JPS63175929A - Dynamic subprogram loader - Google Patents

Abstract

PURPOSE:To eliminate the intervention of an operating system (OS), and also, to make it unnecessary to execute reloading at the time of linking a subprogram, by making a link of the subprogram static. CONSTITUTION:Each main program object is not a subprogram number, but uses a static link system having an address, and after a relocatable program loader has rearranged, a subprogram, a procedure part in a program area of a system is scanned, and when a data pattern consisting of a subprogram link instruction (CALL) and an old address of the subprogram is detected, the old address is rewritten to a new address. In such a way, at the time of linking the subprogram, the intervention of the OS is eliminated, and reloading of the program after the subprogram has been rearranged is not required.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a loader for allocating programs to a computer system, and particularly to a method for dynamically relocating subprograms.

[Conventional technology]

Subprograms (subroutines, macroroutines) in online real-time computer systems.

There are two types of linking methods to functions, etc.), and they are called dynamic linking and static linking, respectively. The dynamic link method shown in Figure 3 is one in which the operating system (hereinafter abbreviated as oS) manages all of the subprograms by sequentially numbering them, and stores the address of the corresponding subprogram in O8. are doing. A general program object only has a subprogram number, and when O8 finds a subprogram link instruction during program execution, it refers to the subprogram address table and sets it in the execution instruction address counter (PC). The advantage of this method is that once the main program has been placed, even if it has to be relocated due to modification or malfunction of the subprogram, and it cannot be placed in the same area, all that is required is to rewrite the O8 internal program address table value. There is no need to re-read programs linked to the program. This effect is significant because subprograms are used in many programs due to their nature.

On the other hand, the static link method shown in Figure 4 is a method in which each program has the address of the subprogram.The advantages are that the structure is simple and that the processing time is fast because it can be linked to the subprogram without O8 intervention. That's true.

Related technologies include, for example, Japanese Patent Application Laid-Open No. 59-18
There is No. 0736.

[Problem that the invention attempts to solve]

Each of the above conventional techniques has the following problems.

The dynamic link method has disadvantages in that the O8 intervenes when linking the subprogram, which increases processing time, and that the subprogram address management table must be secured in advance in the O8.

The disadvantage of static type is that subprogram relocation causes
All programs using the subprogram must be reloaded.

SUMMARY OF THE INVENTION An object of the present invention is to provide a sub-program loader that eliminates the above-mentioned drawbacks and retains its advantages.

[Means for solving problems]

The above object is achieved by firstly eliminating the intervention of O8 when linking subprograms, and secondly by providing a means for eliminating the need for program reloading after subprogram relocation. In other words, each main program object is statically linked with an address rather than a subprogram number, and after relocating the subprogram, the relocatable program loader scans the procedure section in the program area of the system and issues a subprogram link instruction (CALL). )
When a data pattern consisting of an old address of the subprogram is detected, the above is achieved by rewriting the old address with a new address.

[Effect]

As described above, since the link to the subprogram is static, there is no intervention from O8, and since the subprogram address in each main program is rewritten, there is no need for reloading. This eliminates the aforementioned drawbacks and makes it possible to realize a relocatable program loader that retains only the advantages.

〔Example〕

Two embodiments of the present invention will be described below with reference to FIG.

This example is composed of five logics as shown in the figure.

If the subprogram to be loaded has already been registered, logic A saves the address of the corresponding subprogram.

Subsequently, logic B is relocated to a new area different from the subprogram. This is the same as a conventional relocatable program loader.

Relocation, logic, E scans all program areas of the system and rewrites the addresses of the subprograms.

FIG. 2 shows the operation of the logic logic E. Programs can be divided into two types: resident in the main memory and non-resident, but logically E applies to both programs. The program is the procedure department.

Although it is divided into a constant part, a work area, etc., the pattern consisting of the CALL instruction and address in this example is unique only within the procedure division, so it is logical.

E searches only the procedure division. If the pattern is found, the stone printing program address is rewritten to a new address.

According to this embodiment, even if the address changes due to relocation of a subprogram, there is no need to reload other programs linked to it, and no intervention by O8 is required when the program is executed.

〔Effect of the invention〕

According to the present invention, since it is a static link type, O8
No intervention is required, resulting in faster processing speeds. Furthermore, since the addresses of subprograms scattered within each program are rewritten by the loader when subprograms are relocated, there is no need to reload programs. Furthermore, in the past, it was necessary to stop the system when replacing a common subprogram, but in the present invention, addresses are sequentially rewritten, and during that time, the old subprogram remains, so there is no need to stop the system. effective.

[Brief explanation of the drawing]

Fig. 1 is a flowchart of a dynamic subprogram loader according to an embodiment of the present invention, Fig. 2 is an explanatory diagram of operation, Fig. 3 is an explanatory diagram of a conventional subprogram dynamic link method, and Fig. 4 is a similar diagram. FIG. 2 is an explanatory diagram of a subprogram static link method. Agent Patent Attorney Katsuo Ogawa −′ City 1st Ward

Claims (1)

[Claims] 1. In the relocatable program loader of an online real-time computer system, search the old subprogram address save logic that stores the address of the old subprogram, the new subprogram relocation logic, and all procedure divisions in the program area of the system. (A dynamic subprogram loader characterized by having logic for rewriting the old address to a new address when a pattern consisting of a subprogram link instruction and an old subprogram address is found.