nld and noft Manual

nld Utility

nld and noft Manual—520384-003

2-25

Run-Time Consistency Checking

If a program tries to use a processor that does not support IEEE floating-point format,

then process creation error code 64 occurs (IEEE floating-point support not

available on this processor). To determine whether a processor can run IEEE

floating-point instructions, programs can call the PROCESSOR_GETINFOLIST_

procedure. For more information about this procedure, see the Guardian Procedure

Calls Reference Manual.

Run-time consistency checking includes a check for floattype consistency between

the program file and the user library file, if one is used. IEEE and Tandem floating-point

formats use different data formats and calling conventions: IEEE floating-point values

are typically passed in IEEE floating-point registers, while Tandem floating-point values

are passed in general purpose registers. Problems can occur if a function using one

floating-point format calls a function using the other format. Checks are performed at

process creation to ensure that the user library (if there is one) has a floattype

compatible with that of the program file. If not, the program is not allowed to run.

If the floattype attribute of the program file is IEEE and floattype attribute of the

user library file is Neutral, the program is allowed to run, and the C/C++ run-time

libraries operate in IEEE mode.

Even if the user library is marked with a floattype attribute that conflicts with the

program file, the program can use the library if the program does not call anything in

the user library that uses floating point. In this case, you must mark the program file

with the -set float_lib_overrule on command to disregard the floattype

attribute of the user library file (see Overruling the Run-Time Consistency Check on

page 2-26).

TNS/R Native User Libraries

If you are using a TNS/R native user library, the library file must use the same floating-

point format as the program, and the library must be marked accordingly. If the user

library does not use floating point at all, you can mark the library NEUTRAL_FLOAT

using the -set floattype or -change floattype option. Then the user library

can be used by any kind of program.

Table 2-5. Incompatible Program File and User Library floattypes (TNS)

floattype in program file floattype in user library file

IEEE Tandem

Tandem IEEE

Neutral IEEE