JPH07210397A - Program conversion method and program execution method using the same - Google Patents

Abstract

PURPOSE:To provide program conversion and execution methods for converting an instruction sequence in a first program to an instruction sequence of a second instruction set without analyzing a program. CONSTITUTION:A program conversion part 115 in a program conversion application 113 receives the input of the first program 111, uses an instruction conversion information table 117, acquires a conversion instruction routine number for specifying a conversion instruction routine corresponding to the instruction sequence in the first program and converts the instruction sequence in the first program to the instruction sequence of the second instruction set by using a conversion instruction routine file 119 corresponding to the acquired conversion instruction routine number. Thus, the number of the instructions of the second instruction set to be executed per instruction of the first program is reduced.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an instruction sequence in a first program represented for a first processor having a first instruction set to a second processor having a different second instruction set. And a program execution method using the program conversion method.

[0002]

2. Description of the Related Art In recent years, due to advances in computer technology,
Processors are changing rapidly. As a result, the program represented for the previous processor cannot be executed on the new processor. In order to execute the program represented for the first processor having the first processor by the second processor having a different second instruction set, an instruction sequence in the program represented for the first processor is used. Is converted into an instruction sequence for a different second processor and executed by the second processor.

For example, Proceedings of the 1
992 ACM International Conference on Super Computing (Proceeding of
the1992 ACM International Conference on Supercomp
In uting (ICS '92), p198 to p215) (publicly known example 1), the program analysis can be performed before the program execution, and the instruction sequence in the first program can be converted into the instruction sequence of the prepared second instruction set. Converts into a program in which the instructions for executing the corresponding instruction sequence of the second instruction set are embedded. Then, the converted program is executed by the second processor prepared with an interpreter simulating the first processor.

Here, this interpreter is represented by an instruction sequence of the second instruction set. At that time, the instruction sequence of the first program having no corresponding instruction sequence of the second instruction set is executed by the interpreter, and the instruction sequence of the first program having the corresponding instruction sequence of the second instruction set corresponds. The instruction sequence of the second instruction set is executed.

Further, Japanese Patent Laid-Open No. 2-5139 (known example 2)
Then, analyze the instruction flow of the first program by graph,
The instruction flow of the second instruction set is converted, and the first program is simulated by the second processor.

According to this method, it is detected in the program conversion process by the graph analysis whether or not the subsequent instruction uses the condition code, and the condition code setting process of the instruction which does not need the condition code setting is omitted. As a result, the number of instructions of the second instruction set executed per instruction of the first program is reduced as compared with the case where the instructions of the first program are converted into instruction sequences of the second instruction set for each instruction. However, the processing performance is improved.

Further, Japanese Patent Laid-Open No. 2-5140 (known example 3)
Now, check the memory update instruction in the first program,
The instruction sequence of the second instruction set converted according to the instruction before the change is determined if the instruction changes the subsequent instruction in the first program and the instruction is changed during the program execution. To be executed.

In this specification, "instruction sequence"
Is used as a term indicating a single instruction or a plurality of consecutive instructions.

[0009]

In the known example 1, when the first program is analyzed and the instruction sequence in the first program can be converted into the corresponding instruction sequence of the second instruction set,
The instruction for executing the instruction sequence of the corresponding second instruction set is embedded, but the program analysis is required before the execution.

Further, in the known example 2, the instruction flow of the first program is analyzed, and the number of instructions to be executed is reduced by omitting the instruction sequence for performing unnecessary processing from the second program, and the processing performance is improved. is doing. However, also in this case, program analysis is required. As described above, in addition to the known examples 1 and 2, the program analysis is required, and the program analysis method is complicated.

A first object of the present invention is to provide an instruction sequence for a second processor having a second instruction set different from the instruction sequence in the first program represented for the processor having the first instruction set. With the program conversion method for converting into 1 to 1 instructions of the first program, rather than analyzing the instructions of the 1st program into the instruction sequence of the 2nd instruction set without analyzing the program. It is to provide a program conversion method for reducing the number of instructions of the second instruction set to be executed.

Further, in the known example 3, by checking the memory update instruction in the first program, when the instruction is changed during the execution of the program, the second instruction converted according to the instruction before the change is executed. Although this prevents the set of instruction sequences from being executed, this method requires examination of memory update instructions.

A second object of the present invention is to provide a second processor having a different second instruction set from the instruction sequence in the first program represented for the first processor having the first instruction set. If the instruction is converted during the program execution without investigating the instruction in the program by the program execution method that is converted to the instruction sequence for execution by the second processor, it is converted according to the instruction before the change. Another object of the present invention is to provide a program execution method for preventing an instruction sequence of the second instruction set from being executed.

[0014]

In order to achieve the first object, in the program conversion method of the present invention, a second instruction set called a conversion instruction routine corresponding to an instruction sequence of the first instruction set is used. Obtaining a conversion instruction routine number designating a conversion instruction routine corresponding to an instruction string in the first program using an instruction conversion information table having a conversion instruction routine number designating an instruction string; Converting the instruction sequence in the first program into an instruction sequence of the second instruction set by using a conversion instruction routine file having a plurality of conversion instruction routine files. Then, the conversion step is performed.

In order to achieve the second object,
In the program execution method of the present invention, a step of acquiring a conversion instruction routine number designating a conversion instruction routine corresponding to an instruction sequence in the first program using the instruction conversion information table; Using the step of converting the instruction sequence in the first program into the instruction sequence of the second instruction set, and the step of executing the converted instruction sequence in the second processor, the execution step, This is performed every time the instruction sequence in the first program is converted into the instruction sequence of the second instruction set by the program conversion method.

[0016]

According to the program conversion method of the present invention, the conversion instruction routine number designating the conversion instruction routine corresponding to the instruction sequence in the first program is acquired using the instruction conversion information table, and the acquired conversion instruction routine number is acquired. In response to the,
The conversion instruction routine file is used to convert the instruction sequence in the first program into the instruction sequence of the second instruction set. In the conversion instruction routine, when the instruction sequence in the first program is an instruction sequence consisting of a plurality of instructions, it is possible to delete an instruction sequence of unnecessary processing or use a high function instruction in the second instruction set. .

From the above, the instruction sequence in the first program represented for the first processor having the first instruction set is converted into the instruction sequence for the second processor having the different second instruction set. The program conversion method executes per instruction of the first program, rather than analyzing each instruction in the first program into an instruction sequence of the second instruction set without analyzing the program. It is possible to reduce the number of instructions in the second instruction set.

Further, according to the program execution method of the present invention,
The conversion instruction routine number that specifies the conversion instruction routine corresponding to the instruction string in the first program is obtained using the instruction conversion information table, and the instruction string in the first program is acquired using the conversion instruction routine file. Every time the program is converted into the instruction sequence of the second instruction set by the program conversion method, the converted instruction sequence is executed by the second processor.
Further, if the conversion instruction routine file is not provided with a conversion instruction routine of an instruction string in which a memory update instruction is further followed by an instruction, even if the memory update instruction changes the instruction, it does not affect the program execution.

As described above, the instruction sequence of the first program represented for the first processor having the first instruction set is converted into the instruction sequence for the second processor having a different second instruction set, In the program execution method executed by the second processor, if the instruction is changed during the program execution without investigating the instruction in the program, the instruction of the second instruction set converted according to the instruction before the change It becomes possible to prevent the columns from being executed.

[0020]

First, a program conversion method will be described as a first embodiment. The feature of the first embodiment is that by obtaining information such as the instruction sequence and the conversion instruction routine number from the instruction conversion information table, the instruction sequence in the first program can be set in the second instruction set without analyzing the program. It is to convert to the instruction sequence of.

FIG. 1 is a block diagram showing the outline of a program conversion system according to an embodiment of the present invention. In FIG. 1, 1
Reference numeral 01 is an input to the program conversion system, 103 is a CPU of the program conversion system, and 105 is a memory of the program conversion system.

The program conversion system takes as input a first program 111 represented for a first processor having a first instruction set.

The memory also has a program conversion application 113 for converting the instruction sequence in the first program into the instruction sequence of the second instruction set. The program conversion application is called a program conversion unit 115 that converts an instruction sequence in the first program into an instruction sequence of the second instruction set, and a conversion instruction routine corresponding to the instruction sequence of the first instruction set. It has an instruction conversion information table 117 having a conversion instruction routine number that specifies an instruction sequence of the second instruction set, and a conversion instruction routine file 119 having a plurality of conversion instruction routines.

The program conversion unit 115 takes out an instruction sequence from the first program 111, converts it into an instruction of the corresponding second instruction set using the instruction conversion information table and the converted instruction routine file, and converts the converted instruction. Row 121
Is stored in memory.

FIG. 2 is a diagram showing a first program example. In the first program example 131, the instruction TM is a comparison instruction, and the condition code is set based on the comparison result. The instruction BC is a conditional branch instruction. If the branch condition is satisfied, the instruction BC branches to the designated address, and if not satisfied, the subsequent instruction (instruction L in this example) is executed.

The command in the first program that is an input is
If each instruction can be identified like assembler code,
An instruction number is attached to each instruction as an instruction identifier. When each instruction cannot be identified like an object code, an address may be attached and each instruction may be identified by the start address of the instruction. Hereinafter, as the first program, a program in which each instruction is given an instruction number will be taken up.

In order to convert the instruction sequence in the first program into the instruction sequence of the second instruction set, the program conversion application holds an instruction conversion information table and a conversion instruction routine file in addition to the program conversion unit.

FIG. 3 is a part of an example of the instruction conversion information table. The instruction conversion information table in the program conversion method of the present invention includes an instruction sequence indicating a sequence of instructions of the first instruction set having a corresponding conversion instruction routine and an instruction sequence in which the instruction sequence is further followed by instructions. As information, the possibility of the subsequent instruction indicating the presence or absence, the conversion instruction routine number that specifies the conversion instruction routine corresponding to the instruction order, and the number of converted instructions indicating the number of instructions of the first instruction set appearing in the instruction order are used as information. It is defined by the table format 131.
When an address is assigned to the instruction of the first program and each instruction is identified by the start address of the instruction, it is possible to give the number of bytes occupied by the instruction order as an alternative to the number of converted instructions. . Further, the conversion instruction routine file has a plurality of conversion instruction routines corresponding to the instruction sequence of the first instruction set, and each conversion instruction routine can be designated by a conversion instruction routine number.

Here, as the instruction sequence, [AR r
1, r2], [TM 10 (r2), 0x55], [TM
10 (r2), 0x55-BC 0x5,10 (r2, r
3)], [L r3,10 (r1)] are defined. At this time, AR, TM, BC, and L are mnemonics of the instruction, and [TM 10 (r2), 0x55-BC 0x.
5, 10 (r2, r3)] is an “instruction [TM 10 (r2),
0x55] followed by the instruction [BC 0x5, 10 (r2,
r3)] follows ”.

In the instruction conversion information table of FIG. 2, the instruction [TM 10 (r2), 0x55] has a possibility that there is an instruction sequence in which the instructions continue, and the instruction [TM10 (r2), 0x55]
The conversion instruction routine number corresponding to (r2), 0x55] is routine 2, and the number of converted instructions is 1. In addition, the instruction sequence [TM 10 (r2), 0x55-BC0x5, 10
In (r2, r3)], there is no possibility that there is an instruction sequence in which more instructions continue, the corresponding conversion instruction routine number is 21, and the number of converted instructions is 2.

FIG. 4 is a portion of an example conversion instruction routine file. The conversion instruction routine file in the program conversion method of the present invention has a conversion instruction routine corresponding to the instruction sequence of the first instruction set. Here, as the instruction sequence, the instruction [TM 10 (r2),
0x55] and the conversion instruction routine 135 of the conversion instruction routine number 2 and the instruction [TM 10 (r2), 0x5
5-BC 0x5,10 (r2, r3)], the conversion instruction routine 137 of the conversion instruction routine number 21 is shown.

FIG. 5 is a flow chart showing the program conversion processing of the first embodiment.

First, step (hereinafter referred to as S) 301
To set the conversion target instruction number to 1. Next, in S303, it is determined whether or not the conversion target instruction number is equal to the conversion end number for ending the program conversion processing. If they are equal, the program conversion processing is ended, and if they are not equal, the instruction conversion routine after S305. The number acquisition process and the conversion process are executed.

In step S305, the program conversion unit
The “instruction designated by the conversion target instruction number” is taken out from the first program and used as the conversion target instruction sequence. Next, S30
In step 7, the conversion target instruction sequence is searched for in the conversion command information table. If the conversion target command sequence exists in the command conversion information table, the process proceeds to step S309. If not, the process proceeds to step S317. S
At 309, the conversion instruction routine number and the number of converted instructions corresponding to the conversion target instruction sequence are obtained from the conversion instruction information table. Then, in S311, whether or not there is a possibility of a subsequent instruction in the conversion target instruction sequence is determined using the conversion information table. If there is a possibility of a subsequent instruction, the process proceeds to S313, and if not, the process proceeds to S317. In S313, when the value obtained by adding 1 to the conversion target instruction number is not equal to the conversion end number, the process proceeds to S315, and when the value is equal, the process proceeds to S321. S31
In 5, the "instruction string obtained by adding the next instruction to the conversion target instruction string" is set as the conversion target instruction string, and the process proceeds to S307 and S30.
Processing after 7 is performed.

In S307, if there is no conversion target instruction sequence in the instruction conversion information table, or in S311,
If there is no possibility that there is an instruction sequence in which further instructions continue, the process proceeds to S317. In S317, the conversion instruction routine specified by the conversion instruction routine number is taken out from the conversion instruction routine file, and the instruction of the second instruction set is acquired. Convert to a column. Then, in S319, the "value obtained by adding the number of converted instructions to the conversion instruction number" is set as the conversion instruction number, the process proceeds to S303, and the processes after S303 are performed.

If the value obtained by adding 1 to the conversion target instruction number is equal to the conversion end number in S313, the process proceeds to S321. In S321, the conversion instruction routine designated by the conversion instruction routine number is extracted from the conversion instruction routine file. , The instruction sequence of the second instruction set is converted, and the program conversion process is terminated.

Regarding the program conversion method, [TM 10 (r2), 0 in the first program example shown in FIG. 2 is used.
x55] and [BC 0x5, 10 (r2, r3)] instruction sequences.

In describing the program conversion method of the present invention, it is assumed that the current conversion target instruction number is 5 and the conversion end number is 100. In addition, the command [TM 10
(r2), 0x55] sets the condition code to 3, and the instruction [BC 0x5, 10 (r2, r3)] is an instruction that satisfies the branch condition. The instruction conversion information table uses the instruction conversion information table shown in FIG. 3, and the conversion instruction routine file uses the conversion instruction routine file shown in FIG.

In S303 of FIG. 5 showing the program conversion processing of the present invention, the current conversion target instruction number is 5, which is not equal to the conversion end number 100, so the routine proceeds to S305.

In S305, the instruction [TM 10 (r2), 0
x55] as the conversion target instruction sequence, and in S307, the instruction [TM 10 (r2), in the instruction conversion information table shown in FIG.
0x55] is the instruction order, the process proceeds to S309. In step S309, the conversion instruction routine number 2 and the converted instruction number 1 are obtained from the conversion instruction information table. And
In step S311, the instruction [TM 10 (r2), 0x55] may have an instruction sequence in which more instructions continue.
Proceed to 13. In S313, since the conversion target instruction number 5 plus 1 is not equal to the conversion end number 100,
Proceed to S315. In step S315, the "conversion target instruction sequence [TM
10 (r2), 0x55] and the next instruction [BC 0x5, 1
0 (r2, r3)] as a sequence of instructions ”is the sequence of instructions to be converted.

Next, in step S307, the instruction sequence [TM10 (r2), 0x55-B in the instruction conversion information table shown in FIG. 3 is used.
Since C 0x5,10 (r2, r3)] is the instruction order, the process proceeds to S309. In step S309, the conversion instruction routine number 21 and the converted instruction number 2 are obtained from the conversion instruction information table. Then, in S311, the instruction sequence [TM10 (r2),
0x55-BC 0x5,10 (r2, r3)] has no possibility that there is an instruction sequence in which more instructions continue, so S317
Proceed to.

In step S317, the conversion instruction routine designated by the conversion instruction routine number 21 is fetched from the conversion instruction routine file shown in FIG. 4 and converted into the instruction string of the second instruction set. Then, the process proceeds to S319, where the conversion target command number 5 plus the number of converted commands 2 is 7 is the conversion target command number, and the process proceeds to S303 where the conversion command routine number acquisition process and conversion process for the commands after the conversion target command number 7 are performed. I do.

FIG. 6 shows [TM 10 (r2), 0x55] and [BC 0x5, 1 in the program shown in FIG.
0 (r2, r3)] is an instruction sequence example 141 when the instruction sequence is converted into the instruction sequence of the second instruction set for each instruction in the program without performing the program analysis. FIG. 7 shows an instruction sequence example 143 when the program conversion method of the present invention converts into an instruction sequence of the second instruction set.

In FIGS. 6 and 7, the instruction [TM10 (r2), 0x55] sets the condition code 3 and the instruction [BC 0x5, 10 (r
2, r3)] is an instruction executed when a branch is taken. When converted for each instruction, 16 instructions are executed, whereas when converted by the program conversion method of the present invention, 13 instructions are executed, and thus, by the program conversion method of the present invention, It can be seen that the number of instructions of the second instruction set executed per instruction of the first program has decreased.

When the conversion is performed for each instruction, the instruction [TM
10 (r2), 0x55], the condition code is set as a result of the comparison process, and the instruction [BC 0x5, 10 (r2, r
3)], the branch processing is performed after checking the condition code set as the branch determination processing with the branch condition. On the other hand, when the two instructions are collectively converted, the comparison process and the branch determination process are simultaneously performed using the high-performance instruction of the second instruction set. In this way, when a plurality of instructions in the first program are collectively converted into an instruction sequence of the second instruction set, the number of instructions of the second instruction set executed per instruction of the first program can be reduced. You can

Next, a program execution method will be described as a second embodiment. The feature of the second embodiment is that the instruction sequence converted every time the instruction sequence of the first program is converted into the instruction sequence of the second instruction set by using the above-described program conversion method for the program execution method. To perform.

The outline of the program execution system will be described with reference to the block diagram of FIG. In FIG. 8, 151
Is an input to the program execution system, 153 is a CPU of the program execution system, 155 is a memory of the program execution system, and 157 is a magnetic memory for holding the output of the program execution system.

The program execution system takes as input a first program 161 represented for a first processor having a first instruction set.

The memory also includes a program conversion application 163 for converting the instruction sequence in the first program into the instruction sequence of the second instruction set. The program conversion application is referred to as a program conversion unit 165 that converts an instruction sequence in the first program into an instruction sequence of the second instruction set and a conversion instruction routine corresponding to the instruction sequence of the first instruction set. It has an instruction conversion information table 167 having a conversion instruction routine number designating an instruction sequence of the second instruction set, and a conversion instruction routine file 169 having a plurality of conversion instruction routines.

The program conversion unit 165 extracts an instruction sequence from the first program 161, converts it into a corresponding instruction of the second instruction set using the instruction conversion information table and the converted instruction routine file, and converts the converted instruction. Row 171
Is stored in the memory, the CPU 153 executes the converted instruction sequence, and outputs the program execution result 173.

As in the first embodiment, the first program to be an input has an instruction number attached to each instruction.

Here, the instruction conversion information table and the conversion instruction routine file in the program execution method of the present invention have the same structure and contents as the instruction conversion information table and the conversion instruction routine file in the program execution method. However, in order to achieve the second object, the instruction order, which is the information of the instruction conversion information table, does not specify the instruction order in which the memory update instruction is followed by further instructions.

FIG. 9 is a flow chart showing the program execution processing of this embodiment.

First, in S351, the conversion target instruction number is set to 1
To Next, in S353, it is determined whether or not the conversion target instruction number is equal to the conversion end number for ending the program conversion processing. If they are equal, the program conversion processing is ended, and if they are not equal, the instruction conversion routine after S355. Executes number acquisition processing, conversion processing, and execution processing.

The program conversion unit, at S355,
The “instruction designated by the conversion target instruction number” is taken out from the first program and used as the conversion target instruction sequence. Next, S35
In step 7, the conversion target instruction sequence is searched for in the conversion command information table. If the conversion target command sequence exists in the command conversion information table, the process proceeds to S359, and if not, the process proceeds to S367. S
At 359, the conversion instruction routine number and the number of converted instructions corresponding to the conversion target instruction sequence are obtained from the conversion instruction information table. Then, in step S361, it is determined whether or not there is a possibility of a subsequent instruction in the conversion target instruction string using the conversion information table. If there is a possibility of a subsequent instruction, the processing proceeds to step S363, and if not, the processing proceeds to step S367. If the value obtained by adding 1 to the conversion target instruction number is not equal to the conversion end number in S363, S36
5, and if they are equal, proceed to S373. In step S365, the "instruction string in which the next instruction is added to the conversion target instruction string" is set as the conversion target instruction string, the process proceeds to step S357, and the processes in and after step S357 are performed.

If there is no conversion target instruction string in the instruction conversion information table in S327, or if there is no possibility of a subsequent instruction in S361, the process proceeds to S367, but S36.
In step 7, the conversion instruction routine designated by the conversion instruction routine number is extracted from the conversion instruction routine file, converted into the instruction string of the second instruction set, and the converted instruction string is stored in the memory. In S369, the program execution unit retrieves the converted instruction sequence from the memory and executes it. And
In S371, "value obtained by adding the number of instructions to be converted to the instruction number to be converted" is set as the instruction number to be converted, and the process proceeds to S353, where S3
The processing after 53 is performed.

In S363, if the value obtained by adding 1 to the conversion target instruction number is equal to the conversion end number, the process proceeds to S373, but in S373, the conversion instruction routine designated by the conversion instruction routine number is extracted from the conversion instruction routine file. , To the instruction sequence of the second instruction set. S375
Then, the program execution unit executes the converted instruction sequence,
The program conversion process ends.

[0058]

According to the program conversion method of the present invention,
A program conversion method for converting an instruction sequence in a first program represented for a first processor having a first instruction set into an instruction sequence for a second processor having a different second instruction set, Rather than converting the instructions of the first program into instruction sequences of the second instruction set for each instruction without analyzing the program, the second instruction set of the second instruction set executed per instruction of the first program It becomes possible to reduce the number of instructions. Therefore, the first program can be executed at high speed in the second processor without requiring a complicated program analysis method.

Further, according to the program execution method of the present invention, the first program for the first processor having the first instruction set is replaced with the instruction sequence for the second processor having the different second instruction set. In the program execution method of converting to, and executing in the second processor, if the instruction is converted during the program execution without examining the instruction of the program, the second instruction converted according to the instruction before the change It becomes possible to prevent the instruction sequence of the set from being executed.

[Brief description of drawings]

FIG. 1 is a block diagram showing an outline of a program conversion system according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a first program example.

FIG. 3 is an explanatory diagram showing an example of an instruction conversion information table.

FIG. 4 is an explanatory diagram showing an example of a conversion instruction routine file.

FIG. 5 is a flowchart showing a program conversion process of the first embodiment.

FIG. 6 is an explanatory diagram showing an example of an instruction sequence when converted for each instruction.

FIG. 7 is an explanatory diagram showing an example of an instruction sequence when converted by the program conversion method of the present invention.

FIG. 8 is a block diagram showing an outline of a program execution system of a second embodiment.

FIG. 9 is a flowchart showing a program execution process of the second embodiment.

[Explanation of symbols]

101 ... Input to program conversion system, 103 ... CPU of program conversion system, 105 ... Memory of program conversion system, 111 ... First program, 11
3 ... Program conversion application, 115 ... Program conversion unit, 117 ... Command conversion information table, 119 ...
Conversion instruction routine file, 121 ... Converted instruction string.

Claims (2)

[Claims] 1. An instruction sequence in a first program represented for a first processor having a first instruction set is converted into an instruction sequence for a second processor having a different second instruction set. In the program conversion method, an instruction conversion information table having a conversion instruction routine number that specifies an instruction string of the second instruction set, which is called a conversion instruction routine corresponding to an instruction string of the first instruction set, is used, A step of obtaining a conversion instruction routine number that specifies a conversion instruction routine corresponding to an instruction sequence in one program; and a conversion instruction routine file having a plurality of the conversion instruction routines, using the instructions in the first program. Converting the sequence into an instruction sequence of the second instruction set, the program conversion unit according to the conversion instruction routine number. A method for converting a program, which comprises performing a conversion step. 2. An instruction sequence in a first program represented for a first processor having a first instruction set is converted into an instruction sequence for a second processor having a different second instruction set. A program execution method executed by a second processor, the instruction conversion having a conversion instruction routine number designating an instruction string of the second instruction set, which is called a conversion instruction routine corresponding to the instruction string of the first instruction set. A step of acquiring a conversion instruction routine number designating a conversion instruction routine corresponding to an instruction sequence in the first program using the information table; and a conversion instruction routine file having a plurality of the conversion instruction routines, And converting the instruction sequence in the first program into an instruction sequence of the second instruction set, and executing the converted instruction sequence on the second processor. And the instruction sequence in the first program is
The program executing method is characterized in that the executing step is performed every time the instruction sequence of the instruction set is converted.