KR19990024475A - System and data access method with integrated microcontroller and digital signal processor - Google Patents

Abstract

The present invention discloses a system and data access method having a microcontroller and a digital signal processor integration device. The system comprises a microcontroller and digital signal processor integrating means incorporating an M-bit microcontroller and an N-bit digital signal processor, and outputting N bits of data to the microcontroller and digital signal processor integrating means in response to an address signal of a predetermined bit. And a first and second memory, wherein the method for accessing data of the system is to access data from one of the first memory and the second memory when the device operates as a digital signal processor, or the first memory and the second memory. It simultaneously accesses two necessary data from the second memory, and when the device operates as a microcontroller, simultaneously accesses data stored at the same address of the first memory and the second memory. Thus, memory data can be accessed effectively using a small memory device.

Description

System and data access method with integrated microcontroller and digital signal processor

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital signal processing system, and more particularly, to a system and a data access method including a microcontroller and a digital signal processor integrator capable of effectively accessing memory data when integrating a digital signal processor and a microcontroller.

In general, the application size of a microcontroller (MCU) is large, and code is written using a high-level language and a compiler is used. On the other hand, digital signal processors (DSPs) are often used in places where a large amount of real-time computation is required and have a structure that is difficult for the compiler to achieve the desired performance. Therefore, assembler code is often written by hand.

Increasingly, real-time computational and control-intensive applications are increasing, and microcontrollers or digital signal processors are not satisfactory in these applications. The following methods have been used for such applications.

First, it uses two processors of independent microcontroller and digital signal processor. It does not require new processor, but it is expensive and it is not easy to develop because it uses two different environments.

Secondly, the instructions of the digital signal processor are modified to be suitable for the compiler. However, if the structure of the conventional digital signal processor is maintained, there is a limit to the effect of the microcontroller.

Third, a co-processor is attached to the microcontroller to help real-time calculation of multiplication and accumulation engines. Such a structure does not use resources efficiently and coprocessors. Performance is degraded by communication overhead with the system.

Applications that require a lot of real-time computation and require a lot of control require a new microcontroller and digital signal processor integrated processor that combines the functionality of a microcontroller with the capabilities of a digital signal processor. There are many problems in the development of microcontroller and digital signal processor integrated processors, one of which is the bus structure for memory access. To perform real-time calculations, a Harvard structure, in which the program and data buses are separated, is appropriate. In general, 32-bit microcontrollers are suitable for supporting large memory areas and performing 32-bit integer operations in applications that require a large amount of real-time computation and require a lot of control, whereas 16-bit is sufficient for digital signal processors. . Thus, 32-bit microcontrollers and 16-bit digital signal processors are used in these applications. However, when these two processors are combined into one, the number of bits differs from each other, which causes a problem. In particular, you must decide what memory structure and bus to use when accessing memory devices. A common method used by microcontrollers that support both 32-bit and 16-bit data at the same time is to disable the 16-bit bus when not using 16-bit data with 32-bit memory and 32-bit data bus. Get 32 bits and use only 16 bits. However, this structure is wasteful in terms of digital signal processors.

It is an object of the present invention to provide a system having a microcontroller and digital signal processor integrator that can effectively use a memory device.

Another object of the present invention is to provide a data access method of a system having a microcontroller and a digital signal processor integrator that can effectively access a memory device in the microcontroller and a digital signal processor integrator.

In order to achieve the above object, a system having a microcontroller and digital signal processor integrator according to the present invention includes a microcontroller and digital signal processor integrating means incorporating an M-bit microcontroller and an N-bit digital signal processor, and a predetermined bit address signal. First memory means for responsively outputting N bits of data to said microcontroller and digital signal processor integrating means, and N bits of data to said microcontroller and digital signal processor integrating means in response to said predetermined bit address signal; And second memory means for outputting.

A data access method of a system having a microcontroller and digital signal processor integrating apparatus of the present invention for achieving the above another object is a microcontroller and digital signal processor integrating means incorporating an M-bit microcontroller and an N-bit digital signal processor. First memory means for outputting N bits of data to said microcontroller and digital signal processor integrator in response to a bit address signal, and N bits of data to said microcontroller in response to said predetermined bit address signal; A data access method of a device having a second memory means for outputting to a signal processor integrating means, wherein the device operates from a digital signal processor from a memory device of either the first memory means or the second memory means. Access or access your data Simultaneously accesses two necessary data from the first memory means and the second memory means, and simultaneously accesses data stored at the same address of the first memory means and the second memory means when the apparatus operates as a microcontroller It is characterized by.

1 is a block diagram illustrating a data access method of a system having a microcontroller and digital signal processor integrator according to the present invention.

Hereinafter, a system and a data access method including a microcontroller and a digital signal processor integrating apparatus of the present invention will be described with reference to the accompanying drawings.

In digital signal processing, two 16-bit data can be imported in one cycle in order to perform multiplication and accumulation (MAC) instructions in one cycle. The two buses that import two data may use the program bus temporarily as a data bus like TI or Analog Device's digital signal processor, and use two data areas and two buses as DSP group's digital signal processor. One data area may be used in a limited way, or two symmetrical data areas and two buses, such as SGS-Thomsonm and Motorola's DSPs, may be used. The present invention is based on the use of two symmetrical data areas and two data buses. When 32-bit data is required, the two data buses use 16 data bits to obtain 32 bits by using two data buses. will be.

1 is a block diagram illustrating a data access method of a system including a microcontroller and a digital signal processor (MCU-DSP) integrator according to the present invention. The MCU-DSP integrator 10, XRAM 20, and It consists of the YRAM 30.

The structure shown in Fig. 1 has a structure similar to that of the digital signal processor memory structure using two symmetrical 16-bit data regions and two 16-bit buses.

The two 16-bit data areas are distinguished by one bit of memory address. For example, XRAM 20 corresponds to an even memory address and YRAM 30 corresponds to an odd memory address. If a 32-bit memory address is used, the address XA of the XRAM 20 is the upper 31 bits of the memory address and the least significant bit is used for chip selection.

The operation of the structure shown in Fig. 1 when accessing 16-bit data output memory devices (XRAM, YRAM) using a 32-bit memory address is explained as follows.

First, when the MCU-DSP integrated device operates as a digital signal processor to access 16-bit data, one 16-bit data may be accessed from one of the memory devices of the XRAM 20 or the YRAM 30. When a digital signal processor requires two 16-bit data to perform an operation such as a MAC command, two 16-bit data may be simultaneously accessed from the memory devices of the XRAM 20 and the YRAM 30. . In this case, the least significant bit of the XRAM 20 and the least significant bit data of the YRAM 30 are used for chip selection. The least significant bit of the XRAM 20 is 0, the least significant bit of the YRAM 30 is 1, and the rest By changing the upper 31-bit address, two 16-bit data required can be accessed.

Next, when the MCU-DSP integrated device is used as a microcontroller to access 32-bit data required by the microcontroller, upper 16-bit data is stored in the XRAM 20 and lower 16-bit data is stored in the YRAM 30. If so, the memory addresses XA and YA may be set by making the upper 31-bit address of the XRAM 20 and the upper 31-bit address of the YRAM 30 the same and setting the least significant bits for chip selection to 0 and 1. In this way, the memory devices XRAM and YRAM output the upper 16-bit data and the lower 16-bit data corresponding to the same address. In this way, two 16-bit memory devices can be used to access 32-bit data.

The system and data access method with the MCU-DSP integrated device of the present invention can effectively access memory data even when the device operates as a microcontroller using a memory device suitable for a digital signal processor.

Therefore, the system and data access method including the MCU-DSP integrated device of the present invention can effectively access the memory data using a small memory device.

Claims (6)

Microcontroller and digital signal processor integrating means incorporating an M-bit microcontroller and an N-bit digital signal processor; First memory means for outputting N bits of data to said microcontroller and digital signal processor integrating means in response to a predetermined bit of address signal; And And second memory means for outputting N-bit data to said microcontroller and digital signal processor integrating means in response to said predetermined bit address signal. The system as claimed in claim 1, wherein a specific bit signal of the predetermined bit address signal is used as a selection signal for selecting the first memory means and a second memory means, and the remaining bit signal is used as an address signal. . 3. The system according to claim 2, wherein the upper and lower data of M-bit data for use of a microcontroller are stored in predetermined same address areas of the first memory means and the second memory means, respectively. Microcontroller and digital signal processor integrating means incorporating an M-bit microcontroller and an N-bit digital signal processor; First memory means for outputting N bits of data to said microcontroller and digital signal processor integrating means in response to a predetermined bit of address signal; And a second memory means for outputting N bits of data to said microcontroller and digital signal processor integrating means in response to said predetermined bit address signal. When the device operates as a digital signal processor, data from one of the first memory means or the second memory means is accessed, or two pieces of data from the first and second memory means are required. And simultaneously access the data stored at the same address of the first memory means and the second memory means when the device operates as a microcontroller. The system according to claim 4, wherein a specific bit signal of the predetermined bit address signal is used as a selection signal for selecting the first and second memory means and the remaining bit signal is used as an address signal. Data access method. 6. The system data according to claim 5, wherein upper and lower data of M-bit data for use of a microcontroller are stored in predetermined same address areas of the first memory means and the second memory means, respectively. Access method.