JPH0660205A - Data driving type information processor - Google Patents

Abstract

PURPOSE:To provide the data driving type information processor which can improve reliability of the processing, in the case of processing a data flow program containing a selection structure for selecting an instruction group to be executed under a prescribed condition. CONSTITUTION:By containing a data pair generating part 2 provided with an instruction executing part 21 for executing instructions SW and SWC for selecting the instruction to be executed in the next time, and allowing only the data pair generating part 2 to have the execution of the instruction SW and SWC, a function being equivalent to a function realized in two parts of an arithmetic processing part 3 and an external program storage part 4 up to the present is realized in one part, therefore, in the case a result of operation from the arithmetic processing part 3 receives a second rank instruction from the external program storage part 4 in a program storage part 1, it is secured to receive correctly a second rank instruction.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data driven type information processing apparatus, and more particularly to a data driven type information processing apparatus for processing a data flow program including a selection structure.

[0002]

2. Description of the Related Art FIG. 2 is a block diagram of a conventional data driven type information processing apparatus.

FIG. 3 is a field configuration diagram of a data packet processed in a data driven type information processing apparatus applied to a conventional example and an embodiment of the present invention.

FIG. 4 is a diagram showing a part of the stored contents of the program storage unit applied to the prior art and one embodiment of the present invention.

In FIG. 3, the data packet includes a destination field F1, a command field F2, a data 1 field F3 and a data 2 field F4. Destination information is stored in the field F1, command information is stored in the field F2, and various data including operand data is stored in the field F3 or F4.

The information processing apparatus of FIG. 2 includes a program storage unit 1, a data pair generation unit 5, an arithmetic processing unit 6 and an external program storage unit 7.

The data flow program shown in FIG. 4 is stored in the program storage unit 1. The data flow program shown in FIG. 4 includes a plurality of next destination information, a plurality of next instruction information, and a plurality of copy presence / absence information. When the program storage unit 1 receives the data packet shown in FIG. 3, the program storage unit 1 receives the data packet shown in FIG.
Then, the next program data, that is, the next destination information, the next instruction information, and the copy presence / absence information are read out. The read destination information and command information are stored in fields F1 and F2 of the input data packet, and the input packet is sent out. The program data stored at the next address may be read based on the copy presence / absence information read at this time, which will be described later in detail.

The data pair generation unit 5 includes a program storage unit 1
Waits for the data packet output from. That is, when two different data packets having the same destination information are detected, the operand data of one of the two data packets, for example, the content of the field F3 in FIG. 3 is stored in the field F4 of the other data packet, The other data packet is output. The data packet output from the data pair generation unit 5 is sent to the arithmetic processing unit 6 and the external program storage unit 7. The arithmetic processing unit 6 inputs the data packet output from the data pair generating unit 5, performs a predetermined arithmetic processing on the input data packet, stores the result in the field F3 or F4 of the input packet, and stores the result. The input packet is output to the program storage unit 1. The arithmetic processing unit 6 further includes a SW and a SWC instruction execution unit 61, and details of this instruction execution unit will be described later.

The external program storage unit 7 is provided to virtually expand the storage capacity of the program storage unit 1, and stores a data flow program as shown in FIG. 4 in advance. The program stored here includes program data stored in the storage unit 1. The external program storage unit 7 operates to receive the data packet output from the data pair generation unit 5, read the instruction to be executed next and download it to the program storage unit 1. More specifically, the external program storage unit 7 addresses the next destination information, the next instruction information, and the copy presence / absence as shown in FIG. 4 by addressing based on the destination information stored in the field F1 of the input packet. No information is read, the read information is stored in the data field F3 or F4 of the input data packet, and the input packet is stored in the program storage unit 1.
Send to. Incidentally, when the copy presence / absence information read at this time is the copy “present”, the program data stored at the next address is also downloaded to the program storage unit 1 similarly to the reading. The program storage unit 1 stores the field F3 of the input packet in the addressing area based on the destination information of the data packet input from the storage unit 7.
Alternatively, the next program data stored in F4 is written.

When the program is read from the external program storage unit 7, it is confirmed whether or not the program data stored in the addressing area based on the destination information is already downloaded to the program storage unit 1, and if so, the instruction is read. If not, the program data is downloaded to the program storage unit 1 as described above. Therefore, since the program storage unit 1 is a memory having a relatively small capacity (high-speed access), only the program data to be executed is downloaded and its effective use is achieved. The external program storage unit 7
Has an SW and SWC instruction execution unit 71 and operates to operate the destination information of the input data packet to select the program data to be downloaded, the details of which will be described later.

As described above, the data packet continues to rotate with the program storage unit 1, the data pair generation unit 5, the arithmetic processing unit 6, and the program storage unit 1, and at the same time, the program storage unit 1, the data pair generation unit 5, and the external program storage unit. By continuing to rotate with the unit 7 and the program storage unit 1, the calculation based on the data flow program stored in the program storage unit 1 proceeds. It should be noted that, of the two packets that are simultaneously output from the data pair generation unit 5, the circulation speed of the packet that passes through the storage unit 7 is faster than that that that passes through the arithmetic processing unit 6.

The program storage unit 1 and the data pair generation unit 5 are connected by the data transmission lines 10a and 10b, and the data pair generation unit 5 and the external program storage unit 7 are connected by the data transmission line 11. Further, the data pair generation unit 5 and the arithmetic processing unit 6 are connected by the data transmission line 13. The external program storage unit 7 and the program storage unit 1 are connected by a data transmission line 12, and the arithmetic processing unit 6 and the program storage unit 1 are connected by a data transmission line 14.

FIG. 5 is a flow graph showing a data flow program including copy processing.

In the data flow program to be executed, when data copying is required as shown in FIG. 5, that is, a certain operation result, for example, another operation in which the output of the addition instruction (+) in FIG. For example, when referred to by the multiply (*) and subtract (-) instructions in FIG. 5, the copy process is performed in the program storage unit 1. As shown in FIG. 4, copy presence / absence information is stored in the program storage unit 1. First, the information stored in the addressed area is read based on the destination information of the input data packet. At this time,
If the copy presence / absence information is "absent", the read destination information and instruction information are stored in the fields F1 and F2 of the input packet, the input packet is output, and the process is terminated. On the other hand, if the copy presence / absence information is “present”, the read destination information and instruction information are stored in the fields F1 and F2 of the input packet, and the input packet is output and stored at the next address. Destination information, instruction information, and copy presence / absence information are read. If the copy presence / absence information read out subsequently is "absent", the contents of the data field of the input data packet are stored in the data field F3 and the next destination information currently read in the destination field and the instruction field. And the data packet in which the command information is stored is output, and the process ends. If the copy presence / absence information read out subsequently is “present”, the copy process is further similarly performed.

FIG. 6 is a diagram showing a flow graph representing an example of a data flow program including a selection structure.

In FIG. 6, N1 to N8, N4 'and N6' are nodes. D1-D8, D4 'and D
6'is a node number of each node, that is, destination information. op1 to op6, op4 ', op6', SW and SWC are operation codes of each node, that is, instruction information. However, op2 and op3 are condition determination commands, and the condition determination result according to the input data is "true".
/ Output as "false". SWC and SW are commands that have the function of deriving the left input data as it is to the left output when the right input data is “true”, and deriving the left input data as it is to the right output when it is “false”. is there. Therefore, the flow graph of FIG. 6 follows nodes N4 and N4 ′ when the output of node N2 is “true”, according to the output of node N3 when “false”, and node N5 when the output of node N3 is “true”. Represents a structure for selectively executing nodes N6 and N6 'when "false".

FIG. 7 is a diagram showing the stored contents of the program storage unit 1 corresponding to the data flow program represented by the data flow graph of FIG. The flow of processing according to the selection structure of the data flow program will be described with reference to FIGS. 2, 6 and 7.

First, it is assumed that a data packet corresponding to the calculation result of the node N1 is output from the calculation processing unit 3 and input to the program storage unit 1 through the data transmission line 14.

In the program storage unit 1, the stored contents shown in FIG. 7 are read out by addressing based on the destination information D1 included in the input data packet, and the destination information D7, the instruction information SWC and the copy presence / absence information are read. "Nothing" is obtained. The read destination information D7 and instruction information SWC are stored in fields F1 and F2 of the input data packet, respectively, and the input packet is provided to data pair generation unit 5. Similarly, the destination information and the instruction information of the data packet having “true” / “false” as data according to the condition determination result of the node N2 are updated in the program storage unit 1, and the destination information D7 and the instruction information S.
After the data packet having the WC is generated, it is given to the data pair generation unit 5. As a result, the data pair generation unit 2 generates a data pair for the node N7, and the data packet storing the data pair is given to the arithmetic processing unit 6 and the external program storage unit 7. If the condition determination result of the node N2 is “true”, the instruction execution unit 61 of the arithmetic processing unit 6 controls the left input data packet corresponding to the node N7, that is, the SWC to pass through the arithmetic processing unit 6 as it is. The passed data packet is input to the program storage unit 1.

The instruction execution unit 7 of the external data storage unit 7
In 1 as well, since the condition determination result of the node N2 is "true", it is confirmed whether or not the instruction corresponding to the node N7 has already been downloaded to the program storage unit 1, and if so, nothing is output, If not, control is performed so that the instruction corresponding to the node N7 is downloaded to the program storage unit 1 by the address designation based on the destination information D7 of the input packet.

Thus, the node N7, that is, the instruction S
WC is executed by the arithmetic processing unit 6 and the external program storage unit 7.
It will be done in parallel in.

Operation processing unit 6 for program storage unit 1
Since the input data packet from has destination information D7, by address designation based on this destination information D7 in the program storage unit 1, as shown in FIG. Op4 and copy presence / absence information "present" is read. In this case, the program storage unit 1 outputs the next data packet having the destination information D4 and the instruction information op4, and since the read copy presence / absence information is “present”, the destination information D7 specifies it. By designating the address next to the address, the copy information, that is, the next destination information D4 ', the next instruction information op4', and the next copy presence / absence information "absence" is read out and the copy processing is performed. Therefore, the destination information D
4 and the command information op4, and the data packet having the destination information D4 'and the command information op4' are continuously output from the program storage unit 1. This means that the left input data packet of the node N7 is selectively derived to the left output of the node N7 on the data flow program.

On the contrary, the condition judgment result of the node N2 is "false".
If so, the destination data of the input data packet corresponding to the node N7, that is, the SWC is added to the destination information by the SWC in the instruction execution unit 61 of the arithmetic processing unit 6 and the value 2 is added to "D7 +".
After being updated to 2 ″, it is output from the arithmetic processing unit 6 and input to the program storage unit 1.

Also, in the instruction execution unit 71 of the external program storage unit 7, when the condition determination result of the node N2 is "false", the destination information is updated to "D7 + 2" and then to "D7 + 2". Instruction is program memory 1
Check if it has been downloaded to. If it is, nothing is output. If it is not downloaded, "D7 +
The program data read by the address designation based on 2 "is loaded into the program storage unit 1.

Operation processing unit 6 for program storage unit 1
Address information based on the destination information D7 + 2 of the data packet from the next destination information D as shown in FIG.
8 and the data packet having the command information SW are output. This means that the left input data packet of node N7 is selectively derived to the right output of node N7 on the data flow program.

As described above, when the condition determination result of the node N2 is "false", the destination information D8 and the command information S
A data packet having W is given to the data pair generation unit 5.

On the other hand, a data packet having “true” / “false” as data according to the result of the condition determination of the node N3 is also input to the program storage unit 1. The program storage unit 1 reads out the next destination information D8, the next instruction information SW, and the copy presence / absence information “absence” as shown in FIG. 7 by addressing based on the destination information of the input packet. The read destination information D8 and command information SW are stored in fields F1 and F2 of the input data packet, respectively, and the input packet is given to data pair generation unit 5. Therefore, the data pair generation unit 5 generates a data pair for the node N8 and supplies the data pair to the arithmetic processing unit 6 and the external program storage unit 7 in parallel.

If the condition judgment result of the node N3 is "true"
If so, the instruction execution unit 61 of the arithmetic processing unit 6 determines that the node N
8, that is, the left input data packet corresponding to SW is directly output from the arithmetic processing unit 3 and given to the program storage unit 1.

Also, the instruction execution unit 71 of the external program storage unit 7 also confirms whether the instruction corresponding to the node N8 has been downloaded to the program storage unit 1 because the condition determination result of the node N3 is "true". If not, nothing is output, and if not, the program storage unit 1
, The instruction corresponding to node N8 is downloaded.

On the other hand, since the data packet input from the arithmetic processing unit 6 to the program storage unit 1 has the destination information D8, this destination information D8 is stored in the program storage unit 1.
The next destination information D5, the instruction information op5, and the copy presence / absence information "absence" are read out by the address designation based on.
The read destination information D5 and the instruction information op5 are stored in the fields F1 and F2 of the input data packet from the arithmetic processing unit 6, respectively, and the input data packet is sent to the data pair generation unit 5. Therefore, only the data packet having the destination information D5 and the instruction information op5 is output.

This means that the left input data packet of the node N8 is selectively derived to the left output of the node N8 on the data flow program.

On the contrary, the condition judgment result of the node N3 is "false".
Then, in the instruction execution unit 61 of the arithmetic processing unit 6,
For the destination information of the input data packet corresponding to the node N8, that is, SW, 1 is added by the instruction SW, and “D8 +
After being updated to 1 ″, the input packet is output from the arithmetic processing unit 6 and input to the program storage unit 1.

Also, in the instruction execution unit 71 of the external program storage unit 7, the condition determination result of the node N3 is "false".
Therefore, after the destination information is updated to D8 + 1, it is confirmed whether or not the instruction corresponding to D8 + 1 is downloaded to the program storage unit 1. If it is, nothing is output, and if not, the program is stored. The instruction corresponding to D8 + 1 is loaded into the copy 1.

In the program storage unit 1, D8 + is included in the destination information of the data packet input from the arithmetic processing unit 3.
Since 1 is stored, a data packet storing the next destination information D6 and the instruction information op6 is output by the program storage unit 1 by addressing based on the destination information D8 + 1. In this case, as can be seen from FIG. 7, the read copy “present” / “absent” information is “present”, so the address next to the address designated by D8 + 1 is designated and the copy processing is performed. Done. Therefore, the next destination information D6 'and the second instruction information op6'
And the next copy presence / absence information "absence" is read out, a data packet in which D6 'is stored in the destination information, op6' is stored in the command information, and the contents of the data field of the input data packet is stored in the data field is also generated and output. To be done.

This means that the left input data packet of the node N8 is selectively derived to the right output of the node N8 on the data flow program.

As described above, in the conventional selective execution of the data flow program, the instruction SW, the instruction SWC, or the combination of these two instructions selectively switches the destination of the left input data according to the true / false value of the right input data. It is realized by the combination. The instructions SW and SWC are stored in the arithmetic processing unit 6 (instruction executing unit 61) and the external program storage unit 7.
By being executed in parallel by the (instruction execution unit 71), the program storage unit 1 correctly executes the waiting of the operation result and the next instruction.

[0037]

In the conventional data driven type information processing apparatus, when the processing of the data flow program including the selection structure is realized, the instructions SW and SWC are inside the arithmetic processing section 6 (instruction executing section 61). Also, the data having the same content is executed in parallel inside the external program storage unit 7 (instruction execution unit 71). Therefore, the content of update processing of the destination information according to the condition determination result regarding the instruction SW or the instruction SWC inside the arithmetic processing unit 6 and the content of update processing of destination information according to the condition determination result regarding the instruction SW or the instruction SWC inside the external program storage unit 7 Only when the two coincide with each other, the waiting of the operation result and the next instruction is correctly executed in the program storage unit 1. However, if the destination information update processing result no longer matches that of the instruction execution unit 61 due to some error in the instruction execution unit 71, the next instruction cannot be correctly downloaded to the program storage unit 1 and the data Flow program processing cannot be executed correctly.

As described above, since the same instruction (SW, SWC) is executed at a plurality of places in the same information processing apparatus, the probability that the instruction will be erroneously executed becomes extremely high, which in turn results in the information processing. There is a problem that the reliability of processing in the device is reduced.

Therefore, it is an object of the present invention to improve the reliability of the processing when processing a data flow program including a selection structure for selecting an instruction group to be executed according to a predetermined condition. It is to provide an information processing device.

[0040]

A data driven information processing apparatus according to the present invention is an apparatus for processing information using a data packet consisting of a destination field, a command field, a first data field and a second data field. Yes, it comprises a program storage means, a program loading means, a computing means, and a data pair generating means.

The program storage means stores a first data flow program consisting of a plurality of destination information and a plurality of instruction information, inputs the data packet, and performs addressing based on the contents of the destination field. The next destination information and the next instruction information of the first data flow program are read out, and the respective pieces of information are stored in the destination field and the instruction field of the input data packet, and the input packet is output.

The program loading means stores a second data flow program including the first data flow program stored in the program storage means, inputs a data packet, and performs addressing based on the contents of the destination field. Read out the next destination information and the next instruction information of the second data flow program, and load this into the addressing area of the program storage means based on the contents of the destination field of the input packet.

The arithmetic means inputs the data packet, decodes the instruction information in the instruction field of the input packet, and outputs the first and second data fields of this input packet to 2 bits.
Predetermined arithmetic processing is performed on one data, the result is stored in the first data field of the input packet, and the input packet is output to the program storage means.

The data pair generation means waits for the data packet output from the program storage means. More specifically, the data packet output from the program storage means is input, and the destination information is different.
The first of one of the two data packets
Has a function of storing the data in the data field of the other data packet in the second data field of the other data packet and outputting the other data packet in parallel to the program loading means and the arithmetic means. In particular, when executing a data flow program including a selection structure for selecting an instruction group to be executed according to a predetermined condition in this information processing device, the data pair generation means uses the destination information of the input data packet as the above-mentioned predetermined information. It further includes destination operation means that operates based on the condition to selectively specify the destination of this data packet.

[0045]

When the data driven type information processing apparatus according to the present invention is configured as described above and executes a data flow program including a selection structure for selecting an instruction group to be executed according to a predetermined condition in the information processing apparatus. The destination information of each data packet given in parallel from the data pair generating means to the program loading means and the computing means is
The destination operation means of the data pair generation means operates in the same manner on the basis of the above-described predetermined condition to select and specify in advance.
The program loading means inputs this packet and loads the information read by addressing based on the destination information of the input packet into the addressing area of the program storing means based on the destination information of the input packet. The calculation means is
The packet output from the data pair generating means is input, arithmetic processing is performed, and the input packet storing the arithmetic result is output to the program storage means. In the program storage means, the information stored in the addressing area based on the destination information of the data packet from the arithmetic processing means, that is, from the program loading means based on the destination information of the data packet output from the data detection means. The information loaded in the program storage means is read. Therefore,
The data packet storing the calculation result output from the calculation processing means in the program storage means can correctly receive the next destination information and command information.

[0046]

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

FIG. 1 is a block diagram of a data driven type information processing apparatus according to an embodiment of the present invention.

In FIG. 1, the information processing apparatus includes a program storage unit 1, a data pair generation unit 2, an arithmetic processing unit 3 and an external program storage unit 4.

The program storage unit 1 has the same function as a conventional program storage unit. The data pair generation unit 2 further includes a SW and SWC instruction execution unit 21. Instruction execution unit 2
In the generation unit 2, a data pair is detected, and the command information of the detected data packet is command SW or command SWC.
, The destination information of the other data packet is updated according to the condition determination result (true / false) stored in the data field of the one data packet, and then the other packet is stored in the arithmetic processing unit 3 and the external program storage. It operates so as to output to the unit 4. The other operations of the data pair generation unit 2 are the same as those of the conventional data pair generation unit described above.

The arithmetic processing unit 3 does not include the SW and SWC instruction executing units included in the conventional arithmetic processing unit. The arithmetic processing unit 3 performs a predetermined arithmetic processing on the input data packet and outputs the result to the program storage unit 1.
To output to. However, when the command information stored in the input data packet is the command SW or the command SWC, the arithmetic processing unit 3 operates to output the input packet as it is to the program storage unit 1.

The external program storage unit 4 does not include the SW and SWC instruction execution unit included in the conventional external program storage unit. The external program storage unit 4 inputs the data packet output from the data pair generation unit 2, reads the instruction to be executed next, which is stored in the designated address area based on the destination information of the input packet, if necessary, It operates to download to the program storage unit 1. At the time of this download, if the instruction is already downloaded to the program storage unit 1, this download is not performed.

As described above, as can be seen by comparing the information processing apparatus according to the present embodiment shown in FIG. 1 with the conventional information processing apparatus shown in FIG. The SW and SWC instruction execution units provided in
Since the functions are integrated and provided in one place in the device, the size reduction of the information processing device is promoted more than before.

Next, the operation when the data flow program including the selection structure according to the data flow graph shown in FIG. 6 is processed in the information processing apparatus shown in FIG. 1 will be described.

First, it is assumed that the data packet corresponding to the calculation result in the node N1 in FIG. 6 is output from the calculation processing unit 3 and input to the program storage unit 1 through the data transmission line 14.

The arithmetic processing unit 3 in the program storage unit 1
Information D stored in the data packet input from
By the address designation based on 1, the storage contents shown in FIG. 7 are read out, and the destination information D7, the instruction information SWC and the copy existence / non-existence information “absence” are obtained. Therefore, the data pair generator 2 is provided with a data packet storing the destination information D7 and the instruction information SWC.

On the other hand, the data packet storing "true" / "false" as data according to the condition determination result of the node N2 has the destination information and the instruction information updated in the program storage unit 1, and has the destination information D7 and the instruction information SWC. After the data packet is generated, it is given to the data pair generation unit 2.

The data pair generator 2 detects a data pair for the node N7 shown in FIG. If the condition determination result of the node N2 is "true", the node N2
7, that is, the left input data packet corresponding to SWC is output from the data pair generation unit 2 as it is, and the arithmetic processing unit 3
And are provided in parallel to the external program storage unit 4.

The arithmetic processing unit 3 inputs the data packet given from the data pair generating unit 2, and when the command information stored therein is the command SWC, the input data packet is sent as it is, It is given to the program storage unit 1.

Further, the external program storage unit 4 stores the command information stored in the input data packet, that is, the node N.
It is confirmed whether the instruction corresponding to 7 is downloaded to the program storage unit 1, and if it is, nothing is output. If not, the node N7 is sent to the program storage unit 1.
The instruction corresponding to is loaded as usual.

In the program storage unit 1, since the input data packet from the arithmetic processing unit 3 has the destination information D7, this destination information D7 is stored in the program storage unit 1.
7, the next destination information D4, the instruction information op4, and the copy presence / absence information "present" are read out by the address designation based on the. In this case, a copy process similar to the conventional one is performed. Therefore, from the program storage unit 1, the data packet storing the destination information D4 and the instruction information op4 and the data packet storing the destination information D4 'and the instruction information op4' are continuously arranged.
Sent to. This means that the left input data packet of the node N7 shown in FIG. 6 is selectively derived to the left output of the node N7 on the data flow program.

On the contrary, the condition judgment result of the node N2 is "false".
If so, the left input data packet corresponding to the node N7, that is, the SWC is the instruction execution unit 21 of the data pair generation unit 2.
At 2, the destination information is output by the data pair generation unit 2 after the value 2 is added by the instruction SWC and updated to “D7 + 2”. The data packet output from the generation unit 2 is given to the arithmetic processing unit 3 and the external program storage unit 4 in parallel. The arithmetic processing unit 3 inputs the data packet output from the data pair generation unit 2, and outputs the input data packet as it is to the program storage unit 1 when the command information stored therein is the command SWC. . On the other hand, the external program storage unit 4 stores the command information D7 + 2 stored in the data packet input from the data pair generation unit 2.
Whether the instruction corresponding to is downloaded to the program storage unit 1, outputs nothing if it is, and if not, the next program data stored in the addressing area based on the destination information D7 + 2. Is read and loaded into the program storage unit 1.

Operation processing section 3 for program storage section 1
Stores the destination information D7 + 2.
The program storage unit 1 reads out the next destination information D8, the instruction information SW, and the copy presence / absence information “absence” as shown in FIG. 7 by addressing based on the destination information D7 + 2 stored in the input packet. Therefore, the data packet storing the destination information D8 and the command information SW is sent from the program storage unit 1 to the data pair generation unit 2. This means that the left input data packet of the node N7 is selectively derived to the right output of the node N7 on the data flow program.

Similarly, regarding the program processing after the node N7, the update processing of the destination information when selecting the instruction group according to the instruction SWC or the instruction SW2 is performed by the data pair generation unit 2
This is performed only in the instruction execution unit 21.

As described above, in the data driven type information processing apparatus of FIG. 1, the instruction S for selecting an instruction to be executed is executed.
Since the processing of W or the instruction SWC is performed only in the data pair generation unit 2, a function equivalent to the function conventionally realized in two places of the arithmetic processing unit and the external program storage unit is provided in the information processing device. Can be realized in one place,
In the program storage unit 1, the reliability of the operation in which the calculation result of the calculation processing unit 3 receives the next program data is improved.

[0065]

As described above, according to the present invention, the data pair generating means has the function of selecting an instruction group to be executed according to a predetermined condition as the destination operating means, so that the program loading means and the arithmetic operation are always performed. Since the data packet in which the same destination information is stored is supplied to the processing means, the reliability of the operation of reading the destination information and the command information whose operation result of the arithmetic processing means is next in the program storage means becomes extremely high. The reliability of processing in the information processing device is improved.

Further, the processing for selecting the instruction group to be executed according to a predetermined condition is the data pair generation means (destination operation means).
Since it is performed only in the information processing device, the same function, which has been conventionally realized in two places of the arithmetic means and the program loading means, can be realized in one place in the information processing device. Can be realized.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of a data driven type information processing apparatus according to an embodiment of the present invention.

FIG. 2 is a block configuration diagram of a conventional data driven type information processing apparatus.

FIG. 3 is a field configuration diagram of a data packet processed in a data driven type information processing apparatus applied to the related art and one embodiment of the present invention.

FIG. 4 is a diagram showing a part of stored contents of a program storage unit applied to a conventional example and an embodiment of the present invention.

FIG. 5 is a diagram showing a flow graph of a data flow program including copy processing.

FIG. 6 is a diagram showing a flow graph representing an example of a data flow program including a selection structure.

7 is a diagram showing the stored contents of a program storage unit corresponding to the data flow program represented by the data flow graph of FIG.

[Explanation of symbols]

1 Program Storage Section 2 Data Pair Generation Section 8 Arithmetic Processing Section 9 External Program Storage Section 21 SW and SWC Command Execution Section SWC and SW Instruction for selectively switching the destination of the left input data depending on the true / false value of the right input data. In each drawing, the same reference numerals indicate the same or corresponding parts.

Claims (1)

[Claims] 1. A destination field, an instruction field, and a first field.
A data driven type information processing apparatus for processing information using a data packet composed of a data field and a second data field, storing a first data flow program composed of a plurality of destination information and a plurality of instruction information, By inputting the data packet and performing addressing based on the contents of the destination field, the next destination information and the next instruction information of the first data flow program are read out, and the respective information is input to the input data packet. And a second data flow program including the first data flow program, which stores the program in the destination field and the command field and outputs the input packet. Addressing based on the content of Therefore, program load means for reading the next-order destination information and the next-order instruction information of the second data flow program and loading them into the addressing area of the program storage means based on the contents of the destination field of the input packet, A data packet is input, the command information in the command field of the input packet is decoded, two data in the first and second data fields of the input packet are subjected to predetermined arithmetic processing, and the result is input to the data packet. Packet first
Of the two data packets having the same destination information and the arithmetic means for storing the input packet in the data field and outputting the input packet to the program storage means, and waiting for the data packet output from the program storage means. Data pair generation for storing data in the first data field of one data packet in the second data field of the other data packet and outputting the other data packet in parallel to the program loading means and the arithmetic means And a means for executing a data flow program including a selection structure for selecting an instruction group to be executed according to a predetermined condition in the information processing apparatus, the data pair generation means is The destination information of the data packet is operated by operating the destination information based on a predetermined condition. Previously it was provided with selectively further destination operating means for specifying the data driven information processor.