CN100407139C - File system and method for constructing enhanced functions thereof, operating system, recording device - Google Patents

Abstract

The invention relates to a file system under an operating system. The file system includes: an entry interface conversion unit, a management unit, a physical file system unit, a top drive unit, and an exit interface conversion unit. The interface conversion unit of the entrance is connected to the application program unit and the management unit, the management unit is connected to the interface conversion unit of the entrance and the physical file system unit, the physical file system unit is connected to the management unit and the top drive unit, and the top drive unit is connected to the physical file system unit and the exit An interface conversion unit, the outlet interface conversion unit connects the top drive unit with the drive unit of the operating system. The file system of the present invention reduces the VFS unit under the Linux standard file system, reduces the size of the kernel, can realize special transaction processing customized by various users, and can conveniently realize transplantation across operating systems. The invention also includes an operating system with the file system of the invention, and relates to a method for constructing and enhancing the function of the file system.

Description

File system and method for constructing enhanced functions thereof, operating system, recording device

technical field

The present invention relates to a file system under an operating system, and more specifically, the present invention relates to a file system with enhanced functions and an operating system with the file system of the present invention. The invention also relates to a method for constructing and enhancing the function of the file system.

Background technique

In a computer system, a file is the logical organization and presentation of data. Usually, when accessing data in a specific file, the file can be displayed in a recognizable logical form only through the positioning and organization of the file system. In the prior art, an operating system can generally support a plurality of different physical file system units, for example, the physical file system supported by the Linux operating system (Linux is a registered trademark or trademark of Linus Torvalds in the United States and other countries) includes ext2, ext3, udf, jfs, etc. In order to support these different physical file systems in a unified way, in the Linux kernel, between the application program unit and the actual physical file system unit, there is a unified interface unit VFS for different physical file system units ( Virtual File System)-Virtual file system, which acts as a switch to provide support for different physical file system units. Therefore, the virtual file system can also be called a virtual file system unit. VFS provides a set of standard operation functions to manage and operate each specific file system. In actual application, various file operation instructions issued by the user unit are received by the VFS unit through system calls. After unified scheduling and management, the VFS unit sends the operation instructions to the specific physical file system unit for processing according to the settings. , to achieve the operation requested by the user.

FIG. 1 shows a standard file system under the Linux operating system in the prior art. As shown in FIG. 1 , a standard file system includes a VFS unit 10 , a physical file system unit 20 , which may include specific physical file systems such as ext2, ext3, udf, and jfs, and a drive unit 30 . As mentioned above, the Linux operating system adopts the VFS structure to uniformly schedule and manage each specific physical file system under the standard file system.

Although the above-mentioned standard file system is adapted to the coordinated control of multiple physical file systems, it has the following disadvantages and deficiencies:

First, the lack of sufficient flexibility makes it difficult for users to implement special functions that meet their own requirements. Under the above-mentioned file system, the scheduling of each process, buffer maintenance, equipment maintenance, mutual coordination, and the realization of specific functions are all constrained by the file system of the VFS and the operating system. It is difficult for users to customize the file system according to their own requirements. Add or delete specific functions, or perform process scheduling required by users.

Second, in order to adapt to the compatibility of multiple file systems, the existing file system has a complex structure and a large scale. When we need to customize the operating system with a small kernel and only need one or two specific file systems at this time, the standard file system with VFS as the main body is too bloated, and many unnecessary structures cannot be deleted. system burden. Especially when we target embedded systems, this problem is particularly prominent.

Third, it is not convenient for cross-platform portability of the file system. When we need to transplant a target file system from an operating system platform to a Linux operating system platform, especially when various algorithms customized by ourselves are written inside the target file system, or the target file system is based on the original operating system Its requirements are organized. At this time, it will be very difficult to transplant it across operating systems to adapt to the VFS standard file system. In order to access each specific target file system, VFS provides multiple data interface structures such as super_block, file, inode, dentry, etc. Due to the different operating system platforms, it is very difficult for a file system that is not originally on the Linux platform to find, adapt to, and access the various interfaces of the above-mentioned VFS, that is, the data interface structures of super_block, file, inode, dentry, etc. matter.

Chinese patent application, publication number CN 1553325A, proposes a method for transparently expanding file system functions. Its file system includes the virtual file system VFS unit and the underlying file system. A mezzanine file system is also set between the standard call interface of the VFS unit and the underlying file system interface. When an application sends a file system request, the system calls The interface first submits the request to the VFS unit processing function; the VFS submits the request to the mezzanine file system through the standard calling interface; after the special transaction processing is performed by the mezzanine file system, the transaction is submitted to the underlying specific file system; finally The actual data access or input and output is realized by the underlying file system.

Although the application has added its own customized mezzanine file system, it can realize some self-defined special functions. However, due to the existence of VFS, system process scheduling, buffer maintenance, device maintenance, memory management, etc. are still carried out by VFS, so that the user-defined mezzanine file system unit is used to realize the space and possibility of user-customized functions. Greatly reduced. At the same time, because this structure adds new content on the basis of the original structure, the volume of the entire kernel is further increased, which is not conducive to the use of resource-limited systems such as embedded systems. At the same time, in the process of porting the file system across operating system platforms, this application is no different from the existing Linux standard file system, both of which contain VFS units, and porting is still very difficult. Due to different operating system platforms, the upper and lower interfaces of the file system are also different. It is very difficult to extract the Linux interface from the non-Linux system file system and access the Linux system. This invention does not completely solve the above problems. .

Therefore, there is a need for a file system with enhanced functions, so that users can flexibly add and delete various functions customized by themselves, and the file system customized by users can be easily transplanted across operating system platforms, and can minimize Small amount of code modification reduces the size of the kernel.

Contents of the invention

The purpose of the present invention is to propose a file system with enhanced functions, so that users can flexibly add and delete various functions customized by themselves; the file system customized by users can be conveniently transplanted across operating system platforms; Minimize the amount of code modification and reduce the size of the kernel; and this kind of file system does not cause any damage to the original file system, and can be started or closed as needed.

The present invention provides a file system under an operating system. The file system includes: an entry interface conversion unit, a physical file system unit, a top drive unit, and an exit interface conversion unit, wherein the entry interface conversion unit is connected to an application A program unit and the physical file system unit, the physical file system unit is connected to the interface conversion unit of the entry and the top drive unit, and the top drive unit is connected to the interface conversion unit of the physical file system unit and the exit unit, the interface conversion unit of the outlet connects the top drive unit and the drive unit of the operating system.

The present invention provides a file system under an operating system. The file system includes: an entry interface conversion unit, a management unit, a physical file system unit, a top drive unit, and an exit interface conversion unit, wherein the entry interface conversion unit connecting the application program unit and the management unit, the management unit connecting the interface conversion unit of the entry and the physical file system unit, the physical file system unit connecting the management unit and the top drive unit, the The top drive unit is connected to the physical file system unit and the interface conversion unit of the exit, and the interface conversion unit of the exit is connected to the top drive unit and the drive unit of the operating system.

In the file system of the present invention, the interface conversion unit of the entry is used to convert the input command into a format accepted by the management unit, and the management unit processes the input command and data, and provides a physical The support of the file system, the physical file system unit includes the physical file system for processing commands and data from the management unit, after the commands and data are processed by the top drive unit, they are converted to the interface of the exit The interface conversion unit of the outlet converts the commands and data from the top drive unit into control commands and data accepted by the drive unit of the operating system, and provides them to the drive unit of the operating system.

In the file system of the present invention, the physical file system unit includes at least a standard physical file system complying with standard protocols, a user-customized non-standard physical file system, an added physical file system, and user-defined file systems transplanted from other operating systems. One of the physical file systems, the management unit includes general and special functions required for managing and implementing these physical file systems.

In the file system of the present invention, specific function customization can also be added to the management unit according to user needs, and the function requirement can be realized and the function can be provided according to the added function customization.

In the file system of the present invention, the file system may also include a target file system transplanted from another operating system with information related to the original operating system or the original hardware device, and the target file system is related to the Some or all of the management unit, physical file system unit, and top drive unit corresponding to the file system.

The invention also provides an operating system with the file system of the invention. The operating system may further include a standard file system including VFS.

In the operating system, the file system of the present invention may be a module that can be loaded into or unloaded from the operating system at any time. The file system of the present invention can also be added to or removed from the operating system kernel as part of the operating system kernel.

The present invention also provides a method for constructing and enhancing the function of a file system. The method includes the steps of: setting a file system including an entry interface conversion unit, a physical file system unit, a top drive unit, and an exit interface conversion unit, wherein, The application program unit is connected to the interface conversion unit of the entry, the interface conversion unit of the entry is connected to the physical file system unit, the physical file system unit is connected to the top drive unit, and the top drive unit is connected to the exit An interface conversion unit, the interface conversion unit of the outlet is connected to the drive unit of the operating system. The method further includes the step of converting the input command through the interface conversion unit of the entry and providing it to the physical file system unit, the physical file system unit includes a physical file system, and processes the interface conversion from the entry The command and data of the unit, after being processed by the top drive unit, the command and data will be sent to the interface conversion unit of the outlet, and the command and data from the top drive unit will be converted into the operated by the interface conversion unit of the outlet The drive unit of the system accepts the control commands and data that the hardware device can recognize, and provides them to the drive unit of the operating system.

The present invention also provides a method for constructing and enhancing file system functions, said method comprising the steps of: setting a file system including an entry interface conversion unit, a management unit, a physical file system unit, a top drive unit, and an exit interface conversion unit , the application program unit is connected to the interface conversion unit of the portal, the interface conversion unit of the portal is connected to the management unit, the management unit is connected to the physical file system unit, and the physical file system unit is connected to the top drive unit , the top drive unit is connected to the interface conversion unit of the outlet, and the interface conversion unit of the outlet is connected to the drive unit of the operating system, wherein the interface conversion unit of the entrance is used to convert the input command into The format accepted by the unit, the management unit processes the input commands and data, and provides support for the physical file system. The physical file system unit includes the physical file system and is used to process the input from the management unit Commands and data, after the commands and data are processed by the top drive unit, they are sent to the interface conversion unit of the outlet, and the interface conversion unit of the outlet converts the commands and data from the top drive unit into the driver of the operating system The control command and data accepted by the unit and recognized by the hardware device are provided to the drive unit of the operating system.

The method for constructing and enhancing the file system function of the present invention also includes the step of realizing general and special transaction processing, the step includes: setting a standard physical file system complying with the standard protocol and/or a user-customized file system in the physical file system unit For non-standard physical file systems, the management unit provides general and special functions required to manage and implement standard and/or non-standard physical file systems. At the same time, specific function customization can be added to the management unit according to user needs, so as to realize specific functional requirements. Furthermore, in special cases, the management unit can be omitted, and the application program commands can directly enter the physical file system layer through the interface conversion layer of the entry.

The method for building and enhancing file system functions of the present invention includes the steps of adding and deleting a certain function, and the steps include: adding or deleting a corresponding physical file system in a physical file system unit, or adding or deleting a corresponding physical file system in an existing physical file system The steps of adding and deleting the part corresponding to the function of the file system, modifying the management unit and the top drive unit, and providing or deleting the corresponding function support and driving function.

The method for constructing and enhancing the function of the file system of the present invention also includes the steps of adding function customization in the management unit according to user needs to realize the function requirement and provide the function.

The method for constructing and enhancing file system functions of the present invention includes the step of transplanting user-customized functions across platforms, and the steps further include: adding user-customized physical files transplanted from another operating system to the physical file system unit. The file system modifies the management unit and the top drive unit to provide general and special functions and driver functions needed to manage and implement the migrated physical file system.

The method for constructing and enhancing file system functions of the present invention also includes the step of accessing a target file system transplanted from another operating system with information related to the original operating system or original hardware device, and the steps further include A step of setting a management unit, a physical file system unit and a top drive unit corresponding to the file system in the target file system, and connecting these units with corresponding parts of the file system.

The present invention also provides a file recording device with the above-mentioned file system, which is characterized in that it includes: an application program unit, running data adapted to other file systems; and a hardware device, receiving data from the drive unit of the operating system, wherein , the entry interface conversion unit converts the data from the application program unit adapted to the operation of the other file systems into data adapted to the operation of the file system; the management unit is used to control the physical file system unit, for The hardware device schedules the process of the data to be received; the physical file system unit is used to process the data from the management unit; the top drive unit is used to process the data from the physical file system unit, and It is sent to the export interface conversion unit; the export interface conversion unit converts the data adapted to the operation of the file system from the top drive unit into data suitable for the operation of the other file systems, and sends it to the The driver unit of the above operating system.

The file system of the present invention reduces the number of VFS units, and sets management units as required. Compared with the prior art, the file system of the present invention reduces the size of the kernel, can realize special transaction processing customized by various users, and can be conveniently implemented. Migration across operating systems.

Description of drawings

Fig. 1 shows the VFS file system under Linux;

Fig. 2 shows the file system of the present invention of Linux and the standard VFS file system;

Fig. 3 shows the file system in the uITRON system;

Fig. 4 shows, according to an embodiment of the present invention, the schematic diagram that file system is transplanted to Linux system from uITRON system;

Fig. 5 shows an embodiment of the present invention, in the file system of the present invention, adds the schematic diagram of real-time function to UDF physical file system;

Fig. 6 shows a schematic diagram of modifying the management unit according to an embodiment of the present invention; and

Fig. 7 shows a schematic diagram of a modification to a top drive unit according to an embodiment of the present invention.

Detailed ways

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

The right side of Fig. 2 shows the file system of the present invention. As shown in FIG. 2 , the file system of the present invention includes: an interface conversion unit 1 for the file system entry, a management unit 2 , a physical file system unit 3 , a top drive unit 4 , and an interface conversion unit 5 for the exit. The application program unit is connected to the interface conversion unit 1 of the portal, the interface conversion unit 1 of the portal is connected to the management unit 2, the management unit 2 is connected to the physical file system unit 3, the physical file system unit 3 is connected to the top drive unit 4, and the top drive unit 4 is connected to the The export interface conversion unit 5 is connected with the drive unit of the operating system.

When an application program sends a request to the file system, the command is submitted to the interface conversion unit 1 of the entry through a specific method, such as a system call or an extension based on an existing system call, and the input is converted through the interface. The commands are accepted by management unit 2 that supports user-defined physical file systems. The management unit 2 provides special transaction processing and is responsible for realizing various customizations required by users, such as process scheduling methods, real-time algorithms, memory management, and so on. If there is no need for these special customizations, the management unit 2 here can be omitted. After that, enter the physical file system unit 3, where the physical file system can be a standard file system that complies with standard protocols, or a non-standard file system customized by the user's own algorithm, and it can also be transplanted from other platforms The target file system contains more "traces" of the original operating system. The "trace" here means that the outer encapsulation of the standard physical file system implemented by the original operating system platform can be retained, for example, the VFS-based encapsulation added by the Linux operating system to the standard physical file system. After being processed by the physical file system unit, commands and data are converted into commands recognized by the operating system through the top drive unit and the interface conversion unit of the exit, and the output file system part is accepted by the original underlying drive unit of the operating system for data conversion. further processing.

The file system of the present invention can omit the management unit 2 located between the entry interface conversion unit 1 and the physical file system unit 3 . Therefore, the file system of the present invention can include the interface conversion unit 1 of the file system entry, the physical file system unit 3, the top drive unit 4, and the interface conversion unit 5 of the exit, wherein, since the management unit 2 is omitted, the interface conversion of the entry Unit 1 is directly connected to physical file system unit 3 . Omitting the management unit 2 is a special case, at this time, the file system can also provide the required transaction processing.

The file system of the present invention reduces the number of VFS units, and sets management units according to needs. Compared with the prior art, the file system of the present invention reduces the size of the kernel, and can realize special transaction processing customized by users, and can be conveniently implemented across Operating system porting.

The file system of the present invention can be used alone in an operating system, and can also be used in parallel with a standard file system. The file system of the present invention can be based on the existing file system in the operating system, add a file system customized by the user in parallel, increase the user's customization of the specific function of the file system, dynamically load, and facilitate cross-operating system transplantation , to reduce the size of the core, to achieve a variety of special applications.

In the operating system with the file system of the present invention, a compilation switch can be set up in the application program, which is used to provide the choice of whether the command of the application unit enters the standard file system or enters the file system of the present invention. The command of the application unit can enter the file system of the present invention through various methods, including, but not limited to the following ways:

New system calls are added, and these newly added system calls point to the file system described in the present invention. Set up the compilation switch in the application program, when a new type of file system needs to be used, turn on the compilation switch, the operation command of the file system of the application unit points to the newly added system call, and the command of the application unit enters through the newly added system call The file system of the present invention. Such as turning on the compilation switch, so that the operating commands of the file system, such as open, close..., etc. point to the newly added system calls, so that the commands of the application unit enter the file system of the present invention in the operating system kernel through these newly added system calls When it is not necessary to use the file system algorithm based on the file system of the present invention to realize, the compilation switch can be closed, so that the operation command of the application program still points to the entry of the standard file system under Linux, and the operation under the standard file system is performed.

It is also possible to add a new control command word in the system call to point to the entry function of the file system of the present invention in the operating system to call the file system of the present invention. For example, add a new control command word in the original file system call operation function, such as ioctl. Through the compilation switch of the application program, the application layer file operation functions such as open, close... are represented by the corresponding newly added command words in the ioctl, and enter the file described in the present invention in the kernel of the operating system through the original system call system. In the kernel, the command represented by the new command word of the above-mentioned ioctl command is transferred to the entry of the customized physical file system unit in the entry conversion unit.

Obviously, when the file system of the present invention is used in parallel with the standard file system, those skilled in the art can also use other methods to connect the commands of the application unit to the file system of the present invention.

In the file system of the present invention, its management unit 2 can manage and schedule each specific physical file system in the physical file unit, and can be used to realize various self-defined file system functions required by users, such as process scheduling, various Algorithms, memory management, etc., of course, can also be used to achieve some general functions. If the user has no special function requirements corresponding to the management unit for the target file system, the management unit can be canceled. In this way, the user's request command directly enters the physical file system unit through the interface conversion unit of the entry.

In the file system of the present invention, when a user performs cross-operating system platform transplantation on a target file system, the packaging of the original operating system for the physical file system unit can be retained to the greatest extent, greatly reducing the workload during cross-platform transplantation and complexity. The "encapsulation of the physical file system unit" here is just like the "trace" mentioned above. It is a layer of encapsulation added by an operating system outside the specific physical file system in order to maintain its specific functions and operations. Layer encapsulation is just like the VFS unit in the Linux operating system, or the GFM unit in the uITRON operating system.

In the file system of the present invention, when the cross-platform transplantation of the physical file system is carried out, the encapsulation of the physical file system unit by the original operating system as described above can be retained. In order to reduce the workload, only the interface transfer work of the entry is required. , for the target file system that needs to be transplanted, the transplanting work can be successfully completed without a lot of code modification work. However, in order to reduce the size of the kernel and adapt to a system with limited resources, such as an embedded operating system, if the target file system used is a specific target file system, the application unit can directly pass through the entry conversion unit. Commands are transmitted to the management unit or specific target file system, reducing the VFS unit, which can reduce the size of the kernel. In embedded systems with limited system resources, it has great advantages.

In the file system of the present invention, the physical file system unit may include a standard physical file system complying with a standard protocol, and/or a user-customized non-standard physical file system, and may also include an increased physical file system, including standard/non-standard Physical file systems, and user-customized standard/non-standard physical file systems transplanted from other operating systems, etc. Apparently, as required, the physical file system unit of the present invention may include one or more physical file systems mentioned above, and of course may also include other content or parts capable of realizing a certain function, and so on. In this specification, users not only include the end users who use the operating system containing the file system of the present invention, but also include other parties related to the file system of the present invention or the operating system containing the file system of the present invention, including, but not limited to providing file systems and manufacturers and suppliers that provide services such as adding, deleting, cross-platform porting, etc.

In the file system of the present invention, the specific function customization can also be realized by adding the function customization according to the needs in the management unit, such as adding the specific function customization according to the needs, and providing the function to meet the needs.

In the file system of the present invention, the management unit can include several modules that provide general and special functions, and the top drive unit can include several modules with general or special drive functions to support general and special physical file systems in the physical file system unit , to provide general and special transaction processing and to complete the special customization requirements of other functions of users. When the function of a certain item needs to be added or deleted according to user requirements, the physical file system can be introduced into the physical file system unit or deleted from it, and the management unit and the top drive unit can be modified accordingly, such as adding or deleting the corresponding module to modify it. When a certain physical file system needs to be transplanted across operating systems according to user requirements, the physical file system can be transplanted to the physical file system unit of the file system of the present invention, and completed by corresponding modification of the management unit and the top drive unit. Therefore, the file system of the present invention has sufficient flexibility, can add or delete a certain function according to user requirements, and is easy to transplant.

For an operating system with the file system of the present invention, the file system of the present invention can be used as a part of the kernel of the operating system. As the file system of the present invention becomes a part of the operating system kernel through static compilation, it can be automatically loaded each time it is started, and it can also be used as a module by means of a module, which can be dynamically loaded into the operating system when needed. operating system, or dynamically uninstalled from the operating system. When loading, the file system of the present invention intercepts the command sent by the application program to the original file system; after unloading, the original operation process is restored, and the VFS unit of the standard file system accepts the command of the application program.

Next, according to an embodiment of the present invention, namely, by transplanting the target file system customized by oneself from the uITRON operating system to the specific embodiment of the file system of the present invention of the Linux system, the file system of the present invention will be described for other platforms. The transplantation of the target file system that contains more "traces" of the original operating system is easy to implement.

The following first introduces the technical background of the transplantation process. As shown in Figure 3, under uITRON, there are physical file systems, namely UDFM and RFM, and GFM unit. At the same time, in UDFM and RFM, the process scheduling method, data structure, memory allocation method, etc. are According to the user's own customization. That is to say, the target file system to be transplanted here is not a standard VFS file system, but a file system specially customized by the user.

According to the usual transplanting method, it is considered to readjust the inside of the target file system, separate the pure physical file system and the self-customized algorithm unit, and then use the Linux file system to represent the VSF file system The four interface data structures super_block, file, inode, and dentry are used to re-express the interface of the target file system, and then reintegrate the self-customized algorithm into the Linux operating system, which will be very difficult and difficult to complete.

As shown in Figure 3, the file system in the uITRON system includes GFM, UDFM, RFM, RE_RWDRIVE, RE_PRDRIVE, RE_PRWDRIVE, RE_RDRIVE, RE_UPPDRIVE, etc. The standard file system in the Linux system includes VFS, UDF, CDROM.C, etc. structure. At this point we need to adopt the file system described in the present invention. According to the file system of the present invention, each part of the file system in the uITRON operating system is set as a management unit corresponding to the file system of the present invention, a physical file system unit, and a top drive unit, wherein the GFM unit is set to support the file system. A unified management unit for process scheduling, memory allocation, etc. UDFM and RFM are physical file systems that combine specific file systems with specific algorithms. The following RE_RWDRIVE, RE_PRDRIVE, RE_PRWDRIVE, RE_RDRIVE, RE_UPPDRIVE are top drive units, which are used to realize specific drive functions. According to the direction of the process line indicated by the arrow in Figure 3, each part of the file system in the uITRON operating system is connected with the corresponding part in the file system of the present invention, thereby transplanted to the file system of the present invention, as shown on the right side of Figure 4 . At this point, the transplanted target file system is arranged in the manner described in the present invention. It includes I/F change 1, GFM, UDFM & RFM, RE_RWDRIVE & RE_PRDRIVE & RE_PRWDRIVE & RE_RDRIVE & RE_UPPDRIVE, I/F change 2, which respectively correspond to the entry interface conversion unit, management unit, and physical file in the file system of the present invention System unit, top drive unit, export interface conversion unit, like this, the customized content of the user who needs to be transplanted has all been integrated in the file system of the present invention.

After transplantation, the application unit command passes through the entry interface conversion unit 1, and the entry interface conversion unit 1 converts the input command into a format accepted by the management unit 2, and the management unit 2 processes the input command and data, and provides a physical file System support, the physical file system unit includes UDFM and RFM, processing commands and data from the management unit 2, then the commands and data are processed through the top drive unit 4, and then the command request is transmitted to the bottom layer through the interface conversion unit 5 at the exit The drive unit, and then make the drive work according to the set function. Apparently, the management unit 2 and the top drive unit 4 can be modified during migration, so as to be suitable for connection with corresponding parts in the file system of the present invention.

Among them, since the target file system involves calls to C language standard library functions and some functions related to the operating system, after we transplant the target file system, we need to rewrite such related functions.

In the above embodiment, the method of transplanting the target file system as a whole is adopted, which avoids the interface connection of the special data structure according to the needs of VFS, and the process of this interface connection is very difficult, and at the same time, the process does not change the file The internal structure, process, and algorithm of the system, this process only involves the rewriting of the interface part, and the required standard C language library and related parts of the system are added. This process is simple, clear, and with low workload. It can quickly and effectively perform cross-system transplantation, and establish the required target file system in Linux.

At the same time, because the newly transplanted part can be dynamically transplanted in the form of modules, the file system of the present invention has great flexibility, and can be simply added and uninstalled at any time. After uninstalling, there is no impact on the original operating system kernel, especially the original file system.

The present invention also provides a method for constructing and enhancing file system functions, said method comprising the steps of: setting a file system including an entry interface conversion unit, a management unit, a physical file system unit, a top drive unit, and an exit interface conversion unit , the application program unit is connected to the interface conversion unit of the portal, the interface conversion unit of the portal is connected to the management unit, the management unit is connected to the physical file system unit, and the physical file system unit is connected to the top drive unit , the top drive unit is connected to the interface conversion unit of the outlet, and the interface conversion unit of the outlet is connected to the drive unit of the operating system, wherein the interface conversion unit of the inlet is used to convert the input command into the The format accepted by the management unit, the management unit processes the input commands and data, and provides support for the physical file system, and the physical file system unit includes the physical file system for processing commands from the management unit and data, after the command and data are processed by the top drive unit, they are transferred to the interface conversion unit of the outlet, and the interface conversion unit of the outlet converts the command and data from the top drive unit into the drive unit for the operating system Accepted control commands and data that can be recognized by hardware devices, and provide them to the drive unit of the operating system.

In the method of the present invention, the management unit can also be omitted. In this case, the method of the present invention includes setting the file system of the present invention to include an interface conversion unit of the entry, a physical file system unit, a top drive unit, and an interface conversion unit of the exit, wherein, since the management unit is omitted, the entry The interface translation unit is directly connected with the physical file system unit. , the method of the present invention also includes the step of converting the input command through the interface conversion unit of the entry and providing it to the physical file system unit, the physical file system unit includes the physical file system, and processes commands and data from the interface conversion unit of the entry , and provide the processed command and data to the interface conversion unit of the outlet through the top drive unit, and convert the command and data from the top drive unit into the drive unit accepted by the operating system through the interface conversion unit of the outlet. After the control commands and data recognized by the hardware device are provided to the drive unit of the operating system. The method of the present invention omits the management unit in the file system of the present invention is a relatively special case, and the necessary transaction processing can also be provided after being omitted.

The method for constructing and enhancing the file system function of the present invention also includes the step of realizing general and special transaction processing, the step includes: setting a standard physical file system complying with the standard protocol and/or a user-customized file system in the physical file system unit For non-standard physical file systems, the management unit provides general and special functions required to manage and implement standard and/or non-standard physical file systems.

The method for constructing and enhancing the file system function of the present invention also includes the step of realizing the cross-platform transplantation of the physical file system. The step includes: transplanting the physical file system from an operating system to the physical file system unit, modifying the management A unit that provides the general and specific functions needed to manage and implement the physical file system being migrated.

The method for constructing and enhancing file system functions of the present invention also includes the step of accessing a target file system transplanted from another operating system with information related to the original operating system or original hardware device, and the steps further include A step of setting a management unit, a file algorithm implementation unit and a top drive unit corresponding to the file system in the target file system, and connecting with corresponding parts of the file system.

In the file system of the present invention, the cross-operating system transplantation of the target physical file system can also be carried out. By modifying the management unit and the top driver unit, it can provide the passage and special functions of the target physical file system that supports transplantation. Migrate the target physical file system from another operating system to the file system of the present invention.

The following uses an embodiment to illustrate the addition of special functions required by users in the file system of the present invention. In this embodiment, a user-customized function can be added to the file system by adding corresponding functions to the existing physical file system and making corresponding modifications to the management unit and the top drive unit.

Now, how to add the real-time function to the file system of the present invention on the basis of the UDF physical file system.

The protocol specification of UDF2.01 includes two parts, one is the standard part and the other is the additional part. The standard part implements the main algorithm of the UDF file system; and in the additional part, the specification defines some parameters and methods required for real-time operations. In different existing Linux kernel versions, such as the UDF 2.01 version implemented in the Linux 2.4.20 kernel and the UDF 2.5 version implemented in Linux 2.6.1, they are all implementations of the standard part of the protocol. However, the additional part in the protocol is not implemented, that is, the real-time function is not supported.

For the existing Linux file system, although the real-time function specified in the additional part can be realized through efforts, the above-mentioned real-time function does not show good performance because the current Linux file system adopts the VFS file system. performance. In the VFS file system, data process management, scheduling, and memory management are all implemented through several data structures in the VFS, such as inodes. When data is transmitted in the above manner, all transactions are performed by a special process scheduling algorithm in the Linux kernel. However, the Linux kernel is not specially designed for real-time processing, which makes this indirect method cause some processes with high user priority to not be processed the fastest. The real-time function cannot be fully realized. At present, the UDF part in various versions of Linux does not include the real-time function of the extended part of the standard UDF protocol, which is also the root cause.

The file system of the present invention removes the VFS unit and introduces a user-defined user management unit. Therefore, the above-mentioned real-time functions can be well supported through a special design, and the shortcoming of poor support for real-time functions under the standard Linux file system can be solved.

FIG. 5 shows the file system of the present invention under the present embodiment. On the basis of the standard UDF physical file system, the process or method of increasing the real-time function in the file system of the present invention is as follows:

Modify the management unit so that in addition to the conventional related implementations of process scheduling and memory management, it is also necessary to implement the UDF protocol and the user's special customization of real-time performance according to real-time requirements. This particular customization, referred to herein as "Modification 1", is shown in FIG. 5 .

In the UDF physical file system unit, add various file system-related algorithms, protocols, and regulations that specifically implement the UDF real-time function requirements. If the UDF already includes the real-time function, this step can be omitted.

Modify the top drive unit. Here, in addition to the conventional implementation of the drive, it is also necessary to add the implementation of the UDF protocol and the user's special customization for real-time requirements, which is called "Modification 2" in Figure 5.

After the real-time function is added through the above method, the file system of the present invention can provide the real-time function to realize this special transaction processing. The specific process is as follows:

1. The operation request of the application program first enters the input interface conversion unit, where the data and parameters of the operation request will be translated into a command language that the user-defined management unit can understand.

2. Then, enter the management unit, where according to the real-time requirements, in addition to the conventional related implementation of process scheduling and memory management, the UDF protocol and the user's special customization of real-time performance must also be implemented.

3. After that, the data stream enters the UDF physical file system unit, and implements various file system-related algorithms, protocols, and regulations required by UDF.

4. After outputting from the UDF algorithm implementation unit, the data flow enters the top drive unit. Here, in addition to the conventional implementation of the drive, the UDF protocol and the user's special customization for real-time requirements must also be implemented.

5. Finally, enter the export interface conversion unit, where various data and parameters in the data stream are converted into a format that the underlying drive unit can understand, and input into the underlying file system.

In the method of the present invention, "modification 1" is carried out by adding modules such as real-time judgment and real-time processing in the management unit 2 . After the data flow enters the management unit, its process can be carried out as shown in Figure 6:

1. After entering the management unit, first enter the "real-time judging module" 100, and according to the description of the corresponding flag bit, the system judges whether the data flow has real-time requirements.

2. If there is a real-time requirement, enter the "real-time processing module" 102, and customize and realize corresponding real-time functions according to the UDF protocol specification. After this step is complete, go to step 4.

3. If there is no real-time requirement, enter the "general data processing operation module" 101, and proceed to step 4 after performing corresponding processing.

4. The above-mentioned output data flow enters the "other function management module" 103 for processing other functions, and then completes the realization of this part, and enters the algorithm implementation unit of UDF.

In the method of the present invention, "modification 2" is performed by modifying or adding modules such as data stream type or real-time control in the top drive unit 4 . After the data stream enters the top drive unit, its process can be carried out as shown in Figure 7:

1. After the data stream is processed by the UDF physical file system unit, it is output to the top drive unit. Here first judge the type of incoming data flow, this function is realized by the 104 module. According to the three situations of "general data flow", "streaming media data flow" and "real-time data flow", respectively enter the corresponding data processing module for processing. The above three modules are respectively represented as module 105, module 106 and module 107 in the figure.

2. After processing, the data flow enters other control parts in the top drive unit, module 108 in the figure.

3. After all processing by the top drive unit, the data flow enters the outlet interface conversion unit.

In order to make this file system more flexible in actual use, this system can be designed as an installable module. Without recompiling all source files of Linux content, this installable module can be dynamically inserted into the running In the kernel, make it become an organic whole of kernel, thereby realize each special function that the novel file system of the present invention has; Or remove the module that has been installed from kernel, thereby cancel Linux kernel for novel file System and support for various special functions. For example, in Linux, its specific implementation method is realized by calling (module_init) (initializing the new file system) and (module_exit) (launching the new file system). When initializing the new file system, register the new file system; and when launching the new file system, cancel the registered new file system.

In the method for constructing and enhancing the file system function of the present invention, it may also include the step of adding function customization, such as some specific function customization, in the management unit as needed to provide the customized function and realize the function requirement.

The present invention also provides a method for enhancing file system functions by adding or deleting a certain function in the file system of the present invention. For example, in the physical file system unit, add or delete the corresponding physical file system according to the needs or customization, such as the physical file system that provides the function, or add or delete the corresponding part corresponding to the function in the existing physical file system, such as For the part with this function, by modifying the management unit and the top drive unit, such as according to the added physical file system or the corresponding part added in the physical file system, the management unit and the top drive unit are modified accordingly to provide management and realize the added physical file system. The general and special functions and driver functions required by the file system or corresponding parts, and then realize the functions, or modify the management unit and the top drive unit according to the deletion and physical file system or delete the corresponding part from the physical file system, and delete the corresponding The support and driving function of this function are implemented to delete this function from the file system of the present invention.

The file system under an operating system described in the present invention, as well as the operating system with the file system of the present invention and a method for constructing and enhancing file system functions can be used in, but not limited to, the following products and technologies: DVD Video recorders, DVD cameras, hard disk video recorders, hard disk cameras, and products, equipment and technologies related to recording media such as DVDs and hard disks.

Finally, it should be noted that: the above embodiments are only used to illustrate and not limit the technical solutions of the present invention, although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: the present invention can still be modified or Equivalent replacements, any modifications or partial replacements that do not depart from the spirit and scope of the present invention shall fall within the scope of the claims of the present invention.

Claims (20)

1. the file system under the operating system, described file system comprises: the interface conversion unit of inlet, the physical file system unit, top drive unit, the interface conversion unit of outlet, wherein, the interface conversion unit of described inlet connects application program unit and described physical file system unit, described physical file system unit connects the interface conversion unit and the described top drive unit of described inlet, described top drive unit connects the interface conversion unit of described physical file system unit and described outlet, and the interface conversion unit of described outlet connects the driver element of described top drive unit and operating system.

2. the file system under the operating system, described file system comprises: the interface conversion unit of inlet, administrative unit, the physical file system unit, top drive unit, the interface conversion unit of outlet, the interface conversion unit of wherein said inlet connects application program unit and described administrative unit, described administrative unit connects the interface conversion unit and the described physical file system unit of described inlet, described physical file system unit connects described administrative unit and described top drive unit, described top drive unit connects the interface conversion unit of described physical file system unit and described outlet, and the interface conversion unit of described outlet connects the driver element of described top drive unit and operating system.

3. file system as claimed in claim 2, it is characterized in that: the interface conversion unit of described inlet is used for the command conversion of input is become the form of being accepted by described administrative unit, described administrative unit, order and data to input are handled, support to physical file system is provided, described physical file system unit comprises described physical file system, be used to handle order and data from described administrative unit, after order and data are handled via described top drive unit, interface conversion unit to described outlet, control command and data that driver element that the interface conversion unit of described outlet will be converted to the system of being operated from the order and the data of described top drive unit is accepted, and provide it to the driver element of operating system.

4. file system as claimed in claim 3, it is characterized in that described physical file system unit comprises the standard physical file system of deferring to standard agreement at least, the non-standard physical file system of customization, the physical file system that increases, a kind of with from other operating system transplantations physical file system that come, customization, described administrative unit comprises and is used for management and realizes the needed general and specific function of these physical file systems.

5. file system as claimed in claim 3 is characterized in that described administrative unit also comprises the customizing functions that adds as required, provides this function according to the customizing functions that is added.

6. file system as claimed in claim 3, it is characterized in that, described file system also comprises the target file system of coming from another operating system transplantation, have the information relevant with origin operation system or former hardware device, described target file system comprise with the corresponding administrative unit of described file system, physical file system unit and top drive unit partly or entirely.

7. operating system that has as any one described file system of claim 1-6.

8. operating system as claimed in claim 7, further comprise the standard file system that contains VFS, wherein, set up compiler toggle in the application program, be used to provide the order of applying unit to enter standard file system and still enter selection as any one described file system of claim 1-6.

9. operating system as claimed in claim 8, it is characterized in that, open described compiler toggle, the operational order of the file system of applying unit is pointed to the system call that increases newly, the order of applying unit enters any one described file system as claim 1-6 by the described system call that increases newly.

10. operating system as claimed in claim 8, it is characterized in that, increase new control command word in system call, point to the entrance function in the described operating system, call any one described file system as claim 1-6 as any one described file system of claim 1-6.

11. operating system as claimed in claim 7 wherein, as a module, is loaded into described operating system or from the unloading of this operating system as any one described file system of claim 1-6 at any time.

12. operating system as claimed in claim 7 wherein, can be used as the part of operating system nucleus as any one described file system of claim 1-6.

13. a method that makes up and strengthen file system function, described method comprises step:

Step 1, setting comprises the interface conversion unit of inlet, the physical file system unit, top drive unit, file system with the interface conversion unit that exports, wherein, application program unit connects the interface conversion unit of described inlet, the interface conversion unit of described inlet connects described physical file system unit, described physical file system unit connects described top drive unit, described top drive unit connects the interface conversion unit of described outlet, the driver element of the interface conversion unit attended operation system of described outlet, and

Step 2, after changing, the order that will import by the interface conversion unit of described inlet offers described physical file system unit, described physical file system unit comprises physical file system, processing is from the order and the data of the interface conversion unit of described inlet, after order and data are handled via described top drive unit, interface conversion unit to described outlet, the driver element that interface conversion unit by described outlet will be converted to the system of being operated from the order and the data of described top drive unit is accepted, control command and data that hardware device can be discerned, and provide it to the driver element of operating system.

14. a method that makes up and strengthen file system function, described method comprises step:

Step 1, setting comprises the interface conversion unit of inlet, administrative unit, the physical file system unit, top drive unit, file system with the interface conversion unit that exports, wherein, application program unit connects the interface conversion unit of described inlet, and the interface conversion unit of described inlet connects described administrative unit, and described administrative unit connects described physical file system unit, described physical file system unit connects described top drive unit, described top drive unit connects the interface conversion unit of described outlet, the driver element of the interface conversion unit attended operation system of described outlet, and

Step 2, interface conversion unit by described inlet becomes the form of being accepted by described administrative unit with the command conversion of importing, by described administrative unit the order and the data of input are handled, support to physical file system is provided, in described physical file system unit, comprise described physical file system, handle order and data thus from described administrative unit, after order and data are handled via described top drive unit, interface conversion unit to described outlet, the driver element that interface conversion unit by described outlet will be converted to the system of being operated from the order and the data of described top drive unit is accepted, control command and data that hardware device can be discerned, and provide it to the driver element of operating system.

15. as structure as described in the claim 14 and strengthen the method for file system function, it is characterized in that, described method comprises the step that realization is general and particular transaction is handled, described step comprises: defer to the non-standard physical file system of the standard physical file system of standard agreement and/or customization and provide management and realization standard and/or the needed general and specific function of non-standard physical file system by described administrative unit in the unit setting of described physical file system.

16. as structure as described in claim 14 or 15 and strengthen the method for file system function, it is characterized in that described method is included in the described administrative unit and adds customizing functions according to user's needs, realizes the step of this functional requirement.

17. as structure as described in claim 14 or 15 and strengthen the method for file system function, it is characterized in that, described method comprises the step of increase, a certain function of deletion, described step comprises: increase, delete corresponding physical file system or having the corresponding step partly of physical file system increase, deletion and this function in the physical file system unit, revise described administrative unit and top drive unit, the step that provides or delete the function corresponding support and drive function.

18. as structure as described in claim 14 or 15 and strengthen the method for file system function, it is characterized in that, described method comprises the step of the function of cross-platform transplanting customization, described step further comprises: the physical file system that increases the customization of coming from another operating system transplantation in described physical file system unit, revise described administrative unit and top drive unit, management is provided and realizes transplanted described physical file system needed general and specific function and driving function.

19. method as claim 14 or 15 described structures and enhancing file system function, it is characterized in that, the step that described method comprises is that access is come from another operating system transplantation, have the target file system of the information relevant with origin operation system or former hardware device, described step further is included in described target file system setting and the corresponding administrative unit of described file system, physical file system unit and top drive unit, and the appropriate section of described these unit and described file system is carried out step of connecting.

20. the file recording device with the described file system of claim 2 is characterized in that, comprising:

Application program unit, operation adapts to the data of alternative document system; And

Hardware device receives the data from the driver element of described operating system, wherein,

Described ingress interface converting unit will be converted to the data that adapt to described file system operation from the data of the described alternative document of the adaptation of described application program unit system operation;

Described administrative unit is used to control described physical file system unit, and the process of the data that will receive described hardware device is dispatched;

Described physical file system unit is used for handling the data from described administrative unit;

Described top drive unit is used to handle the data from described physical file system unit, and sends it to described discharge coupling converting unit;

Described discharge coupling converting unit will be converted to the data of the described alternative document of adaptation system operation from the data of the described file system operation of the adaptation of described top drive unit, and send it to the driver element of described operating system.

Images