JPS5856040A - Data processor - Google Patents

Abstract

PURPOSE:To reduce a program area by providing in a data processor a means for storing coded data a means for arithmetic processing which corresponds to a data form according to its storage contents. CONSTITUTION:When a data form setting instruction is executed, a data form specified in the instruction is coded and outputted to an internal data bus 1 and at the same time, a write control signal 3 to a code storage circuit 2 is generated to fetch coded data form information on the bus into the code storage circuit 2. The information written in the circuit 2 is held until the data form setting instruction is executed, and then supplied, as it is, to an instruction execution control circuit 4. According to the data form information, this circuit 4 performs arithmetic processing which corresponds to a data form specified by an arithmetic processing circuit 5. Consequently, the instruction system of the data processor is made simple and regular, and simular processing parts in a program are made common to reduce the capacity of a program memory.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a data processing device having an instruction system that does not include data format designation in the instruction code.

Digital computers, including arm microcomputers, which are generally defined as information processing devices, use a data representation format based on binary numbers, but depending on the purpose of the computer, M
There are various data formats that are suitable for use in scientific and technical calculations. Computers that handle virtual data mainly handle data in floating point format. Furthermore, when processing decimal numbers, considering the efficiency of memory usage for storing lO-decimal data, it is necessary to
acked BCD (Binary Coded De
-cima l = binary coded lO base) format is advantageous, and from the viewpoint of ease of programming, it is more convenient to directly process data input from an input device such as a keyboard, and the ASCII (American
St-andard Code for Info
A method is used in which a one-digit numerical value is stored for every 7 or 8 bits, such as in the rmaNon Interchan-ge) format.

Generally, the formats of data processed by digital computers include bit, bit, unsigned binary, PackedBCD, and U.
Examples include npacked BCD, key a, 5-fi, floating point, etc.

When such a variety of data formats are handled by a single computer, it has been common practice to prepare dedicated instructions for each data format for operations such as operations on the data.

However, in this case, the number of types of instructions provided by the computer increases, and it is necessary to use these instructions appropriately when creating a program. For example, taking a microprocessor as an example, the types of addition processing are 8-bit binary numbers, 16-bit binary numbers,
A base number addition instruction, an addition correction instruction for the Packed BCD format, and an addition correction instruction for the Un-packed BCD format are provided. In addition, some general-purpose large-scale computers are equipped with instructions such as binary addition, logical addition, decimal addition, and floating point addition as types of addition instructions. Furthermore, similar types of instructions such as subtraction, multiplication, and division are provided.

Having such a wide variety of arithmetic instructions poses an obstacle for programmers to understand and memorize instructions, and from the perspective of computer architecture designers, it is difficult to allocate machine language code to individual instructions by the number of bits. This limitation caused the code system to become unorganized.

One way to solve this problem is to create an attribute field in the data itself to express the format of the data, so that the instruction code does not contain any information about the format of the data to be manipulated. There is no way.

In this case there is certainly one process (e.g. transfer).

1 each for addition, subtraction, 9 multiplication, division, comparison, etc.)
All you need to do is to provide different types of instructions (such instructions are called generic instructions), but since you have to check and process the attribute field attached to the data each time you perform an operation, the processing speed becomes slower and the data There was a drawback that the amount of memory required for storage increased.

The present invention has been made in view of the above-mentioned disadvantages.In a computer that operates on multiple types of data formats, only the basic instructions for each process can be used without preparing instructions for each data format individually. General-purpose data that allows you to operate any of multiple data formats, simplifying programming, and reducing the program memory area by standardizing processing routines for each data format. The purpose is to provide processing equipment.

In order to achieve the above objects, the data processing device of the present invention includes: a memory circuit for encoding and storing the format of data to be manipulated during program processing; a means for encoding and writing the data format into the memory circuit; and a processing circuit that performs processing according to a data format specified by information held in the circuit.

An embodiment of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a circuit diagram showing the main parts of the data processing apparatus in this embodiment.

In FIG. 1, a code storage circuit 2 having a plurality of pits connected to an internal data bus 1 stores the type of data format to be operated as a code. The code information can be written into the code storage circuit 2 using a data format setting command held in a program storage area such as a ROM (Read Only Memory) by the data processing device.

When this data format setting command is executed, the data format specified in the command is encoded and output to the internal data bus 1, and at the same time, a write control signal 3 to the code storage circuit 2 is generated. Coded data format information on the internal data bus l is taken into the code 6 self-storage circuit 2.

The data format information written in the code storage circuit 2 is held until the data format setting command is executed again, and is supplied as is to the instruction execution control circuit 4.

The instruction execution control circuit 4 controls the arithmetic processing circuit 5 to perform arithmetic processing corresponding to the specified data format according to the data format information.

For example, if the arithmetic processing circuit 5 is an arithmetic circuit capable of processing data with a width of 16 bits, if the specified data format is a signed 16-bit binary number, it will be expressed as a normal two's complement number. Control to perform binary operation, Pa
When the cked BCD format is specified, the instruction execution control circuit 4 controls the 1O base correction circuit for every 4 bits included in the arithmetic processing circuit 5 after the binary number operation. As a more complicated example, when a 32-bit floating point format is specified, the operation of the arithmetic processing circuit 5 is controlled by a fixed floating point processing microprogram included in the instruction execution control circuit 4, and the floating point format is Processing of exponent part, mantissa part, sign, etc. of decimal point data is 16-bit.

Finally, a 32-bit operation result is obtained.

FIG. 2 shows an example of the data format encoded and stored in the code storage circuit 2. As shown in FIG. In this case, the number of data formats is 16 or less, so the code storage circuit 2 is 4 bits,
This configuration is sufficient.

In this way, according to this embodiment, the data processing device is provided with means for storing coded data for distinguishing data formats, and means for performing arithmetic processing according to the data format according to the stored contents. Since there is no need to increase the number of types of instructions, the instruction system of the data processing device can be approximated to a simple and orderly format. As a result, programming becomes easier, and the instructions themselves become more versatile, independent of the data format, making it easier to perform similar processing in programs that were previously prepared for each data format. It is also effective to reduce the amount of program memory by sharing parts.

Let me present an example as an application of this embodiment.

For example, if an application program includes subprograms that perform similar processing on two or more types of data formats, general computers use different instructions even for the same arithmetic processing when the data formats are different. Although it is not easy to share subprograms, it is relatively easy to share similar subprograms because arithmetic instructions, etc. can be expressed independently of the data format in a computer using the data processing device of the present invention. .

Immediately after execution of an instruction to set a value that encodes the data format in the code d self-memory circuit 2 in FIG. The data formats to be operated on can be switched, and different data formats can be processed while using the same subroutine. In this case, the larger the amount of subprograms that can be shared, the greater the effect of reducing the program memo IJf area.

In FIG. 1, the code storage circuit 2 may be any circuit that can store logic information, such as a normal flip-flop, register, or RAM. Further, the basic data width of the data processed by the arithmetic processing circuit 5 is generally considered to be 8°, 16.32 bits, but is not particularly limited thereto.

Furthermore, the data format to be operated is not limited to the format illustrated in FIG. 2.

[Brief explanation of the drawing]

FIG. 1 is a main circuit block diagram showing an embodiment of a data processing device of the present invention, and FIG. 2 is a diagram illustrating types of data formats to be processed by the data processing device. l...Internal data bus, 2...Code h self-storage circuit. 3...Write control signal, 4...Instruction execution control circuit. No. l closed

Claims (1)

Claims: Storage means for encoding and storing the format of data to be manipulated during program processing; and control means for encoding and setting the data format in the storage means. A data processing device comprising: arithmetic processing means for performing processing corresponding to the data format based on output information from the ml storage means.