JPH08129376A - One-chip memory device - Google Patents

Abstract

PURPOSE: To make a frame buffer for, for example, a high-speed graphic display compact by making it possible to obtain a memory circuit incorporating a circuit to execute a read modifying write operation at an arbitrary data width. CONSTITUTION: An arithmetic circuit which consists of a memory element 2 enabling the arbitrary execution of reading out, writing and storing of data and an arithmetic and logic unit 1 for calculating the first data from the outside and the second data in this memory element 2 is provided with a register 3 for storing the assigned arithmetic function codes and a register (mask circuit 4) for storing the assigned writing control data. The calculation and the writing control of bit units are executed in accordance with the output data of this arithmetic function code storage register 3 and the writing control data storage register. As a result, the assignment of the modifying calculation of read modifying writing does not depend any more on the data width of the writing data.

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a storage element, and more particularly to a storage circuit suitable as a frame buffer for a high-speed graphic display. 2. Description of the Related Art With the improvement of display resolution, graphic display devices have begun to require a large-capacity memory for storing display information, that is, a frame buffer. However, since increasing the capacity of the frame buffer leads to an increase in the number of memory accesses when displaying graphic data, it is necessary to reduce the number of memory accesses in order to speed up the display. As a means for reducing the number of times of memory access, there is a method of executing arithmetic processing inside a graphic display frame buffer. FIG. 2 shows an example of a frame buffer using this method. In FIG. 2, 1 is an arithmetic unit having a 16-bit length, 2 is a memory for storing graphic data, 3 is an arithmetic function specifying register of the arithmetic unit, 4 is a write mask circuit, and D15 to D0 are 16-bit data from the data processing device. , DO15 to DO0 are read data from the memory, FC3 to FC0 are calculation function specifying data for the arithmetic unit, M15 to M0 are write control signals to the memory, A23 to A1 are 23-bit address signals from the data processing device, and WE is data processing. A write control signal from the device, FS is a latch control signal for the arithmetic function designation register, and MS is a latch control signal for the write mask circuit. The reason why the number of memory accesses is reduced in the configuration of FIG. 2 will be described. When writing a figure on a bitmap type graphic display, the figure is expressed by a set of points, and therefore the figure is drawn by repeating the point drawing. For this reason, access to the frame buffer is performed not in units of 16 bits, but in units of 1 bit or 4 bits smaller than the data width forming the memory. Also,
Generally, when writing a point, an operation with write data is required, so that an operation with memory data and a write in bit units are required. Since ordinary memory does not have these functions, the calculation is executed inside the data processing device that performs graphic drawing processing, the data of the memory address to be written is read, the bit calculation is executed, and then the data is written to the same address. Has been realized. Therefore, even when writing 1-bit data, it is necessary to access the memory twice. In the frame buffer of FIG. 2, the arithmetic unit 1 performs the memory data and the operation of the data processing device to write the data in bit units by the write mask circuit 4, and the memory access necessary for writing the 1-bit data is the data access. It only needs to be done once with the processor. Access to memory 2 is required twice: read and write, but read / write is
Since there is an access mode called “modify / write” that realizes reading and writing in one operation, it can be realized in one operation. As described above, the frame buffer shown in FIG. 2 is effective for speeding up the graphic display, but requires a large number of circuits to be added around the memory element, so that the reliability is reduced and the cost is high. There is a problem that becomes. For the frame buffer shown in FIG. 2, for example, Nikkei Electronics 1984.8.27, “1280 × 1024 pixel graphic display frame buffer 64KRA with nibble mode”
Design with M ”(P.227 to 245). SUMMARY OF THE INVENTION An object of the present invention is to provide a memory circuit for realizing a compact frame buffer for high speed graphic display in order to solve the above problems. [0006] A storage element capable of arbitrarily reading, writing and storing data, and a computing unit for computing first data from the outside and second data in the storage element. The memory circuit is provided with a register for storing the specified arithmetic function code and a register for storing the specified write control data, and the operation and bit unit are performed based on the output data of the arithmetic function code storage register and the write control data storage register. Writing control is performed. With the above configuration, the designation of the modify operation of read modify write does not depend on the data width of the write data, so that the read modify write operation is executed with an arbitrary data width. A memory circuit having a built-in circuit can be realized, and for example, a frame buffer for high-speed graphic display can be made compact. An embodiment of the present invention will be described in detail below with reference to the drawings. First, the concept of the present invention will be described. In order to reduce the peripheral circuits of the frame buffer memory shown in FIG. 2, an IC (Integrated Memory), an arithmetic unit, an arithmetic function designation register, and a write mask circuit are integrated.
It is possible to make an integrated circuit). In the current graphic display, since a logical operation is mainly required as an arithmetic function, the arithmetic unit can be divided into bit units of arithmetic data. Even when using arithmetic operations, by adding a circuit that handles carry signals,
In principle, bit division is possible. Since the write mask circuit 4 is a circuit that performs write control in bit units, it is obvious that it can be divided in bit units.
However, the arithmetic function designation register 3 has a bit length determined by the number of arithmetic functions of the arithmetic unit 1 and is not related to the bit length of the arithmetic data (here, 16). Can not. Therefore, it is necessary to have the arithmetic function designation register 3 for each divided unit. Thus, it is useless to have the same function for each divided unit, but the degree of integration of the IC increases every year, and the number of elements used as a peripheral circuit with respect to the number of memory elements when integrated is increased. This is not a problem because the ratio of numbers is so small that it does not reach 1%. In the case of integration, having the arithmetic function designation register 3 for each division unit is not so problematic as described above, but it is necessary to divide the frame buffer shown in FIG. 2 into data bit units. Has a problem. In order to use the frame buffer shown in FIG. 2, it is necessary to write arithmetic function data in the arithmetic function designation register 3 and write mask data in the mask circuit 4 before actually performing memory access.
In the frame buffer of FIG. 2, both data use the data signals D <b> 15 to D <b> 0 from the processing device as input signals. Therefore, if the data is divided into bit units, it becomes a 1-bit signal. But,
Only two types of calculations can be specified in the calculation function specification register 3. Thus, it is a problem that the number of arithmetic functions changes depending on the bit configuration of the memory. According to the present invention, since the arithmetic function is specified by the data bus, the fact that it depends on the bit division of data occurs, and the specification is performed by using an address signal that does not depend on the bit division unlike the data bus. Is. Next, an embodiment of the present invention will be described. FIG.
Is the configuration of the frame buffer memory circuit of the embodiment. 1 is an arithmetic unit, 2 is a memory element, 3 is an arithmetic function designating register, 4 is a write mask circuit, Dj is 1 of 16 bits of the data signal of the data processing apparatus for graphic drawing.
Bit signals, A23 to A1 are address signals of the data processing device, WE is a write control signal of the data processing device, FS is a data set control signal for the arithmetic function designation register 3 and the write mask circuit 4, and DOj is read data of the memory element 2. , DIj are the operation result data of the operation unit 1, W
j is a write control signal for the memory element 2. FIG. 3 shows the configuration of the write mask circuit.
Reference numeral 41 is a write mask data storage register, and 42 is a gate for suppressing the write control signal WE. FIG. 4 shows a configuration example of a frame buffer using the memory circuit of FIG. FIG. 4 shows a 4-bit configuration to clarify the connection relationship. FIG. 5 shows an example in which the memory circuit of the embodiment is applied to a graphic display system. Reference numeral 6 denotes a data processing device, and reference numeral 7 denotes a decoding circuit for generating a set signal FS. The operation of the memory circuit according to the embodiment will be described below. In the embodiment, the memory circuit 5 has a capacity of 800000H to 8F.
It is assigned to address FFFFH. Here, H is an address indicating a hexadecimal number and in units of bytes. Decoding circuit 7 outputs set signal FS at addresses 900000H to 90001FH. The computing function of computing unit 1 is shown in FIG.
16 types. The data processing device 6 is, for example, 900
When FOFFH is written to the address 014H, the decode circuit 7 outputs the set signal FS,
Address signals A4 to A1 are set to 0101B (B is bit data). As a result, the arithmetic unit 1
As shown in the calculation function table, OR is selected as the calculation function. In the write mask circuit 4, one bit of 16-bit data 0F00H from the data processing device 6 is set in the write mask data storage register 41. One bit to be set is the same position as the bit position of the memory element. As a result, F0FFH is set as the write mask data. Next, a case where the data processing device 6 writes F3FFH at the address 800000H will be described. 8000
It is assumed that 0512H is stored in the address 00H. The memory access timing of the data processing device 6 is shown in FIG. Write access to the memory circuit 5 of the data processing device 6 is a read-modify-write operation as shown in FIG. 0512H is read to the DO bus at the timing of read-modify-write, and F3FFH is input to the D bus. At the next modify timing, the arithmetic unit 1 calculates the data of the D bus and the DO bus, and outputs the calculation result to the DI bus. In this case, the value of D bus is F3FFH and the value of DO bus is 0.
Since it is 512H, the data on the DI bus is F7FFH. This is because the arithmetic unit 1 selects the logical sum as the arithmetic function in the above-described operation. Finally, the data F7FFH of the DI bus is written at the timing of the read-modify-write write.
As for the write mask data, F0FFH is set, and as shown in FIG. 3, the bit of the mask data of 0 turns on the gate 42 and the 1 bit of the mask data turns off the gate 42.
Since it is F, only the 4 bits D11 to D8 execute the actual write operation, and the write operation does not occur in the remaining 12 bits. As a result, the data at the address 800000H becomes 0712H. As described above, since a part of the address signal is used as a control signal in this embodiment, a memory circuit for performing read / modify / write which can specify an arithmetic function regardless of the data division method is realized. can do. The only difference between the memory circuit of the embodiment and the ordinary memory IC is the set signal FS for setting the arithmetic function and the write mask data, and the number of pins of the IC is increased by only one pin. This does not pose a problem in forming an IC as it is. For example, a dynamic RAM having a 64K × 1 bit configuration does not use one pin, so that it is possible to use FS for this empty pin. It is apparent that this set signal may be realized in a timing sequence different from that of a normal memory access. For example, as shown in FIG.
It is possible to generate a set signal by the falling edge of the RAS signal and the WE signal, which cannot be obtained in the normal sequence of the RAM. Although the data width is 16 bits and the unit of division is 1 bit in this embodiment, it is obvious that either value may be a value other than the value described in this embodiment. In the embodiment, the designation of the arithmetic function and the designation of the write mask are performed at the same time. However, it is apparent that the designation may be performed separately. Further, the data width for designating the function of the arithmetic unit is also 4
It is clear that other than bits may be used. It is also apparent that the present embodiment may be applied to a memory having a built-in shift register and having a serial output. As is apparent from the above description, according to the present invention, the designation of the modify operation of read-modify-write does not depend on the data width of the write data, so that the read operation can be performed with an arbitrary data width. A memory circuit having a built-in circuit for executing a modify write operation can be realized, and for example, a frame buffer for high-speed graphic display can be made compact.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a memory circuit according to an embodiment. FIG. 2 is a block diagram showing a conventional frame buffer memory. FIG. 3 is a diagram showing a write mask circuit. FIG. 4 is a diagram for explaining a frame buffer configuration of the embodiment. FIG. 5 is a block diagram illustrating a configuration example of a graphic display system. FIG. 6 is a diagram for explaining an arithmetic function. FIG. 7 is a timing chart showing memory access timing. FIG. 8 is a timing chart showing set signal creation timing. [Explanation of Codes] 1 ... Arithmetic unit, 2 ... Memory element, 3 ... Arithmetic function designation register, 4 ... Write mask circuit, D15-D0 ... Input data, A23-A1 ... Address signal, WE ... Write control signal, FS ... Set signal.

─────────────────────────────────────────────────── ───
[Procedure Amendment] [Date of submission] July 17, 1995 [Procedure Amendment 1] [Document name for amendment] Specification [Item name for amendment] Invention title [Amendment method] Change [Amendment content] Name: 1-chip memory device [Procedure amendment 2] [Amendment document name] Specification [Amendment item name] Claims [Amendment method] Change [Amendment content] [Claims] 1. An address terminal (A23 to A1) to which an address signal is externally supplied; a storage element (2) for reading and writing data according to the address signal supplied from the address terminal; and a storage element (2). An arithmetic unit (1) which is connected and arithmetically operates data (Doj) read from the storage element and data (Dj) from outside the chip in a specified arithmetic mode; the address terminals (A23 to A1); Connected to the unit (1), the address is supplied via the address terminals (A4 to A1), and the address is supplied before the access period in which the storage element (2) reads and writes data. The calculation mode designation data supplied from the outside of the chip via the terminal is preset, and the calculation mode (1) designates the calculation mode so as to execute the calculation processing in accordance with the preset calculation mode. 1-chip memory device having a calculation specification register (3) that (9). [Procedure Amendment 3] [Name of Document to be Amended] Specification [Name of Item to Amend] [Correction Method] Change [Details of Amendment] [Problems to be Solved by the Invention] Provided is a one-chip memory device with an arithmetic function capable of setting a desired arithmetic mode from a plurality of arithmetic functions while ensuring versatility with other memory devices without increasing the number of pins of the device. Especially. It is another object of the present invention to provide a one-chip memory device that minimizes the overhead for setting the operation mode. [Procedure Amendment 4] [Document name for amendment] Specification [Item name for amendment] [Correction method] Change [Content of amendment] [Means for solving the problem] In order to achieve the above-mentioned object, A feature of the invention is that an address terminal to which an address signal is externally supplied, a storage element for reading and writing data according to the address signal supplied from the address terminal, and a storage element connected to the storage element, An arithmetic unit that arithmetically operates the data read from the storage element and the data from the outside of the chip in a specified arithmetic mode,
Connected to the address terminal and the arithmetic unit, via the address terminal prior to an access period in which the address is supplied via the address terminal and the storage element reads and writes data. A one-chip memory device having preset operation mode designation data supplied from the outside of the chip, and an operation designation register for designating the operation mode so that the operation unit executes the operation processing according to the preset operation mode. . [Procedure Amendment 5] [Name of document to be amended] Specification [Name of item to be amended] [Correction method] Change [Content of amendment] [Operation] With the above-mentioned configuration, access to the memory element (memory element) The operation mode specification data setting signal is set to the operation specification register via the address terminal at a timing different from the timing of reading the data from the memory and writing the data to the memory element). And holding means for holding the operation mode in accordance with the operation mode designation data input via the address terminal. This makes it possible to specify the operation mode of a one-chip memory device having an operation function having a plurality of operation modes without increasing the number of pins of the one-chip memory device. Further, according to the present invention, since it is not necessary to increase the number of pins of the 1-chip memory device, the 1-chip memory having a plurality of operation modes can be miniaturized, and a general operation function is not provided. It is possible to realize a one-chip memory device that ensures compatibility with various memories. Further, according to the present invention, if a specific one of the plurality of operation modes is held in the holding unit, it is necessary to set the operation mode for data for performing the same operation each time the memory is accessed. Therefore, the overhead for setting the operation mode can be minimized. [Procedure Amendment 6] [Name of Document to Amend] Specification [Name of Item to Amend] [Correction Method] Change [Details of Amendment] [Effect of the Invention] As described above, the present invention provides access to a memory element. The operation mode specifying data setting signal is set to the operation specifying register at a timing different from the (data reading operation from the memory element / data writing operation to the memory element) timing. Configure to input via address terminal,
Holding means is provided for holding the operation mode in accordance with the operation mode designation data input via the address terminal.
Thereby, one with a calculation function having a plurality of calculation modes
The calculation mode of the chip memory device can be specified without increasing the number of pins of the one-chip memory device. Further, according to the present invention, since it is not necessary to increase the number of pins of the 1-chip memory device, the 1-chip memory having a plurality of operation modes can be miniaturized, and a general operation function is not provided. It is possible to realize a one-chip memory device that ensures compatibility with various memories. Further, according to the present invention, if the holding means holds a specific one of the plurality of operation modes,
Since it is not necessary to set the calculation mode for each memory access for the same calculation data, the overhead for setting the calculation mode can be minimized.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G11C 11/401 (72) Inventor Koichi Kimura 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefectural corporation Hitachi Microelectronics Equipment Development Laboratory (72) Inventor Hiromichi Enomoto 1 Horiyamashita, Hadano City Hitachi Ltd. Kanagawa Plant

Claims (1)

[Claims] 1. In a storage circuit that includes a storage element that can arbitrarily read, write, and store data, and an arithmetic unit that operates the first data from the outside and the second data in the storage element, specify a specified arithmetic function code. A register for storing and a register for storing designated write control data are provided, and operation and bit-wise write control are performed based on output data of the arithmetic function code storage register and the write control data storage register. Memory circuit. 2. The memory circuit according to claim 1, wherein a part of an address signal for the memory element is used as data for specifying an arithmetic function. 3. The memory circuit according to claim 1, wherein a part of an address signal for the memory element is used as write control data. 4. The memory circuit according to claim 1, wherein write data for the memory circuit is used as write control data. 5. The memory circuit according to claim 1, wherein the operation by the operation function code and the write control by the write control data are performed by one memory access. 6. The memory circuit according to claim 1, wherein the memory circuit is integrated into one integrated circuit.