Previous topic |
Next topic |
Contents |
Contact z/OS |
Library |
PDF
Multipart program objects z/OS MVS Program Management: User's Guide and Reference SA23-1393-00 |
|
Under certain circumstances, the binder will create a program object with multiple segments. When loading this type of module, each segment has a different load point. The binder currently uses RMODE and time of load (initial or deferred) of the classes as the criteria for splitting a module into segments. If all input is from traditional object modules (not XOBJ or GOFF) or load modules, a multipart module is created only if RMODE(SPLIT) is specified. If using the assembler CATTR support to create user-defined text classes (supported only in GOFF object format), or if GOFF or XOBJ object modules from a compiler are part of the input, there can be text classes other than B_TEXT. The RMODE of these additional text classes is used to assign each of these classes to a segment. In addition, deferred load classes, such as C_WSA, are each placed in a separate segment. There is no binder external to override this division into segments. Certain restrictions apply to multipart modules. If you use the capabilities of the High Level Assembler or the
binder RMODE(SPLIT) option to create multipart program objects, certain
restrictions apply.
If a multipart program object is subsequently loaded through a directed load or by the binder, all text classes are loaded into consecutive storage locations according to the minimum RMODE value for all loaded classes. |
Copyright IBM Corporation 1990, 2014
|