JPH01287745A - Data processor - Google Patents

Abstract

PURPOSE:To process list data of optional depth by storing data following an element address in addresses of an element memory device which are indicated by an overflow information storage device for list data including element data which is deficient in the number of items of the element address part of the element memory device. CONSTITUTION:The value of the overflow flag of a list identification address in an overflow information memory 5 is '1', addresses which are included in the item N of the element address part among addresses of element data are stored in the list identification addresses 21 in respective element memory devices 6, and information on an element is put in a value part 25. Further, addresses in an element memory stored with a part following the address part of the element data is written in the auxiliary address part 23 of the list identification address 21. Further, parts following address parts of the element data in the same element memory device are written in the element address parts of the addresses of respective element memory devices having the same value as said address value. Consequently, list data of optional depth can be processed.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a data processing device mainly intended for use in the field of artificial intelligence.

BACKGROUND OF THE INVENTION In recent years, the field of artificial intelligence has been actively researched as one of computer applications.

In this field, it is necessary to process structured data, and therefore LISP, a language that can handle structured data, is widely used. Since it is inefficient to execute the LISP language on a general-purpose computer, special-purpose machines with various improvements have been developed.

These dedicated machines were improved mainly from a linguistic perspective, and the typical improvements made are shown below.

(1) Primitive functions such as CAR and CDR are executed at the microprogram level.

(2) TA for handling generic data type 7゛
Data format with G.

(3) Provide a node-ware control stack to speed up stack processing.

(Reference 1-LISP Machine” Information Processing Vol. 1゜23
No. 0.8 pp. 752-772) However, although improvements have been made to the execution system of the language as described above, the representation of structure data inside a computer basically uses pointers to represent the order of elements and the method of connection. (hereinafter referred to as the list). Generally, a list is expressed as a binary tree, starting from a starting node and branching left and right sequentially, ending at each leaf node. There are two types of leaf nodes: atom nodes and NIL nodes.

Nodes that are not leaf nodes are list nodes that indicate that the branch is continuing. This list node is used to indirectly represent the position of leaf mates.

In pointer representation, this list structure is expressed as is, and all nodes are expressed as cells connected by addresses, which causes the following problems.

1. Accessing an element at an arbitrary position, searching for an arbitrary element within a list, and decomposing/combining a list involve going through the list, making it inefficient.

2. Matching of lists and searching for arbitrary sublists within a list involve decomposition of the list, which is inefficient.

In order to solve these problems, it is basically necessary to change the expression of list data. That is, if the position of a leaf node can be directly expressed, there is no need to have list node information. Therefore, equivalent list data can be expressed by a table in which leaf position information and leaf information are sequentially arranged. If we use a one-dimensional vector representation in which the list is numbered sequentially in the length direction and items are assigned sequentially in the depth direction as a way to express the node position of this leaf, the list data will be a vector indicating the leaf position information and a leaf node position. It is expressed in a tabular format as a set of data that includes its own information.

FIG. 4 shows an example of list data expression. This shows the graphical representation (1) and tabular representation (2) of list data that becomes (B(())D) when expressed in 8 formulas. In the diagrammatic representation, circles are Represents a list node, and the box enclosed in a square indicates a leaf node.The number string added above each node indicates the node position expressed according to the method described above.Extract this leaf part. The table representation (2) is a table representation (2) in which the element address field contains the node position vector and the value field contains the leaf elements.

When a list is expressed in a table format like this, each element can be accessed without passing through pointers, and the operations on each element can be performed in parallel, which speeds up list processing such as pattern matching. It can be done.

An example of the conventional data processing device mentioned above will be described below with reference to the drawings.

FIG. 6 shows the configuration of a conventional data processing device.

In FIG. 6, 51 is a main memory device, 62 is a main register device, 63 is a main processing unit, 64 is a main data processing section composed of the main memory device 61, a plurality of main register devices 62, and a main processing device 63; 59 is an element memory device; 66 is an element register device; 67 is an element operation device;
is the control device.

The operation of the data processing apparatus configured as described above will be explained below with reference to FIGS. 4, 6, and 6.

First, FIG. 6 is a diagram showing the relationship between the main memory device and the element memory devices.

In FIG. 6, 51 is a main memory device, 66 is an element memory device, 61 is identification information of list data, 62 is an element address field, and 63 is a value field.

List data consists of identification information and a plurality of element data. The identification information 61 of the list data is an address within the element memory and is stored in the main memory device 61. Further, each element data of a list consists of an element address part 62 and a value part 63, and a plurality of element data constituting the same list data is indicated by identification information 61 of the list data in a plurality of element memory devices 66. stored at the same address.

Similarly, the main register device 62 and each element register device 5e
also has a correspondence relationship.

When processing is required for list data, the identification information 61 of the list data is transferred from the main memory device 51 to the main register device 52, and at the same time, the same address is sent to each element processing section 58 by the control device 59, and each element processing In section 58, element data is transferred from the indicated address of element memory device 56 to element register device s6. Furthermore, at the same time as the calculation instruction to the main processing unit 53,
The control device 59 sends the same calculation command to each element processing section 68, and the plurality of element calculation devices 57 can perform calculations on a plurality of element data in parallel.

Problems to be Solved by the Invention However, in the above configuration, there is a limit to the number of items in the element address field that expresses the position of the element in the tabular list data. There is a problem in that it is not possible to process list data that includes list data, that is, list data whose depth exceeds the number of items in the element address field.

In view of the above problems, the present invention provides a data processing device that can process list data of arbitrary depth in parallel processing of tabular list data.

Means for Solving the Problems In order to solve the above problems, the data processing device of the present invention includes a main memory device that stores storage position information of elements of list data, and a main memory device that stores data stored in the main memory device. A main data processing section consisting of a main register device that transfers and stores data, a main processing device that operates on the data stored in the main register device, and an element address section that indicates the position of a node for each element of list data. and a value portion indicating a storage position of a data value; an element register device that transfers and stores the elements stored in the element memory device; and an element register device that stores the elements stored in the element register device. a control device; The system is equipped with the following functions and is capable of processing list data of arbitrary depth.

Effects of the present invention With the above-described configuration, for list data including element data for which the number of items in the element address section of the element memory device is insufficient, the continuation of the element address is added to the address of the element memory device specified by the overflow information storage device. By storing , list data of arbitrary depth can be processed.

Embodiment Hereinafter, a data processing apparatus according to an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 shows the configuration of a data processing device in an embodiment of the present invention.

In FIG. 1, 1 is a main memory device, 2 is a main register device, 3 is a main arithmetic unit, and 4 is a main memory device 1, a main register device 2. A main data processing unit composed of the main processing unit 3, 5
is an overflow information storage device, 6 is an element memory device,
7 is an element register device, 8 is an element operation device, 9 is an element memory device θ, and element register device 7. A plurality of element data processing units constituted by element calculation devices 8, 1o is a control device.

The operation of the data processing apparatus configured as described above will be explained below with reference to FIGS. 1, 2, and 3.

First, FIG. 2 shows that in the data processing apparatus of the above embodiment,
FIG. 6 is a diagram illustrating a method of storing list data that overflows in the depth direction.

In FIG. 2, 1 is a main memory device, e is an element memory device, 5 is an overflow information storage device, 21 is a list identification address, 22 is an overflow flag, 23 is an auxiliary address field, and 24 is an element address field. Let the number be N. Furthermore, 26 is a value part.

Figure 3 shows an example of depth-direction overflow list data, which is a table representation (1) of list data that becomes ((((((((λ B))))))))) when expressed using formula 8. , and element data of the list data in the data processing device of the embodiment.

If the address parts of the plurality of element data constituting the list all fall within the limit on the number of items, the value of the overflow flag 22 of the list identification address 21 in the overflow information memory 6 is 0, and each element memory device Element data is stored in the list identification address 21 in 6, and the storage method and processing method of the element data are the same as in the conventional example.

Among the element data composing the list, if there is element data whose address field has an insufficient number of items, the value of the overflow flag of the list identification address in the overflow information memory 5 is 1, and the value of the overflow flag of the list identification address in the overflow information memory 5 is 1, The list identification address 21 has addresses that fit into the number of items N in the element address part among the addresses of the element data, and element information in the value part 25. Further, in the auxiliary address section 23 of the list identification address 21 in the overflow information memory 6, an address in the element memory that stores the continuation of the address section of the element data is written.

At the address of each element memory device that is the same as this address value, the continuation of the address part of the element data of the same element memory device is written in the element address part. For example, when N is 8, the list data shown in the tabular representation of FIG. 3(1) is expressed as shown in FIG. 3(2) in the data processing apparatus of this embodiment.

Therefore, with the above configuration, list data of arbitrary depth can be processed.

Effects of the Invention As described above, the present invention comprises: a main memory device that stores storage position information of elements of list data; a main register device that transfers and stores data stored in the main memory device;
A main data processing unit consisting of a main processing unit that operates on the data stored in the main register device; an element memory device that stores the elements stored in the element memory device; an element register device that transfers and stores the elements stored in the element memory device; and an element operation device that operates the elements stored in the element register device. It is equipped with a plurality of element data processing sections configured, an overflow information storage device that stores overflow information of list data including element data whose element address section has an insufficient number of items, and a control device. Be able to process list data.

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram of a data processing device according to an embodiment of the present invention, FIG. 2 is an explanatory diagram showing a method of displaying list data overflowing in the depth direction, and FIG. FIG. 3 (2) is an explanatory diagram showing the representation in a data processing device in one embodiment of the present invention, and FIG. FIG. 5 is a data correspondence diagram showing an example of table representation, FIG. 5 is a configuration diagram of a conventional data processing device, and FIG. 6 is an explanatory diagram showing the relationship between a main memory device and an element memory device. DESCRIPTION OF SYMBOLS 1...Main memory device, 2...Main register device, 3...Main processing unit, 4...Main data processing unit, 6... ...Overflow information storage device, 6...Element memory device, 7...
Element register device, 8...Element operation device, 9.
...Element data processing unit, 1o...Control device. Name of agent: Patent attorney Toshio Nakao and 1 other person No. 4
Diagram (A (B (C)) D)

Claims (1)

[Claims] A main memory device that stores storage position information of elements of list data, a main register device that transfers and stores data stored in the main memory device, and a main operation that operates on the data stored in the main register device. an element memory device configured to store each element of list data by an element address field indicating a node position and a value field indicating a storage position of a data value; A plurality of element data processing sections each including an element register device that transfers and stores the elements stored in the memory device, and an element operation device that operates on the elements stored in the element register device, and an element address section. 1. A data processing device, comprising: an overflow information storage device that stores overflow information of list data including element data with insufficient number of items; and a control device, and is capable of processing list data of arbitrary depth.