CN100365576C - Implementation Method of Embedded Operating System Memory Mirroring Startup Optimization - Google Patents

Abstract

The invention discloses a method for realizing the optimization of memory image startup of an embedded operating system. The method of the present invention saves the register content of the running operating system, the memory content in use, and the corresponding state during the operation of the embedded operating system, and saves them on the NOR Flash of the hardware system, and automatically saves the memory content when starting. Mirroring to DRAM restores the state of the embedded operating system at the time of saving. The method of the invention can significantly accelerate the restarting speed, so that the starting speed of the embedded system is significantly improved, and the waiting time of the user is reduced.

Description

Implementation Method of Embedded Operating System Memory Mirroring Startup Optimization

technical field

The invention relates to the field of startup optimization of an embedded operating system, in particular to a method for realizing startup optimization of an embedded operating system memory image.

Background technique

In order to make the development of embedded system more convenient and fast, there needs to be a software module specially responsible for managing memory allocation, interrupt processing, task scheduling and other functions, which is the embedded operating system. Embedded operating system is the system software used to support embedded applications. It is an extremely important part of the embedded system. )wait. The embedded operating system has the basic characteristics of a general-purpose operating system, such as being able to effectively manage complex system resources, abstracting hardware, and providing library functions, drivers, and development tool sets. However, compared with general-purpose operating systems, embedded operating systems have more distinctive features in terms of system real-time performance, hardware dependence, software solidification, and application specificity.

Embedded operating systems can be divided into two categories according to the application occasions: one is non-real-time systems for consumer electronics products, such devices include personal digital assistants (PDAs), mobile phones, set-top boxes (STBs), etc. such as WinCE, PalmOS, Symbian OS; the other is a real-time operating system for control, communication, medical and other fields, such as VxWorks of WindRiver Company, QNX of QNX System Software Company, etc.

Embedded systems are user-oriented, product-oriented, and application-oriented. If there is no established goal and application prospect, its application value will be lost. Therefore, an embedded system is quite sensitive to cost, and its software and hardware design must be efficient. A good embedded system is the smallest system that accomplishes the target function. The power consumption, volume, cost, reliability, speed processing capability, electromagnetic compatibility and other aspects of embedded processors are all constrained by application requirements. These are also hot spots of competition among chip makers. Embedded processors cut and add chip configurations according to the specific requirements of users, so as to achieve ideal performance.

Embedded systems generally require high reliability. In the harsh environment or sudden power failure, the system is required to still work normally. There are also many embedded applications that require real-time functions, which requires the embedded operating system (EOS) to have real-time processing capabilities. Embedded systems are organically combined with specific applications, and its upgrades are carried out simultaneously with specific products. Therefore, once embedded system products enter the market, they will have strong vitality.

The application fields of embedded systems are generally small electronic devices, and the system resources are relatively limited, so the requirements for the kernel are quite high, which are much smaller than traditional operating systems, such as the OSE distributed embedded system launched by ENEA. , the entire kernel is only 5KB. The system software and application software in the embedded system usually have no obvious difference, and their functions and implementation are not required to be too complicated, which is beneficial to control the system cost on the one hand, and to ensure the security of the system on the other hand.

Embedded systems usually require a certain real-time guarantee. In order to improve the execution speed and system performance, the software in the embedded system is generally solidified in the memory chip or the internal storage device of the processor instead of being stored in an external disk such as a disk. in the carrier. Since the computing speed and storage capacity of embedded systems are limited to a certain extent, and most systems must have high real-time guarantees, there are high requirements for software quality (especially reliability).

The hardware resource environment in embedded systems is generally harsh, and the memory of embedded microprocessors and microcontrollers is generally not large. To complete complex functions under such tight resources, it is required that the embedded system software must be as compact as possible. , stable and efficient. In this case, as a mobile embedded device user, if you are facing a general-purpose operating system, you often cannot bear the long startup time. At the same time, for the embedded operating system, you should try to shorten its startup time so To meet the user's requirements, and the boot time has become an important factor affecting the user's product selection, therefore, a new method is needed to reduce the boot time of the embedded operating system.

Contents of the invention

The purpose of the present invention is to provide a method for realizing the optimization of memory image startup of an embedded operating system.

The technical scheme that the present invention solves its technical problem adopts is as follows:

1) On-site state detection of the operating system:

The daemon process detects whether the operating system is in an interrupt processing state, and if so, cancels the operation of saving the image;

2) Mirror the operating system site:

① Check the state of all processes in memory and virtual memory, and set all process states to a suspended state;

② Send a notification that the operating system is suspended to the device registered in the performance management module;

③ Read the contents of the current CPU registers, the contents of the memory in use and the corresponding status, and write them into the hardware system NOR Flash as a memory image;

④Shut down or restart the operating system;

3) The embedded operating system starts:

① Start the operating system kernel, complete the initialization of the process and the basic startup process;

②Read the memory image from the hardware system NOR Flash;

③Restore the CPU register content, memory content and corresponding state saved by the memory mirror;

④ Send a notification of system recovery to the device registered in the performance management module;

5. Restart the process that is placed in the suspended state in step 2) of the process of saving the mirror image.

Compared with the background technology, the present invention has the beneficial effects that:

The present invention is a kind of implementation method of embedded operating system memory image start-up optimization, and its main function is to automatically mirror the memory to the dynamic Random access memory, which restores the state of the embedded operating system at the time of saving. Can significantly speed up restarts.

(1) Efficiency. The memory image of the running state of the embedded operating system is stored in the on-chip memory, making full use of the characteristics of the capacity and speed of the existing embedded CPU's on-chip memory.

(2) The startup time is accelerated. When the embedded operating system is started, starting from the memory image stored on the on-chip memory can significantly accelerate the speed of system booting and restarting, and reduce the waiting time of users.

Description of drawings

Accompanying drawing is the schematic diagram of the process of the present invention.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

A kind of implementation method of embedded operating system memory image start-up optimization, its specific implementation method is as follows:

1) On-site state detection of the operating system:

The daemon process detects whether the operating system is in the interrupt processing state. Interrupt is an important interface for the communication between the core part of the computer and the peripheral equipment. No matter what the core part is doing, if there is an interrupt, it must stop and process. After the program written for this peripheral device is executed, the work just done is resumed. In order for the computer to work properly, interrupts must be reasonably allocated to peripheral devices so that they do not conflict. And if the processor is in the interrupt state, it must first complete the interrupt task. If the operating system is in the interrupt processing state, cancel the saved operation;

2) Mirror the operating system site:

①The operating system checks the state of all processes in the memory, and sets all process states to the suspended state for processes in the following states: new state, corresponding to the state that the process has just been created; running state: the process occupies the processor and is running; ready State: The process has the conditions to run, waiting for the operating system to allocate a processor to run; Waiting state: The process does not have the conditions to run, and is waiting for the completion of an event.

② Send a notification that the operating system is suspended to the device registered in the performance management module;

③The vast majority of calculations are carried out in the microprocessor. Registers are special storage blocks in the processor that are used to retrieve and store processed information. During the running of the embedded operating system, read the contents of the current CPU register.

The core process of the embedded operating system, user process and related information are stored in the memory, and written into the hardware system NOR Flash;

④Shut down or restart the operating system;

3) The embedded operating system starts:

①Start the kernel after booting, and complete the initialization of the process and the basic startup process;

The booting and initialization of the operating system is the first step in the control of the operating system, and it is also an important part that embodies the advantages and disadvantages of the system. First, a part of the kernel is booted with a program to guide other parts of the kernel. After the kernel is decompressed and loaded into memory, the hardware and device drivers are initialized. The kernel will start the initialization program to form the first process of the operating system. The following table is some startup information when the embedded Linux operating system starts:

Linux version 2.4.19-rmk4-pxa2-alvin2(wliu@weiwei-dev)(gcc version 3.2.1)#2 TueAug 26 09:55:45 EDT 2003CPU:Intel XScale-PXA255 revision 6Machine:Intel Sitsang Development PlatformIgnoringed unrecognized 0x00000000Memory clock:99.53MHz(*27)Run Mode clock:398.13MHz(*4)Turbo Mode clock:398.13MHz(*1.0，inactive)On node 0 totalpages:16384

② After the operating system completes the basic initialization work and starts the first process, it detects the hardware system NOR Flash, reads its first page, and checks whether it is a memory image, and reads the memory image from the hardware system NOR Flash;

③ load the CPU register content saved in the memory mirror into the CPU register, and load the memory content saved in the memory mirror, including the process, each data structure of the process, and other data information into the DRAM;

④ Send a notification of operating system recovery to the device registered in the performance management module;

⑤ Restart the process that was suspended during the process of saving the image.

The present invention can be realized by using assembly language, C language or C++ language.

Claims (1)

1. an implementation method of embedded operating system memory image startup optimization, characterized in that the steps of the method are as follows: 1) On-site state detection of the operating system: The daemon process detects whether the operating system is in an interrupt processing state, and if so, cancels the operation of saving the image; 2) Mirror the operating system site: ① Check the state of all processes in memory and virtual memory, and set all process states to a suspended state; ② Send a notification that the operating system is suspended to the device registered in the performance management module; ③ Read the contents of the current CPU registers, the contents of the memory in use and the corresponding status, and write them into the hardware system NOR Flash as a memory image; ④Shut down or restart the operating system; 3) The embedded operating system starts: ① Start the operating system kernel, complete the initialization of the process and the basic startup process; ②Read the memory image from the hardware system NOR Flash; ③Restore the CPU register content, memory content and corresponding state saved by the memory mirror; ④ Send a notification of system recovery to the device registered in the performance management module; 5. Restart the process that is placed in the suspended state in step 2) of the process of saving the mirror image.