contain an error. Statement numbers on specification statements and statement functions are ignored. NN is the error code. For example:

   1  DIMENSION C(10,10)
   2  DIMENSION D(5,5)
   3  DIMENSION E(I,6,6)    (error C08)
   4  DIMENSION F(4,4)
   5  DIMENSION G(2,2))     (error C16)

This example will cause the following error messages to be printed.

    C08 ERROR AT STATEMENT 00000 + 001
    C16 ERROR AT STATEMENT 00000 + 002

In addition to the CNN type of error the following error messages may be encountered:

    PROGRAM LENGTH EXCEEDS CAPACITY

The error occurs when the program in internal compiler format is too large to be contained in core working storage, and the program must be reduced in size in order to compile.

    SUBROUTINE INITIALIZE TOO LARGE

During compilation of Sub-programs a subroutine initialize statement (CALL SUBIN) is generated.

The CALL SUBIN statement initializes all references to "dummy" variables contained within the subprogram to the appropriate core location in the calling program.

The nature of the FORTRAN compiler limits the size of any statement in internal compiler format to 511 words. In the case of CALL SUBIN, the size is calculated by the following formula:

    S = 5 + ARG + N

where ARG is the number of arguments in the subroutine parameter list and N is the total number of times the dummy arguments are used within the subprogram. S is the total size of the CALL SUBIN statement; if S ever exceeds 511, an error occurs and the above error message is printed.

If any of these errors are detected during compilation, the message:

    OUTPUT HAS BEEN SUPPRESSED is also printed.

If a monitor control record is encountered prior to an END card, the message

    MONITOR CONTROL RECORD ENCOUNTERED

is printed, and control is returned to the monitor supervisor.

The above error and the C00 error will interrupt the compilation. In this case the message

    COMPILATION DISCONTINUED

will be printed.

Table 9. FORTRAN Error Codes

* The detection of a code 65, 66, or 67 error prevents any subsequent detection of any of these three errors.

DUP MESSAGES AND ERROR CODES

When a DUP function is performed without error, an informational message is printed on the principal printer.

On a DEFINE VOID, one of the following messages is printed.

    ASSEMBLER VOIDED
    FORTRAN VOIDED

On a DEFINE FIXED AREA, the message is as follows,

    CART ID XXXX CYLS FXA XXXX DBS AVAIL XXXX FLET SECTOR ADDR XXXX

where

• CYLS FXA XXXX is the decimal number of cylinders -l in the Fixed Area. The additional cylinder is used for FLET.
• DBS AVAIL XXXX is the hexadecimal number of disk blocks remaining in the Fixed Area following the last program or data file.
• FLET SECTOR ADDR XXXX is the hexadecimal sector address of the first cylinder in the Fixed Area, i. e., the sector address of FLET.

On a dump of LET or FLET, the printout is followed by a sign-off message.

    END OF DUMPLET/FLET

All other DUP operations are followed by the following message.

    CART ID XXXX DB ADDR XXXX DB CNT XXXX

where

• DB ADDR XXXX is the hexadecimal starting address of the program or data file.
• DB CNT XXXX is the hexadecimal number of disk blocks being deleted, stored, or dumped.

DUP error messages are listed in Table 10.

Table 10. DUP Error Messages

SYSTEM LOADER MESSAGES AND ERROR CODES

No informational messages are printed during an initial load. At the completion of a reload, the following message is printed,

    END RELOAD

Table 11 lists the System Loader Errors.

Table 11. System Loader Errors

Procedure A

If cards are being read from a 1442 Card Read Punch:

1. Lift the remaining cards from the hopper and press nonprocess run out (NPRO).
2. Correct the card in error (first card nonprocessed out) and place the two nonproceased cards ahead of the cards removed from the hopper.
3. Place the deck back in the hopper.
4. Press reader START.
5. Press console PROGRAM START.

If cards are being read from a 2501 Card Reader:

1. Lift the remaining cards from the hopper and press NPRO.
2. 1. Correct the card in error (last card in stocker prior to NPRO) and place this card followed by the single nonprocessed card ahead of the cards removed from the hopper or,
   2. If the error occurred after the PHID card was read and before the type 81 card was read the System Loader is in double buffer mode. Correct the card in error (in this case the second from last card in the stacker when the error occurred) and place the last two cards from the stacker and the nonprocessed card ahead of the cards removed from the hopper.
3. Place the deck back in the hopper.
4. Press reader START.
5. Press console PROGRAM START.

Table 12. FORTRAN I/O Errors

FORTRAN I/O ERRORS

When a FORTRAN I/O error occurs, the system WAITs at $PRET with an Fxxx error code displayed in the accumulator. Table 12 lists the FORTRAN I/O errors.

SUPERVISOR MESSAGES AND ERROR CODES

The monitor Supervisor causes all Monitor system control records to be printed on the principal printer.

During a DCOM update operation (i. e., following each JOB record or user call to SYSUP) the following message is printed.

LOG DRIVE  CART SPEC  CART AVAIL  PHY DRIVE

   XXXX      XXXX        XXXX       XXXX

where

LOG DRIVE is the drive number specified on the JOB card (in the calling sequence of the SYSUP subroutine)

CART SPEC is specified cartridge ID

CART AVAIL Is the available cartridge ID

PHY DRIVE is the physical drive number starting with 0.

The logical drive may be different from the physical drive, e.g., physical drive 0 may be defined as logical drive 2.

One line is printed for each physical drive on the system.

Tables 13, 14, and 15 list Supervisor errors.

Table 13. Phase 1, Monitor Control Record Analyzer Errors

Table 14. Phase 2. System Control Record Program Errors (Phase 2 errors cause execution to be bypassed).

Table 15. SYSUP - DCOM Update Errors (SYSUP errors are also listed with the System Library Utility Subroutine SYSUP).

CORE LOAD BUILDER ERRORS

Except for the core load map described in the Programming Tips and Techniques section and messages R41-R45, the Core Load Builder prints no informational messages. Table 16 lists Core Load Builder Error Messages.

Table 16. Core Load Builder Error Messages

AUXILIARY SUPERVISOR ERRORS

The Auxiliary Supervisor prints no informational messages. Table 17 lists Auxiliary Supervisor Error messages.

ISS SUBROUTINE WAITS

A device not ready or illegal function parameter causes a pre-operative WAIT at $PRET. The ISS subroutine WAITs are listed in Table 18.

Table 17. Auxiliary Supervisor Errors

Table 18. ISS Subroutine WAITs

But wait, there's MORE...