Humanity has come a long way towards the creation of computers, without which it is impossible to imagine modern society with all aspects of its life in the fields of industry, the national economy and household appliances. But even today, progress does not stand still, opening up new forms of computerization. At the center of technological development for several decades now is the structure of the microprocessor (MP), which is being improved in its functional and design parameters.
Microprocessor concept
In a general sense, the concept of a microprocessor is presented as a program-controlled device or system based on a large integrated circuit (LSI). With the help of MP, data processing operations or management of systems that process information are performed. At the first stagesThe development of the MP was based on separate low-functional microcircuits, in which transistors were present in quantities from a few to hundreds. The simplest typical microprocessor structure could contain a group of microcircuits with common electrical, structural and electrical parameters. Such systems are called microprocessor set. Along with the MP, one system could also consist of permanent and random access memory devices, as well as controllers and interfaces for connecting external equipment - again, through compatible communications. As a result of the development of the concept of microcontrollers, the microprocessor kit was supplemented with more complex service devices, registers, bus drivers, timers, etc.
Today, the microprocessor is less and less considered as a separate device in the context of practical applications. The functional structure and principle of operation of the microprocessor already at the design stages are guided by the use as part of a computing device designed to perform a number of tasks related to the processing and management of information. The key link in the processes of organizing the operation of a microprocessor device is the controller, which maintains the control configuration and modes of interaction between the system's computing core and external equipment. An integrated processor can be considered as an intermediate link between the controller and the microprocessor. Its functionality is focused on solving auxiliary tasks that are not directly related to the purpose of the main MT. In particular, these can be network and communication functions that ensure the operation of the microprocessor device.
Classifications of microprocessors
Even in the simplest configurations, MPs have many technical and operational parameters that can be used to set classification features. To justify the main levels of classification, three functional systems are usually distinguished - operational, interface and control. Each of these working parts also provides a number of parameters and distinguishing features that determine the nature of the operation of the device.
From the point of view of the typical structure of microprocessors, the classification will primarily divide devices into multi-chip and single-chip models. The former are characterized by the fact that their working units can function offline and execute predetermined commands. And in this example, MPs will be pronounced, in which the emphasis is on the operational function. Such processors are focused on data processing. In the same group, for example, three-chip microprocessors can be control and interface. This does not mean that they do not have an operational function, but for optimization purposes, most of the communication and power resources are allocated to the tasks of generating microinstructions or the ability to interact with peripheral systems.
As for single-chip MPs, they are developed with a fixed set of instructions and compact placement of all hardwareon one core. In terms of functionality, the structure of a single-chip microprocessor is quite limited, although it is more reliable than segment configurations of multi-chip analogs.
Another important classification refers to the interface design of microprocessors. We are talking about ways to process input signals, which today continue to be divided into digital and analog. Although the processors themselves are digital devices, in some cases the use of analog streams justifies itself in terms of price and reliability. For conversion, however, special converters must be used, which contribute to the energy load and structural fullness of the working platform. Analog MPs (usually single-chip) perform the tasks of standard analog systems - for example, they produce modulation, generate oscillations, encode and decode a signal.
According to the principle of temporary organization of the functioning of the MP, they are divided into synchronous and asynchronous. The difference lies in the nature of the signal to start a new operation. For example, in the case of a synchronous device, such commands are given by control modules, regardless of the execution of current operations. In the case of asynchronous MPs, a similar signal can be given automatically upon completion of the previous operation. To do this, an electronic circuit is provided in the logical structure of the asynchronous type microprocessor, which ensures the operation of individual components in an offline mode, if necessary. The complexity of implementing this method of organizing the work of the MP is due to the fact thatalways at the moment of completion of one operation there are enough certain resources to start the next one. Processor memory is typically used as a prioritizing link in the very choice of subsequent operations.
Microprocessors for general and special purposes
The main scope of general purpose MT are workstations, personal computers, servers and electronic devices intended for mass use. Their functional infrastructure is focused on performing a wide range of tasks related to information processing. Such devices are being developed by SPARC, Intel, Motorola, IBM and others.
Specialized microprocessors, whose characteristics and structure are based on powerful controllers, implement complex procedures for processing and converting digital and analog signals. This is a very diverse segment with thousands of configuration types. The structural features of this type of MP include the use of one crystal as a base for the central processor, which, in turn, can be interfaced with a large number of peripheral devices. Among them are the means of input / output, blocks with timers, interfaces, analog-to-digital converters. It is also practiced to connect specialized devices like blocks for generating pulse-width signals. Due to the use of internal memory, such systems have a small number of auxiliary components that support the operationmicrocontroller.
Microprocessor specifications
Operating parameters define the range of device tasks and the set of components that, in principle, can be used in a particular microprocessor structure. The main characteristics of MP can be represented as follows:
- Clock frequency. Indicates the number of elementary operations that the system can perform in 1 second. and is expressed in MHz. Despite the differences in structure, different MPs mostly perform similar tasks, but in each case it requires individual time, which is reflected in the number of cycles. The more powerful the MP, the more procedures it can perform within one time unit.
- Width. The number of bits that the device can execute at the same time. Allocate bus width, data transfer rate, internal registers, etc.
- The amount of cache memory. This is the memory included in the internal structure of the microprocessor and always operating at limiting frequencies. In the physical representation, this is a crystal placed on the main MP chip and coupled to the microprocessor bus core.
- Configuration. In this case, we are talking about the organization of commands and addressing methods. In practice, the type of configuration can mean the possibility of combining the processes of executing several commands at the same time, the modes and principles of MP operation, and the presence of peripheral devices in the basic microprocessor system.
Microprocessor architecture
By and large, MP is universalinformation processor, but in some areas of its operation, special configurations for the execution of its structure are often required. The architecture of microprocessors reflects the specifics of the application of a particular model, causing the features of the hardware and software integrated into the system. Specifically, we can talk about the provided actuators, program registers, addressing methods and instruction sets.
In the representation of the architecture and features of the functioning of the MP, they often use device diagrams and the interaction of available software registers that contain control information and operands (processed data). Therefore, in the register model there is a group of service registers, as well as segments for storing general-purpose operands. On this basis, the method of executing programs, the scheme of memory organization, the mode of operation and the characteristics of the microprocessor are determined. The general-purpose MP structure, for example, may include a program counter, as well as registers for the status and control of the system operation modes. The workflow of a device in the context of an architectural configuration can be represented as a model of register transfers, providing addressing, selecting operands and instructions, transferring results, etc. The execution of different instructions, regardless of the assignment, will affect the status register, the contents of which reflect the current state of the processor.
General information about the structure of microprocessors
In this case, the structure should be understood not only as a set of components of the working system, but alsomeans of connection between them, as well as devices that ensure their interaction. As in the functional classification, the content of the structure can be expressed through three components - operational content, means of communication with the bus and control infrastructure.
The device of the operating part determines the nature of command decoding and data processing. This complex may include arithmetic-logic functional blocks, as well as resistors for temporary storage of information, including information about the state of the microprocessor. The logic structure provides for the use of 16-bit resistors that perform not only logical and arithmetic procedures, but also shift operations. The work of registers can be organized according to different schemes, which determine, among other things, their accessibility to the programmer. A separate register is reserved for the battery pack function.
Bus couplers are responsible for connections to peripheral equipment. The range of their tasks also includes fetching data from memory and forming a queue of commands. The typical microprocessor structure includes an IP command pointer, address adders, segment registers and buffers, through which links with address buses are serviced.
The control device, in turn, generates control signals, decrypts the command, and also ensures the operation of the computing system, issuing micro-commands for internal MP operations.
Structure of basic MP
The simplified structure of this microprocessor provides two functionalparts:
- Operating room. This unit includes control and data processing facilities, as well as microprocessor memory. Unlike the full configuration, the basic microprocessor structure excludes segment registers. Some execution devices are combined into one functional unit, which also emphasizes the optimized nature of this architecture.
- Interface. In essence, a means of providing communication with the main highway. This part contains the internal memory registers and the address adder.
The principle of signal multiplexing is often used on the external output channels of basic MPs. This means that the transmission of signals takes place over common time-sharing channels. In addition, depending on the current operating mode of the system, the same output can be used to transmit signals for different purposes.
Microprocessor instruction structure
This structure largely depends on the general configuration and the nature of the interaction of the MP functional blocks. However, even at the design stage of the system, developers lay down the possibilities for applying a certain array of operations based on which a set of commands is subsequently formed. The most common command functions include:
- Data transfer. The command performs the operations of assigning the values of the source and destination operands. Registers or memory cells can be used as the latter.
- Input-output. ThroughI/O interface devices transfer data to ports. In accordance with the structure of the microprocessor and its interaction with peripheral hardware and internal units, the commands set the port addresses.
- Type conversion. The formats and size values of the operands used are determined.
- Interruptions. This type of instruction is designed to control software interrupts - for example, it can be a processor function stop while I / O devices start working.
- Organization of cycles. Instructions change the value of the ECX register, which can be used as a counter when executing certain program code.
As a rule, restrictions are imposed on basic commands related to the ability to operate with certain amounts of memory, simultaneously manage registers and their contents.
MP management structure
MP control system is based on the control unit, which is associated with several functional parts:
- Signal sensor. Determines the sequence and parameters of pulses, evenly distributing them in time across the buses. Among the characteristics of the operation of sensors is the number of cycles and control signals required to perform operations.
- Source of signals. One of the functions of the control unit in the structure of the microprocessor is assigned to the generation or processing of signals - that is, their switching within a specific cycle on a specific bus.
- Operation code decoder. Performs decryption of the operation codes present in the instruction register onthis moment. Together with determining the active bus, this procedure also helps to generate a sequence of control pulses.
Of no small importance in the control infrastructure is a permanent storage device that contains in its cells the signals required to perform processing operations. To count commands when processing pulse data, an address generation unit can be used - this is a necessary component of the internal structure of the microprocessor, which is included in the interface unit of the system and allows you to read the details of the memory registers with signals in full.
Microprocessor components
Most of the functional blocks, as well as external devices, are organized between themselves and the central microcircuit MP through the internal bus. It can be said that this is the backbone network of the device, providing a comprehensive communication link. Another thing is that the bus can also contain elements of different functional purposes - for example, circuits for data transfer, lines for transferring memory cells, as well as an infrastructure for writing and reading information. The nature of the interaction between the blocks of the bus itself is determined by the structure of the microprocessor. The devices included in the MP, in addition to the bus, include the following:
- Arithmetic logic unit. As already mentioned, this component is designed to perform logical and arithmetic operations. It works with both numeric and character data.
- Control device. Responsible forcoordination in the interaction of different parts of the MT. In particular, this block generates control signals, directing them to different modules of the machine device at certain points in time.
- Microprocessor memory. Used for recording, storing and issuing information. Data can be associated with both working computational operations and processes serving the machine.
- Math processor. It is used as an auxiliary module to increase the speed when performing complex computational operations.
Features of the structure of the coprocessor
Even within the framework of performing typical arithmetic and logical operations, there is not enough capacity of a conventional MP. For example, the microprocessor does not have the ability to execute floating point arithmetic instructions. For such tasks, coprocessors are used, the structure of which provides for the combination of a central processor with several MPs. At the same time, the logic of the device operation itself has no fundamental differences from the basic rules for constructing arithmetic microcircuits.
Coprocessors execute typical commands, but in close interaction with the central module. This configuration assumes constant monitoring of command queues across multiple lines. In the physical structure of a microprocessor of this type, it is allowed to use an independent module to provide input-output, a feature of which is the ability to select its commands. However, for such a scheme to work correctly, coprocessors must clearly define the source of instruction selection,coordinating interaction between modules.
The principle of building a generalized structure of a microprocessor with a strongly coupled configuration is also connected with the concept of a coprocessor device. If in the previous case we can talk about an independent I / O block with the possibility of its own selection of commands, then a strongly coupled configuration involves the inclusion in the structure of an independent processor that controls command streams.
Conclusion
The principles of creating microprocessors have undergone little change since the advent of the first computing devices. The characteristics, designs and requirements for resource support have changed, which has radically changed the computer, but the general concept with the basic rules for organizing functional blocks for the most part remains the same. However, the future of microprocessor structure development may be influenced by nanotechnology and the advent of quantum computing systems. Today, such areas are considered at the theoretical level, but large corporations are actively working on the prospects for the practical use of new logic circuits based on innovative technologies. For example, as a possible option for the further development of MT, the use of molecular and subatomic particles is not ruled out, and traditional electrical circuits can give way to systems of directed electron rotation. This will make it possible to create microscopic processors with a fundamentally new architecture, the performance of which will many times exceed today's. MP.
