JPH04220736A - Sharing method for binary format data file between different kind of computers - Google Patents

Abstract

PURPOSE:To effectively utilize process data by making computers which have different memory boundary characteristics to read accurately file data of the opposite computers. CONSTITUTION:The different kind of computers A and B read data out of main memories 10 and 11 in order without considering the memory boundary and store them in a shared file memory 12 to generate shared file data previously, and when those shared data are read out of the computers A and B and written on their main memories 10 and 11, address assignment is performed corresponding to the memory boundary characteristics N of the corresponding computers to accurately read the file data of the different kind of computers.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for sharing binary data files between heterogeneous computers having different memory boundary characteristics.

0002

2. Description of the Related Art FIG. 3 shows the storage state of binary format data in the main memories 20 and 21 of computers A and B, respectively, which have different memory boundaries.
The memory boundary of computer A shown by is in units of 4 bytes, and there is a memory boundary every 4 bytes from address 0.
The memory boundaries of computer B, indicated by B), are in units of 1 byte, and there are memory boundaries every 1 byte from address 0.

FIG. 4 shows the contents of data stored in the main memories 20 and 21 of each of the computers A and B, and a 4-byte data identification code ID indicating the attribute of the data is attached at the beginning. For example, 1 byte character type data c and 4 bytes integer type data i are stored.

That is, in the case of computer A, since its memory boundary characteristics are in units of 4 bytes,
1 from the 4th byte after the data identification code id of te
When byte character type data c is stored, 3 bytes
4b from the 8th byte after the free area for
Integer type data i of yte is stored. As a result, on computer A, by determining the attribute and data length of the data from the data identification code id, specifying the 4th byte address reads character data c, and specifying the 8th byte address. In this case, integer type data i will be read. On the other hand, in the case of computer B, its memory boundary characteristics are in 1-byte units, so after the 4-byte data identification code ID, the 4-byte
When 1 byte of character type data c is stored from e-th,
Subsequently, 4 bytes of integer type data i are stored from the 5th byte. As a result, on computer B, by determining the attribute and data length of the data from the data identification code id in the same way as above, specifying the address of the 4th byte reads character data c, and specifying the address of the 5th byte. Then, integer type data i will be read out.

[0005] Here, in the computers A and B, when the data stored in the main memories 20 and 21 is stored as file data on a magnetic disk or the like,
The stored image in the main memories 20 and 21 will be stored in the same data arrangement.

[0006]

[Problems to be Solved by the Invention] As described above, computers A and B of different models having different memory boundary characteristics have different allocation formats for stored data in the main memory or disk memory. Even if an attempt is made to use the file data as is, it is not possible to correctly specify the start address of the desired data and read it out.

That is, when computer A reads the first data identification code ID from the file data of computer B, it first recognizes that there are 1 byte of character type data c and 4 bytes of integer type data i. , Next, after specifying the address of the 4th byte and reading the character type data c, in order to read the above integer type data i, 8 bytes are read according to the memory boundary characteristics of each 4 byte.
Even if the e-th address is specified, the beginning of the integer type data i is not located in that address area, so it is not possible to read and use the file data of the other computer as it is between different types of computers A and B. Can not.

[0008] The present invention has been made in view of the above problems.
To provide a method for sharing binary format data files between computers of different types, which makes it possible to accurately read file data of a partner computer and effectively utilize processed data between computers having different memory boundary characteristics. With the goal.

[0009]

[Means for Solving the Problems] That is, the method of sharing binary data files between different types of computers according to the present invention is to share binary data files stored in the main memory of each different type of computer in a manner unique to each model. A data file means that is sequentially packed and stored regardless of memory boundary characteristics, and when reading file data stored by this data file means, a write address for the main memory is set based on the memory boundary characteristics unique to the model to be read. and file data loading means for loading the file data stored by the data file means into the main memory in accordance with the write address set by the address setting means.

0010

[Operation] In other words, in each of the different types of computers, file data stored in the main memory is sequentially packed and read out regardless of memory boundaries, and this shared data is created in advance as shared data. When read and written to main memory,
By assigning addresses in accordance with the memory boundary characteristics of the relevant model, file data from different models of computers can be read accurately.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a conceptual diagram for explaining a method of sharing binary data files between different types of computers. In the figure, 10 is the main memory of computer A having a 4-byte memory boundary characteristic, and 11 is the main memory of computer A. 1by
The main memory of computer B has a memory boundary characteristic of te. Main memory 10, 1 of this computer A, B
1 contains, for example, 1-byte character type data c, 4b, each with a 4-byte data identification code ID at the beginning.
yte integer type data i, 3 byte character type data a
rray, 4-byte integer type data d is stored. Here, the data identification code id is 1 byte.
, 2 bytes, and 4 bytes.The following data attributes c (character type), i (integer type), array (character type), d (integer type) and their data. It shows the length.

When creating a data file corresponding to the main memory 10 or 11, in either case,
The data stored in the main memory is sequentially read out and stored in the shared file memory 12, with empty areas closed, regardless of the memory boundary characteristics specific to each model.

On the other hand, computer A and computer B
are provided with hardware characteristic description files 13A and 13B each describing memory boundary characteristics N unique to each model. In this case, the hardware characteristic description file 13A of computer A contains its memory boundary characteristic N=4 (
Byte) is set, and the memory boundary characteristic N=1 (byte) is set in the hardware characteristic description file 13B of computer B.

Here, when the file data stored in the shared file memory 12 is read to the main memory 10 of the computer A for use, the write address of the file data to the main memory 10 is set to the computer A. It is set based on the memory boundary characteristic N=4 of the provided hardware characteristic description file 13A.

In other words, if the address specified for writing to the main memory 10 is WP, and the following formula 1 including WP holds true, data is written to the address indicated by the WP, and if the following formula 1 does not hold, then calculates a new write designated address WP using the following equation 2 and writes data. WP mod N=0...Formula 1WP=WP+N
-(WP mod N) ...Formula 2, where mod
means an operation that calculates the remainder.

In other words, if the above formula 1 holds true and WP is divisible by the memory boundary value N (=4), it corresponds to the memory boundary characteristics of every 4 bytes, so the WP is designated as the write address. do. In addition, if the above formula 1 does not hold and the remainder X (=WP
mod N) exists, by adding WP, which is the value obtained by subtracting the remaining X from the memory boundary value N (=4), as a new write specified address, the 4-byte
This corresponds to the memory boundary characteristics of each memory.

FIG. 2 is a flowchart showing the shared data loading process when reading the file data stored in the shared file memory 12 to the main memory of a desired computer. I will explain when to issue it.

First, the start address of the data area secured on the main memory is set as the write designated address WP for the main memory 10, and data stored in the shared file memory 12 is read (steps S1 to S3).
. Then, the data identification code ID located at the beginning of the read data is transferred to the write designated address WP of the main memory 10.
(=0), and since the data length of the data identification code id is specified as 4 bytes, "4" is added to the write designated address WP (WP=4) (steps S4 to S6). Here, based on the data identification code id written in the main memory 10, the content of the data (data attribute and data length) read out subsequent to the data identification code id is determined (step S7). In this case, the first data following the data identification code id is character type data c
Therefore, through step S8, the character type data c is written to the specified write address WP (=4) of the main memory 10, and the specified write address WP is updated according to the write data length (1 byte). (WP=5)(
Steps S9, S10). Then, the process proceeds to step S16, and if there is read data from the shared file memory 12 following the write data c, the process proceeds to step S16.
7, the next data content to be written into the main memory 10 is determined. Here, since the read data following the character type data c is integer type data i, the process advances to step S11 via step S8.

Then, first, based on the above formula 1, the current designated write address WP (=5) for the main memory 10 is
is divisible by the memory boundary value N (=4) set in the hardware characteristic description file 13A of this computer A, that is, whether the remainder X obtained by the remainder calculation in step S11 is "0" or not. Determine (step S12)
. In this case, since X=1 and it does not correspond to the memory boundary characteristics of every 4 bytes, the process advances to step S13.
New write specified address WP(8) based on the above formula 2
Find "=WP(5)+N(4)-X(1)". Then, the integer type data i is written to the specified write address WP (=8) of the main memory 10, and the specified write address WP is written according to the write data length (4 bytes).
(WP=12) (steps S14, S15)
. Then, the process proceeds to step S16, and if there is read data from the shared file memory 12 following the write data i, the process returns to step S7, and the main memory 12
Determine the next data content to be written to. Here, the read data following the above integer type data i is character type data ar
ray, the process advances to step S9 via step S8.

The above character type data array is stored in main memory 1.
The data is written to the specified write address WP (=12) of 0, and the write specified address WP is updated according to the write data length (3 bytes) (WP=15) (steps S9, S10). The process then proceeds to step S16, where the shared file memory 12 following the write data array is
If there is data to be read from the main memory 10, the process returns to step S7 to determine the next data content to be written into the main memory 10. Here, since the read data following the character type data array is integer type data d, the process returns to step S11 via step S8.

Then, as above, first, based on equation 1, the current write designated address WP (=1
5) is divisible by the memory boundary value N (=4) set in the hardware characteristic description file 13A of this computer A, that is, whether the remainder X obtained by the remainder calculation in step S11 is "0". Determine whether or not (step S1
2). In this case, X=3, which means that it does not correspond to the memory boundary characteristics of every 4 bytes, so the process proceeds to step S13, and a new write designated address WP (
16) Find "=WP(15)+N(4)-X(3)". Then, the above-mentioned integer type data d is written to the specified write address WP (=16) of the main memory 10, and the above-mentioned write specified address WP is updated according to the write data length (4 bytes) (WP=20) (step S14
, S15). Then, the process proceeds to step S16, and if there is no read data from the shared file memory 12 following the write data d, the process proceeds to steps S11 to S1.
4 byte in the same way as the address setting process in 4.
Write specified address WP corresponding to memory boundary characteristics for each
After setting , it enters a waiting state for reading new file data.

Therefore, according to the method of sharing binary data files between different types of computers as described above, the file data of computer B stored in the shared file memory 12 is made to match the memory boundary characteristics of computer A. data files can be correctly read into the main memory 10, and data files created in different types of computers can be used effectively.

[0024] The method for sharing binary format data files between different types of computers in the above embodiment is as follows:
This can be realized by simply adding simple arithmetic processing in addition to normal data reading and writing processing between memories, and there is no need to create a complicated system conversion program.

[0025]

As described above, according to the present invention, there is provided a data file means for sequentially packing and storing binary format data stored in the main memory of each of different types of computers, regardless of the memory boundary characteristics specific to each model. and address setting means for setting a write address to the main memory based on memory boundary characteristics specific to the model to be read when reading file data stored by this data file means; and file data loading means for loading the file data stored by the data file means into the main memory according to the write address, so that the file data of the other computer can be accurately read between computers having different memory boundary characteristics. The processed data can be read out and used effectively.

[Brief explanation of the drawing]

FIG. 1 is a conceptual diagram for explaining a method for sharing binary format data files between different types of computers according to an embodiment of the present invention.

FIG. 2 is a flowchart showing shared data loading processing when reading file data stored in a shared file memory to a desired computer in the above embodiment.

[Figure 3] Computers A and B with different memory boundaries
FIG. 3 is a diagram showing the storage state of binary format data in each main memory.

FIG. 4 is a diagram showing data contents stored in the main memories of the computers A and B.

[Explanation of symbols]

10... Main memory of computer A, 11... Main memory of computer B, 12... Shared file memory, 13A, 1
3B...Hardware characteristics description file.

Claims (1)

[Claims] [Claim 1] Data file means for sequentially packing and storing binary format data stored in the main memory of each of different types of computers, regardless of the memory boundary characteristics specific to each model; and When reading file data, an address setting means sets a write address for the main memory based on the memory boundary characteristics specific to the model to be read, and the data file means Binary format data between heterogeneous computers, characterized in that the data is shared between heterogeneous computers having different memory boundary characteristics, comprising a file data loading means for loading stored file data into main memory. How to share files.