FORTRAN PowerStation ERRORS.TXT

• Microsoft FORTRAN PowerStation for MS-DOS, version 1.0 and 1.0a

SUMMARY

The following is an extract of the ERRORS.TXT file located in the \F32\README\ directory of the Microsoft FORTRAN PowerStation installation.

   Part 1    Compiler Error Messages
   Part 2    LINK32 Error Messages
   Part 3    LINK32 Warning Messages
   Part 4    MS-DOS Extender Error Messages
   Part 5    Floating-point Exception Error Messages
   Part 6    Low Level Math Error Messages
   Part 7    Run-time Library Error Messages

MORE INFORMATION

Part 1: Compiler Error Messages

F2202   subprogram: defined with different number of arguments

F2625   subprogram: too many actual arguments

This practice is legal only when the C and VARYING attributes are specified for the subprogram.

Part 2: LINK32 Error Messages

LNK2120 number Undefined Externals The given number of external identifiers could not be resolved. This error is preceded by LINK32 warning LNK4516, once for each identifier.

LNK2123 Can't convert object to Windows NT format A non-COFF object file could not be converted to COFF.

This error is preceded by either LNK2199, which gives a specific CVTOMF error, or LNK4599, which gives a specific CVTOMF warning.

LNK2126 Unknown switch name The specified name is not a valid LINK32 option. The LINK32 options are as follows:

/DEBUG:{NONE|FULL}

 NONE Do not include debugging information.
 FULL Include all debugging information.

 Create CodeView4-format debugging information.

 Create executable file despite unresolved externals.

 Create a map file, and optionally name it.

 Specify a name for the executable file.

 Specify total (reserve) and allocated (commit) stack size.

 Write a version number in the executable-file header.

LNK2199 CVTOMF error : message A non-COFF object file could not be converted to COFF. The message gives the reason for the CVTOMF error. This error is followed by LNK2123.

Other LINK32 Error Messages

Part 3: LINK32 Warning Messages

LNK4505 No modules extracted from libraryname You specified libraryname in the LINK32 -LINK command, but no references from that library were used by your program.

LNK4514 Member module not found You ran LINK32 -LIB and specified /EXTRACT or /REMOVE, but the module was not in the library.

LNK4516 identifier is undefined One of the following may be a cause:

- The program or subprogram called a subprogram in different source files

  with different numbers of arguments.

  noncharacter type, or vice versa. FORTRAN PowerStation passes a hidden
  argument that contains the length of a string.

LNK4599 CVTOMF warning : message A problem occurred during conversion of a non-COFF object file to COFF. The message gives the reason for the CVTOMF warning. The resulting executable file may not run.

Other LINK32 Warning Messages

LNK4501 No object files specified LNK4503 No linker member found LNK4504 Bad object - fixup not aligned LNK4506 symbol multiply defined, defined in objectfile LNK4507 Bad object - fixup to far LNK4508 Bad object - relocationtype is an unknown relocation address LNK4509 Bad object - unmatched PAIR relocation - ignored LNK4510 Invalid version number argument LNK4511 Unknown subsystem argument LNK4512 Found argument, expected i386, M68K, or mips LNK4513 Image size is imagesize, KEY specified maxsize max LNK4515 commitsize commit size larger than reserve, set to number LNK4518 Can't apply gp relative fixup, gp not defined LNK4519 Bad object - incorrect string table end - fixed LNK4520 Based data overlaps code. Base code @ address LNK4521 Based code overlaps data. Base data @ address LNK4522 Found multiple matches for symbol LNK4523 Base adjusted from offset to offset (next 64K multiple) LNK4525 nodefaultlib switch in directive section in filename LNK4526 library out of date LNK4528 Obsolete switch -name ignored LNK4529 bad object file (reloc to comdat "symbol") LNK4530 Invalid file offset offset specified for section LNK4532 "symbol" has a conflicting type in objectfile LNK4533 converting object from OMF to COFF LNK4534 Using default output filename filename

Part 4: DOSXMSF MS-DOS Extender Error Messages

DX1010 DOS-Extended applications nested too deeply One of the following may be a cause:

- The nesting limit was exceeded. A protected-mode application can

  spawn a real-mode application, and a real-mode application can spawn
  a protected-mode application, up to the limit of 5 of each kind of
  application.

  run BINDMSF.EXE.

Causes of this error include the following:

- Accessing memory beyond the range of a string or array. - Passing a value argument to a subprogram that expects a reference

   argument, or vice versa.

If a Microsoft tool generated this error, note the circumstances of the error and notify Microsoft Corporation by following the instructions in the "Microsoft Support Services" section in one of your manuals.

DX1040 dllname: cannot load DLL; MS-DOS error number number The MS-DOS extender cannot load the given DLL, due to an operating-system error. The error is indicated by the given error code. Probably either DEBUG.DLL or DOSXMSF.EXE is corrupt.

DX1070 stubname stub is out of date The program you are attempting to run is not compatible with the version of DOSXNT.EXE available to the program.

The program is a 32-bit MS-DOS-extended application. A compatible version of the Microsoft DOSXNT MS-DOS Extender (DOSXNT.EXE) must exist in the same directory or in a directory specified in the PATH environment variable.

One way this error can result is when Microsoft MASM 6.1 is installed in the same directory as another language product. Either reinstall in separate directories or update from this version of MASM.

DX1140 DOSXNT.386 driver required to run with Windows For an MS-DOS-extended program that uses DOSXNT to run with the Microsoft Windows operating system, the driver called DOSXNT.386 must be specified in the [386Enh] section of SYSTEM.INI as follows:

DEVICE=path\DOSXNT.386

where path is the directory that contains the DOSXNT.386 file.

DX1180 cannot load program : reason The MS-DOS extender could not load the MS-DOS-extended program for the given reason. The reason can be one of the following:

- access denied to file

  Change the access rights for the file.

  Note the circumstances of the error and notify Microsoft Corporation by
  following the instructions in the "Microsoft Support Services" section in
  one of your manuals.

  Remove some TSRs or add more memory.

  Either close some files or increase the number of file handles specified
  in the FILES= statement in CONFIG.SYS.

DX1240 DPMICodeAlloc: could not allocate DPMI resize segment - retcode ------ = number There is insufficient memory for the MS-DOS extender to run.

DX1430 cannot load program : DLL not found Probably DEBUG.DLL is not in a directory specified in the PATH environment variable.

Other DOSXMSF DOS Extender Error Messages

Other DOSXMSF Errors

Part 5: Floating-point Exception Error Messages

M6101 invalid An invalid operation occurred. This error usually occurs when operating on a NAN (not a number) or infinity.

M6102 denormal A very small floating-point number was generated which may no longer be valid due to loss of significance. Normally, calculations proceed on denormals. To get this error message you must call SETCONTROLFPQQ without the FPCW$DENORMAL flag.

M6103 divide by 0 A floating-point operation attempted to divide by zero.

M6104 overflow An overflow occurred in a floating-point operation.

M6105 underflow An underflow occurred in a floating-point operation. Normally, calculations proceed with the underflowed value set to 0.0. To get this error message you must call SETCONTROLFPQQ without the FPCW$UNDERFLOW flag.

M6106 inexact A floating-point operation occurred where the result was not representable as an exact floating-point number. Normally, calculations proceed when an inexact value results, because almost any floating-point operation can produce an inexact result; the actual result is the rounded or truncated result as determined by the current rounding mode. To get this error message you must call SETCONTROLFPQQ without the FPCW$INEXACT flag.

Part 6: Low Level Math Error Messages

M6108 square root The operand in a square-root operation was negative. (Note: The FORTRAN intrinsic function SQRT does not generate this message; instead, SQRT generates the DOMAIN error M6201. You can get this error only if you use the fsqrt instruction in an assembly-languge procedure.

M6110 stack overflow A floating-point expression caused a stack overflow on the math coprocessor or the emulator.

M6111 stack underflow A floating-point operation resulted in a stack underflow on the math coprocessor or the emulator.

M6201 function : DOMAIN error An argument to the given function was outside the domain of legal input values for that function. For example, the following expression causes this error: SQRT(-1.0)

This error calls the matherr() function with the function name, its arguments, and the error type. You can rewrite matherr() to override this behavior.

M6202 function : SING error An argument to the given function was a singularity value for this function. The function is not defined for that argument. For example, the following expression causes this error: LOG10 (0.0)

This error calls the matherr() function with the function name, its arguments, and the error type. You can rewrite matherr() to override this behavior.

M6203 function : OVERFLOW error The given function result was too large to be represented. This error calls the matherr() function with the function name, its arguments, and the error type. You can rewrite matherr() to override this behavior.

M6205 function : TLOSS error A total loss of significance (precision) occurred. This error calls the matherr() function with the function name, its arguments, and the error type. You can rewrite matherr() to override this behavior.

Part 7: Run-time Library Error Messages

F6421 file read-only or locked against writing The program tried to transfer data to a file that was opened in read-only mode or locked against writing. The error message may indicate a CLOSE error when the fault is actually coming from WRITE. This is because the error is not discovered until the program tries to write buffered data when it closes the file.