CA2275970C - Object and method for providing efficient multi-user access to shared operating system kernal code using instancing - Google Patents
Object and method for providing efficient multi-user access to shared operating system kernal code using instancing Download PDFInfo
- Publication number
- CA2275970C CA2275970C CA002275970A CA2275970A CA2275970C CA 2275970 C CA2275970 C CA 2275970C CA 002275970 A CA002275970 A CA 002275970A CA 2275970 A CA2275970 A CA 2275970A CA 2275970 C CA2275970 C CA 2275970C
- Authority
- CA
- Canada
- Prior art keywords
- kernal
- user
- instantiation
- kernel
- memory
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
Abstract
A method for improving the performance of a multi-user operating system includes the steps of providing a kernal mode memory region in which instantiations of kernal objects associated with each user may be stored. Storing instantiations of kernal objects reduces code contention for commonly used kernal objects while allowing the instantiation to retain kernal mode protections.
Description
OBJECT AND METHOD FOR PROVIDING
EFFICIENT MULTI-USER ACCESS TO SHARED
OPERATING SYSTEM KERNAL CODE USING INSTANCING
Field of the Invention The present invention relates to multi-user operating systems, in particular, to providing multi-user functionality in single-user operating systems as well as to improving the performance of multi-user operating systems having operating system code objects shared by multiple users.
Background of the Invention An operating system is software code that provides basic functions allowing a user or users to interact with a computer. For example, an operating system provides the user with commands for opening files, closing files, printing files, displaying data objects on a display screen, as well as many other functions.
In a computer system that is designed to service a single user only, a single copy of software code is required to provide these functions. As the user performs actions requiring operating system intervention, the appropriate section of the operating system code services the user request. In a system that services multiple users, however, it becomes unwieldy to provide a separate copy of the operating system code for each user.
This problem has been addressed in some multi-user operating systems by providing only a single copy of certain segments of the operating system code which are shared among many or all of the users. For example, WINDOWS NTTM, an operating system produced and sold by
EFFICIENT MULTI-USER ACCESS TO SHARED
OPERATING SYSTEM KERNAL CODE USING INSTANCING
Field of the Invention The present invention relates to multi-user operating systems, in particular, to providing multi-user functionality in single-user operating systems as well as to improving the performance of multi-user operating systems having operating system code objects shared by multiple users.
Background of the Invention An operating system is software code that provides basic functions allowing a user or users to interact with a computer. For example, an operating system provides the user with commands for opening files, closing files, printing files, displaying data objects on a display screen, as well as many other functions.
In a computer system that is designed to service a single user only, a single copy of software code is required to provide these functions. As the user performs actions requiring operating system intervention, the appropriate section of the operating system code services the user request. In a system that services multiple users, however, it becomes unwieldy to provide a separate copy of the operating system code for each user.
This problem has been addressed in some multi-user operating systems by providing only a single copy of certain segments of the operating system code which are shared among many or all of the users. For example, WINDOWS NTTM, an operating system produced and sold by
-2-Microsoft Corporation of Redmond, Washington, provides only a single copy of the code necessary to implement graphics functions, i.e. display data objects on a display screen.
This approach has the drawback of not truly providing mufti-user access to graphics functions because multiple user accesses would result in data collisions between individual users' data. Further, even if multiple users could simultaneously perform actions requiring invocation of the graphics code without data collisions, users would experience delays due to code contention.
Summary of the Invention The object and method described herein provide a technique that may allow a portion of system address space appear to be unique for each login instance (i.e. per user or per terminal) and may allow a group of processes serving a given logon instance to access the same global kernal data while keeping data associated with the logon instance separate and isolated from the data of other logon instances. The object and method described herein may also allow a mufti-user operating system to be implemented in a relatively small amount of memory space while avoiding the performance delays associated with code contention for single segments of code.
In brief overview, a special kernal mode memory region that instances data for each user or terminal is created. When a kernal mode module or device driver is loaded by the system, an address for that module may be globally assigned in the same manner as a single instance system, i.e. a single user system. The memory management data structures are subsequently modified to provide a separate memory allocation for any portions of the module to which data can be written. A specific instance is shared among a group of processes associated with a unique logon occurrence. The module may be automatically removed when the last process associated with the logon occurrence terminates.
-2a-According to one aspect of the present invention, in a computer system servicing at least a first user and a second user and including an operating system having a kernal memory space for storing protected code, an apparatus for improving the performance of the system is provided. This apparatus comprises a kernal mode memory region of kernal memory space reserved for storing instantiations of kernal objects, a first instantiation of a kernal object stored in the kernal mode memory region associated with the first user of the system, and a second instantiation of the kernal object stored in the kernal memory mode region associated with a second user of the system. Each instantiation of the kernal object has the same virtual address.
According to another aspect of the present invention, there is provided a method for providing a first user and a second user concurrent access to shared, writeable operating system kernal code. The method includes reserving a first region of kernal mode memory space for the first user and a second region of kernal mode memory space for the second user, creating a first instantiation of a computer kernal object associated with the first user 1 S of the computer system, creating a second instantiation of a computer kernal object associated with the second user of the computer system, storing the first kernal object instantiation in the first kernal mode memory region and the second kernal object instantiation in the second kernal mode memory region, and accessing the first kernal object instantiation and the second kernal object instantiation using the same virtual address.
According to yet another aspect of the present invention, an article of manufacture having computer-readable program code means for providing a first user and a second user concurrent access to shared, writeable operating system kernal code, is provided. This article of manufacture comprises a computer-readable program code means for reserving a -2b-first region of kernal mode memory space for the first user and a second region of kernal mode memory space for the second user, a computer-readable program code means for creating a first instantiation of a computer kernal object associated with the first user of the computer system, a computer-readable program code means for creating a second instantiation of a computer kernal object associated with the second user of the computer system, and a computer-readable program code means for storing the first kernal object instantiation in the first kernal mode memory region and the second kernal object instantiation in the second kernal mode memory region and accessing the first kernal object instantiation and the second kernal object instantiation by the same virtual address.
According to yet another aspect of the present invention, in a computer system servicing at least a first user and a second user and including an operating system having a kernal memory space for storing protected code, an apparatus for improving the performance of the system is provided. This apparatus comprises a kernal mode memory region of the kernal memory space reserved for storing instantiations of kernal objects, a first instantiation of a kernal object stored in a first range of physical memory addresses in the kernal mode memory region and associated with the first user of the system, a second instantiation of the kernal object stored in a second range of physical memory addresses in the kernal mode memory region and associated with the second user of the system, and a page table for each user of the system, wherein each page table stores a mapping between a virtual memory address that is common for each instantiation of the kernal object and each range of physical -2c-memory addresses in the kernal mode memory region occupied by each instantiation of the kernal object.
According to yet another aspect of the present invention, there is provided a method for providing a first user and a second user of a computer system concurrent access to shared, writeable operating system kernal code. The method includes allocating a first region of kernal mode memory space for the first user and a second region of kernal mode memory space for the second user, forming a first instantiation of a computer kernal object associated with the first user of the computer system, forming a second instantiation of the computer kernal object associated with the second user of the computer system, storing the first kernal object instantiation in the first kernal mode memory region and the second kernal object instantiation in the second kernal mode memory region, and accessing a data structure associated with each user of the computer system for obtaining a mapping between a virtual memory address that is common for each instantiation of the computer kernal object and a physical memory address of the first and second kernal object instantiations in the kernal mode memory space.
This approach has the drawback of not truly providing mufti-user access to graphics functions because multiple user accesses would result in data collisions between individual users' data. Further, even if multiple users could simultaneously perform actions requiring invocation of the graphics code without data collisions, users would experience delays due to code contention.
Summary of the Invention The object and method described herein provide a technique that may allow a portion of system address space appear to be unique for each login instance (i.e. per user or per terminal) and may allow a group of processes serving a given logon instance to access the same global kernal data while keeping data associated with the logon instance separate and isolated from the data of other logon instances. The object and method described herein may also allow a mufti-user operating system to be implemented in a relatively small amount of memory space while avoiding the performance delays associated with code contention for single segments of code.
In brief overview, a special kernal mode memory region that instances data for each user or terminal is created. When a kernal mode module or device driver is loaded by the system, an address for that module may be globally assigned in the same manner as a single instance system, i.e. a single user system. The memory management data structures are subsequently modified to provide a separate memory allocation for any portions of the module to which data can be written. A specific instance is shared among a group of processes associated with a unique logon occurrence. The module may be automatically removed when the last process associated with the logon occurrence terminates.
-2a-According to one aspect of the present invention, in a computer system servicing at least a first user and a second user and including an operating system having a kernal memory space for storing protected code, an apparatus for improving the performance of the system is provided. This apparatus comprises a kernal mode memory region of kernal memory space reserved for storing instantiations of kernal objects, a first instantiation of a kernal object stored in the kernal mode memory region associated with the first user of the system, and a second instantiation of the kernal object stored in the kernal memory mode region associated with a second user of the system. Each instantiation of the kernal object has the same virtual address.
According to another aspect of the present invention, there is provided a method for providing a first user and a second user concurrent access to shared, writeable operating system kernal code. The method includes reserving a first region of kernal mode memory space for the first user and a second region of kernal mode memory space for the second user, creating a first instantiation of a computer kernal object associated with the first user 1 S of the computer system, creating a second instantiation of a computer kernal object associated with the second user of the computer system, storing the first kernal object instantiation in the first kernal mode memory region and the second kernal object instantiation in the second kernal mode memory region, and accessing the first kernal object instantiation and the second kernal object instantiation using the same virtual address.
According to yet another aspect of the present invention, an article of manufacture having computer-readable program code means for providing a first user and a second user concurrent access to shared, writeable operating system kernal code, is provided. This article of manufacture comprises a computer-readable program code means for reserving a -2b-first region of kernal mode memory space for the first user and a second region of kernal mode memory space for the second user, a computer-readable program code means for creating a first instantiation of a computer kernal object associated with the first user of the computer system, a computer-readable program code means for creating a second instantiation of a computer kernal object associated with the second user of the computer system, and a computer-readable program code means for storing the first kernal object instantiation in the first kernal mode memory region and the second kernal object instantiation in the second kernal mode memory region and accessing the first kernal object instantiation and the second kernal object instantiation by the same virtual address.
According to yet another aspect of the present invention, in a computer system servicing at least a first user and a second user and including an operating system having a kernal memory space for storing protected code, an apparatus for improving the performance of the system is provided. This apparatus comprises a kernal mode memory region of the kernal memory space reserved for storing instantiations of kernal objects, a first instantiation of a kernal object stored in a first range of physical memory addresses in the kernal mode memory region and associated with the first user of the system, a second instantiation of the kernal object stored in a second range of physical memory addresses in the kernal mode memory region and associated with the second user of the system, and a page table for each user of the system, wherein each page table stores a mapping between a virtual memory address that is common for each instantiation of the kernal object and each range of physical -2c-memory addresses in the kernal mode memory region occupied by each instantiation of the kernal object.
According to yet another aspect of the present invention, there is provided a method for providing a first user and a second user of a computer system concurrent access to shared, writeable operating system kernal code. The method includes allocating a first region of kernal mode memory space for the first user and a second region of kernal mode memory space for the second user, forming a first instantiation of a computer kernal object associated with the first user of the computer system, forming a second instantiation of the computer kernal object associated with the second user of the computer system, storing the first kernal object instantiation in the first kernal mode memory region and the second kernal object instantiation in the second kernal mode memory region, and accessing a data structure associated with each user of the computer system for obtaining a mapping between a virtual memory address that is common for each instantiation of the computer kernal object and a physical memory address of the first and second kernal object instantiations in the kernal mode memory space.
- 3 -.
Brief Description of the Drawines The invention is pointed with particularity in the appended claims. The above and further advantages of this invention may be better understood by referring to the following description taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a block diagram of the virtual memory space of a WINDOWS NT~
process;
FIG. 2 is a block diagram showing the mapping between the virtual memory space and a physical memory space;
FIG. 3 is a block diagram showing the virtual memory space of a WINDOWS NT~
process having a special kernal memory mode region set aside;
FIG. 4 is a block diagram showing two processes which map the same operating system address to separate, instanced copies of the operating system code; and FIG. 5 is a block diagram showing two processes which map the same operating system address to the same instanced copy of the operating system code.
Detailed Description of the Invention Referring now to FIG. 1, the virtual memory space 10 for a WINDOWS NTTT~
process is shown. In general, user-mode memory spaces 14 are assigned on a per-process basis, and virtual system memory space 12 is shared by all processes. Each virtual memory space 10 is segregated into system memory 12 (also referred to throughout the specification as kernal space) and user memory 14 (also referred to throughout the specification as user space).
User memory 14 is a set of paged memory addresses to which user and kernal threads may write. System memory 12, on the other hand, is a set of protected memory addresses. Only kernal threads may write to system memory 12. If a. user thread desires to write to system
Brief Description of the Drawines The invention is pointed with particularity in the appended claims. The above and further advantages of this invention may be better understood by referring to the following description taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a block diagram of the virtual memory space of a WINDOWS NT~
process;
FIG. 2 is a block diagram showing the mapping between the virtual memory space and a physical memory space;
FIG. 3 is a block diagram showing the virtual memory space of a WINDOWS NT~
process having a special kernal memory mode region set aside;
FIG. 4 is a block diagram showing two processes which map the same operating system address to separate, instanced copies of the operating system code; and FIG. 5 is a block diagram showing two processes which map the same operating system address to the same instanced copy of the operating system code.
Detailed Description of the Invention Referring now to FIG. 1, the virtual memory space 10 for a WINDOWS NTTT~
process is shown. In general, user-mode memory spaces 14 are assigned on a per-process basis, and virtual system memory space 12 is shared by all processes. Each virtual memory space 10 is segregated into system memory 12 (also referred to throughout the specification as kernal space) and user memory 14 (also referred to throughout the specification as user space).
User memory 14 is a set of paged memory addresses to which user and kernal threads may write. System memory 12, on the other hand, is a set of protected memory addresses. Only kernal threads may write to system memory 12. If a. user thread desires to write to system
-4-memory 12, it cannot do so directly. Instead, it requests a specific system service which is system code. The operating system temporarily takes control of the user thread, giving the user thread kernal permissions. This allows the thread to write to system memory. When the thread completes its system operation, it returns to user mode. FIG. 1 shows one embodiment of a virtual memory space in which system memory 12 is further subdivided into nonpaged memory addresses 16, paged memory addresses 18, and directly mapped addresses 20.
In the embodiment shown in FIG. l, nonpaged system memory addresses 16 are reserved for system code which is not allowed to be paged out of memory to disk, for example, the system code which handles virtual memory paging may not be paged to disk. Directly mapped addresses 20 represent a range of system addresses that are directly mapped or translated by hardware, e.g.
masking off one or more bits of the address or by subtracting a predetermined constant from the virtual address to generate the physical address. Directly mapped addresses 20 are typically reserved for system code that must execute with minimum delay, for example, the system code which schedules and dispatches execution of threads.
The remainder of system memory in FIG. 1 is paged system memory 18. Paged system memory 18 address ranges are used for all other aspects of system code, e.g.
graphics code Referring to FIG. 2, when a user space process requests operating system code that resides in paged system memory 18 but is not memory-resident, a page table look-up occurs. A page table look-up begins by accessing the appropriate entry, based on the address 36 requested, of the page directory 30 associated with the process. Individual page directories 30 are kept for each process and store mappings to page tables 34. Page tables 34 are kept for each user/terminal and store translations between the user's virtual memory space 10 and physical memory space 40.
In the embodiment shown in FIG. l, nonpaged system memory addresses 16 are reserved for system code which is not allowed to be paged out of memory to disk, for example, the system code which handles virtual memory paging may not be paged to disk. Directly mapped addresses 20 represent a range of system addresses that are directly mapped or translated by hardware, e.g.
masking off one or more bits of the address or by subtracting a predetermined constant from the virtual address to generate the physical address. Directly mapped addresses 20 are typically reserved for system code that must execute with minimum delay, for example, the system code which schedules and dispatches execution of threads.
The remainder of system memory in FIG. 1 is paged system memory 18. Paged system memory 18 address ranges are used for all other aspects of system code, e.g.
graphics code Referring to FIG. 2, when a user space process requests operating system code that resides in paged system memory 18 but is not memory-resident, a page table look-up occurs. A page table look-up begins by accessing the appropriate entry, based on the address 36 requested, of the page directory 30 associated with the process. Individual page directories 30 are kept for each process and store mappings to page tables 34. Page tables 34 are kept for each user/terminal and store translations between the user's virtual memory space 10 and physical memory space 40.
-5--Accordingly, a user/terminal may have multiple page directories 30 but only one page table 34.
The page table 34 may store a translation for every user page or it may store only the most recently used translations.
For example, an operating system may examine only the first n bits of an address to determine which page directory entry 32 the process should access. The page directory entry 32 is accessed and returns the address of a particular page table 34 to use. The page table 34 maps the virtual address 36, i.e. the virtual page requested by the user process, to a page 38 in physical memory 40. That is, the page table 34 stores the address of the physical page 38 in memory associated with that virtual address 36.
Although each process (i.e., each user virtual memory space 10) is provided with its own page directory 30, current operating systems typically provide only one copy of paged operating system code to service user requests. Accordingly, code contention or data collisions occur when multiple users attempt to access the same portion of operating system code at the same time.
For example, the graphics code is operating; system code which implements the "desktop"
with which individual users interact and allows applications to display windows and data within those windows. WINDOWS NTTM provides only one instance of kernal graphics code, which means that only a single user/desktop can be serviced at a time.
Referring now to FIG. 3, code contention is alleviated in systems such as those described above by providing a special kernal mode memory region 22 which allows operating system code to be instanced for each user/terminal, that is, operating system code may instanced on a per process, or per group of processes, basis. The special region 22 may be set aside at system start-up time, or the region 22 may be dynamically set a:>ide based on the occurrence of a
The page table 34 may store a translation for every user page or it may store only the most recently used translations.
For example, an operating system may examine only the first n bits of an address to determine which page directory entry 32 the process should access. The page directory entry 32 is accessed and returns the address of a particular page table 34 to use. The page table 34 maps the virtual address 36, i.e. the virtual page requested by the user process, to a page 38 in physical memory 40. That is, the page table 34 stores the address of the physical page 38 in memory associated with that virtual address 36.
Although each process (i.e., each user virtual memory space 10) is provided with its own page directory 30, current operating systems typically provide only one copy of paged operating system code to service user requests. Accordingly, code contention or data collisions occur when multiple users attempt to access the same portion of operating system code at the same time.
For example, the graphics code is operating; system code which implements the "desktop"
with which individual users interact and allows applications to display windows and data within those windows. WINDOWS NTTM provides only one instance of kernal graphics code, which means that only a single user/desktop can be serviced at a time.
Referring now to FIG. 3, code contention is alleviated in systems such as those described above by providing a special kernal mode memory region 22 which allows operating system code to be instanced for each user/terminal, that is, operating system code may instanced on a per process, or per group of processes, basis. The special region 22 may be set aside at system start-up time, or the region 22 may be dynamically set a:>ide based on the occurrence of a
6 PCT/US98/00105 predetermined event, e.g. the system could set aside the special kernal mode memory region 22 once a second user logs onto the system. The special kernal mode memory region 22 is used for paged operating system code that may be invoked by multiple processes.
As noted above, when a user process requests shared operating system code and if the code is not memory-resident, a page table look-up must occur. The page directory provides an address into the appropriate page table which maps the special kernal mode memory region 22 to physical memory. Physical memory contains an instantiation of the requested operating system code. By appropriately manipulating the entries in the page directory table 20, groups of processes are able to share operating system code instantiations. Individual processes may have page directory entries associated with particular operating system code segments each set to the same page table 30. The page table 30 will map the special kernal mode memory region 22 associated with process to physical memory containing an instantiation of operating system code.
For example, a client node requests the operating system to draw a rectangle on the display screen associated with its terminal. The client node makes its request by, for example, making a call to a dynamically loaded library (DLL) of operating system functions. The operating system accesses the paged operating system memory address corresponding to the requested DLL
module. If that module is memory-resident, no page fault occurs and the rectangle is drawn.
However, if the module is not memory-resident then a page table lookup as described above will occur, the virtual page will paged into physical memory from disk, and then the rectangle will be drawn.
For example, FIG. 4 depicts two processes, which may or may not be associated with the same user/terminal, accessing the same operating system code instantiation.
The virtual memory space 10, 10' of the two processes have page directories 20, 20' and page tables 30, 30'. The intermediate page directories 20, 20' and page tables 30, 30' point to the same instantiation of operating system code 46 stored in physical memory 40. Each process uses the same virtual memory address 44 to access the operating system code instantiation 42 stored in physical memory 40, and each process is able to operate on the same operating system code instantiation.
FIG. 4 depicts normal operation for the sysl:em and method so long as one of the processes does not request that a system write occur. Sharing the same operating system object instantiation between processes minimizes the physical memory which is consumed by the operating system and, since the processes are not writing to the instantiations, data collisions do not occur and code contentions are minimized. However, if a process requests that a write occur to the system object instantiation, a private instantiation must be created for that process. Any method of detecting writes to the system object insl:antiation can be used so long as the write is detected before it occurs.
FIG. S shows the virtual memory to physical memory mapping that exists after a process requests that a write occur to the system code. The: virtual memory space 10, 10' of each process has page directories 20, 20' and page tables 30, 30' pointing to private instantiations of operating system graphics code 42, 42' stored in physical memory 40. Each process continues to access its own graphics code instantiation using the same virtual address 44, but the intermediate page directories 20, 20' and page tables 30, 30' direct those requests to different instantiations of the operating system code 42, 42' stored in physical memory 40. Once a separate instantiation of operating system code is created for a process associated with a user/terminal, all the processes associated with that user/terminal may then access the private instantiation.
This allows all _g_ processes associated with a user/terminal to access the most recent system data associated with its user/terminal.
The redirection of operating system accesses described above may be achieved by providing the operating system with an additional data structure. The data structure stores information about the various operating system code instantiations present in the system. In one embodiment, the data structure includes an identification of the terminal associated with the process, a list of the processes sharing the terminal, a list of the kernal mode modules loaded in the special memory region (name and address), and which page table entries leave been instanced.
Whenever operating system code is accessed, the described data structure can be accessed to determine whether a new instantiation needs to be created or whether an instantiation shared by the process already exist. For example, if a process accesses graphics code and no page fault occurs, the memory access occurs normally as described above. If a page fault occurs, however, the data structure described above is first accessed to determine if the graphics code has been loaded by any other process, regardless of the user/terminal with which it is associated. If it has been loaded. the page directory entries and page table entries for that virtual address are set to direct accesses to the shared instantiation in physical memory. If the process then attempts to write to operating code memory, the write is trapped, a new instantiation of the kernal object 42' is created in physical memory 40, and the page directory entries and page table entries associated with that virtual address are changed to reflect the new physical location of the private instantiation 42'. Once this change is made, all processes associated with that user/terminal will automatically access the private instantiation because only page table per user/terminal is provided. Alternatively, instead of providing an additional, new data structure, an already existing - 9 .-memory management data structure present in the system may be altered to provide similar information.
The method and apparatus of the present invention may be provided as an article of manufacture having computer-readable program means embodied thereon. The article can be any form of computer-readable media, including hardware, e.g. RAM chip, ROM chip, or PROM
chip, as well as traditional form of software distribution, e.g. floppy disk, hard disk, optical disk, magneto-optical disk, or CD-ROM. The computer-readable program means embodied thereon are any form of program means that provide the functions described above, whether as a single monolithic program or as functional modules. The program means may be provided as object code or they may be provided as source code writtE.n in C, C++~ PASCAL, LISP, FORTRAN or some other suitable programming language.
Having described certain embodiments of the invention, it will now become apparent to one of ordinary skill in the art that other embodiments incorporating the concepts of the invention may be used. Therefore, the invention should not be limited to certain embodiments, but rather should be limited only by the spirit and scope of the: following claims.
As noted above, when a user process requests shared operating system code and if the code is not memory-resident, a page table look-up must occur. The page directory provides an address into the appropriate page table which maps the special kernal mode memory region 22 to physical memory. Physical memory contains an instantiation of the requested operating system code. By appropriately manipulating the entries in the page directory table 20, groups of processes are able to share operating system code instantiations. Individual processes may have page directory entries associated with particular operating system code segments each set to the same page table 30. The page table 30 will map the special kernal mode memory region 22 associated with process to physical memory containing an instantiation of operating system code.
For example, a client node requests the operating system to draw a rectangle on the display screen associated with its terminal. The client node makes its request by, for example, making a call to a dynamically loaded library (DLL) of operating system functions. The operating system accesses the paged operating system memory address corresponding to the requested DLL
module. If that module is memory-resident, no page fault occurs and the rectangle is drawn.
However, if the module is not memory-resident then a page table lookup as described above will occur, the virtual page will paged into physical memory from disk, and then the rectangle will be drawn.
For example, FIG. 4 depicts two processes, which may or may not be associated with the same user/terminal, accessing the same operating system code instantiation.
The virtual memory space 10, 10' of the two processes have page directories 20, 20' and page tables 30, 30'. The intermediate page directories 20, 20' and page tables 30, 30' point to the same instantiation of operating system code 46 stored in physical memory 40. Each process uses the same virtual memory address 44 to access the operating system code instantiation 42 stored in physical memory 40, and each process is able to operate on the same operating system code instantiation.
FIG. 4 depicts normal operation for the sysl:em and method so long as one of the processes does not request that a system write occur. Sharing the same operating system object instantiation between processes minimizes the physical memory which is consumed by the operating system and, since the processes are not writing to the instantiations, data collisions do not occur and code contentions are minimized. However, if a process requests that a write occur to the system object instantiation, a private instantiation must be created for that process. Any method of detecting writes to the system object insl:antiation can be used so long as the write is detected before it occurs.
FIG. S shows the virtual memory to physical memory mapping that exists after a process requests that a write occur to the system code. The: virtual memory space 10, 10' of each process has page directories 20, 20' and page tables 30, 30' pointing to private instantiations of operating system graphics code 42, 42' stored in physical memory 40. Each process continues to access its own graphics code instantiation using the same virtual address 44, but the intermediate page directories 20, 20' and page tables 30, 30' direct those requests to different instantiations of the operating system code 42, 42' stored in physical memory 40. Once a separate instantiation of operating system code is created for a process associated with a user/terminal, all the processes associated with that user/terminal may then access the private instantiation.
This allows all _g_ processes associated with a user/terminal to access the most recent system data associated with its user/terminal.
The redirection of operating system accesses described above may be achieved by providing the operating system with an additional data structure. The data structure stores information about the various operating system code instantiations present in the system. In one embodiment, the data structure includes an identification of the terminal associated with the process, a list of the processes sharing the terminal, a list of the kernal mode modules loaded in the special memory region (name and address), and which page table entries leave been instanced.
Whenever operating system code is accessed, the described data structure can be accessed to determine whether a new instantiation needs to be created or whether an instantiation shared by the process already exist. For example, if a process accesses graphics code and no page fault occurs, the memory access occurs normally as described above. If a page fault occurs, however, the data structure described above is first accessed to determine if the graphics code has been loaded by any other process, regardless of the user/terminal with which it is associated. If it has been loaded. the page directory entries and page table entries for that virtual address are set to direct accesses to the shared instantiation in physical memory. If the process then attempts to write to operating code memory, the write is trapped, a new instantiation of the kernal object 42' is created in physical memory 40, and the page directory entries and page table entries associated with that virtual address are changed to reflect the new physical location of the private instantiation 42'. Once this change is made, all processes associated with that user/terminal will automatically access the private instantiation because only page table per user/terminal is provided. Alternatively, instead of providing an additional, new data structure, an already existing - 9 .-memory management data structure present in the system may be altered to provide similar information.
The method and apparatus of the present invention may be provided as an article of manufacture having computer-readable program means embodied thereon. The article can be any form of computer-readable media, including hardware, e.g. RAM chip, ROM chip, or PROM
chip, as well as traditional form of software distribution, e.g. floppy disk, hard disk, optical disk, magneto-optical disk, or CD-ROM. The computer-readable program means embodied thereon are any form of program means that provide the functions described above, whether as a single monolithic program or as functional modules. The program means may be provided as object code or they may be provided as source code writtE.n in C, C++~ PASCAL, LISP, FORTRAN or some other suitable programming language.
Having described certain embodiments of the invention, it will now become apparent to one of ordinary skill in the art that other embodiments incorporating the concepts of the invention may be used. Therefore, the invention should not be limited to certain embodiments, but rather should be limited only by the spirit and scope of the: following claims.
Claims (24)
PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. In a computer system servicing at least a first user and a second user and including an operating system having a kernal memory space for storing protected code, an apparatus for improving the performance of the system comprising:
a kernal mode memory region of kernal memory space reserved for storing instantiations of kernal objects;
a first instantiation of a kernal object stored in said kernal mode memory region associated with the first user of the system; and a second instantiation of the kernal object stored in said kernal memory mode region associated with a second user of the system, wherein each instantiation of said kernal object has the same virtual address.
a kernal mode memory region of kernal memory space reserved for storing instantiations of kernal objects;
a first instantiation of a kernal object stored in said kernal mode memory region associated with the first user of the system; and a second instantiation of the kernal object stored in said kernal memory mode region associated with a second user of the system, wherein each instantiation of said kernal object has the same virtual address.
2. The apparatus of claim 1 further comprising a memory-management data structure providing a separate allocation of memory for each instantiation of said kernal object.
3. The apparatus of claim 1 wherein said kernal object is a graphics subsystem.
4. The apparatus of claim 1 wherein an instantiation of said kernal object is shared by a plurality of processes, each of said plurality of processes associated with a unique login occurrence.
A method for providing a first user and a second user concurrent access to shared, writeable operating system kernal code, the method comprising:
(a) reserving a first region of kernal mode memory space for the first user and a second region of kernal mode memory space for the second user;
(b) creating a first instantiation of a computer kernal object associated with the first user of the computer system;
(c) creating a second instantiation of a computer kernal object associated with the second user of the computer system;
(d) storing the first kernal object instantiation in the first kernal mode memory region and the second kernal object instantiation in the second kernal mode memory region; and (e) accessing the first kernal object instantiation and the second kernal object instantiation using the same virtual address.
(a) reserving a first region of kernal mode memory space for the first user and a second region of kernal mode memory space for the second user;
(b) creating a first instantiation of a computer kernal object associated with the first user of the computer system;
(c) creating a second instantiation of a computer kernal object associated with the second user of the computer system;
(d) storing the first kernal object instantiation in the first kernal mode memory region and the second kernal object instantiation in the second kernal mode memory region; and (e) accessing the first kernal object instantiation and the second kernal object instantiation using the same virtual address.
6. The method of claim 5 further comprising the step of:
(f) removing the first kernal object instantiation when the first user terminates use of the computer system.
(f) removing the first kernal object instantiation when the first user terminates use of the computer system.
7. The method of claim 5 further comprising the step of modifying memory management data structures to provide a separate memory allocation for each of the kernal object instantiations.
8. The method of claim 7 wherein an entry in a first page table is modified to map a first virtual operating system address to the first kernal object instantiation.
9. An article of manufacture having computer-readable program code means embodied thereon for providing a first user and a second user concurrent access to shared, writeable operating system kernal code, the article comprising:
computer-readable program code means for reserving a first region of kernal mode memory space for the first user and a second region of kernal mode memory space for the second user;
computer-readable program code means for creating a first instantiation of a computer kernal object associated with the first user of the computer system;
computer-readable program code means for creating a second instantiation of a computer kernal object associated with the second user of the computer system;
and computer-readable program code means for storing the first kernal object instantiation in the first kernal mode memory region and the second kernal object instantiation in the second kernal mode memory region and accessing the first kernal object instantiation and the second kernal object instantiation by the same virtual address.
computer-readable program code means for reserving a first region of kernal mode memory space for the first user and a second region of kernal mode memory space for the second user;
computer-readable program code means for creating a first instantiation of a computer kernal object associated with the first user of the computer system;
computer-readable program code means for creating a second instantiation of a computer kernal object associated with the second user of the computer system;
and computer-readable program code means for storing the first kernal object instantiation in the first kernal mode memory region and the second kernal object instantiation in the second kernal mode memory region and accessing the first kernal object instantiation and the second kernal object instantiation by the same virtual address.
10. The article of claim 9 further comprising computer-readable program code means for removing the first kernal object instantiation when the first user terminates use of the computer system.
11. The article of claim 9 further comprising computer-readable program code means for modifying memory management data structures to provide a separate memory allocation for each of the kernal object instantiations.
12. The article of claim 11 wherein an entry in a first page table is modified to map a first virtual operating system address to the first kernel object instantiation.
13. In a computer system servicing at least a first user and a second user and including an operating system having a kernel memory space for storing protected code, an apparatus for improving the performance of the system comprising:
a kernel mode memory region of the kernel memory space reserved for storing instantiations of kernel objects;
a first instantiation of a kernel object stored in a first range of physical memory addresses in the kernel mode memory region and associated with the first user of the system;
a second instantiation of the kernel object stored in a second range of physical memory addresses in the kernel mode memory region and associated with the second user of the system; and a page table for each user of the system, wherein each page table stores a mapping between a virtual memory address that is common for each instantiation of the kernel object and each range of physical memory addresses in the kernel mode memory region occupied by each instantiation of the kernel object.
a kernel mode memory region of the kernel memory space reserved for storing instantiations of kernel objects;
a first instantiation of a kernel object stored in a first range of physical memory addresses in the kernel mode memory region and associated with the first user of the system;
a second instantiation of the kernel object stored in a second range of physical memory addresses in the kernel mode memory region and associated with the second user of the system; and a page table for each user of the system, wherein each page table stores a mapping between a virtual memory address that is common for each instantiation of the kernel object and each range of physical memory addresses in the kernel mode memory region occupied by each instantiation of the kernel object.
14. The apparatus of claim 13 further comprising a page directory that stores a mapping of the page table for the first user and at least one process associated with the first user.
15. The apparatus of claim 13 wherein the first and second ranges of physical memory addresses in the kernel mode memory region are identical when a plurality of users access but do not attempt to write to that memory location.
16. The apparatus of claim 13 wherein the first and second ranges of physical memory addresses in the kernel mode memory region differ when a plurality of users attempt to write to the first range of physical memory addresses.
17. The apparatus of claim 13 comprises a data structure storing data that indicates whether the first range of physical memory addresses in the kernel mode memory region can be shared by a plurality of instantiations of the kernel object.
18. The apparatus of claim 13 wherein the first and second ranges of physical memory addresses in the kernel mode memory region are dynamically reserved upon the occurrence of a predetermined event.
19. The apparatus of claim 18 wherein the predetermined event is when the second user logs onto the computer system.
20. The apparatus of claim 18 wherein the predetermined event corresponds to a start-up sequence of the computer system.
21. A method for providing a first user and a second user of a computer system concurrent access to shared, writeable operating system kernel code, the method comprising:
(a) allocating a first region of kernel mode memory space for the first user and a second region of kernel mode memory space for the second user;
(b) forming a first instantiation of a computer kernel object associated with the first user of the computer system;
(c) forming a second instantiation of the computer kernel object associated with the second user of the computer system;
(d) storing the first kernel object instantiation in the first kernel mode memory region and the second kernel object instantiation in the second kernel mode memory region; and (e) accessing a data structure associated with each user of the computer system for obtaining a mapping between a virtual memory address that is common for each instantiation of the computer kernel object and a physical memory address of the first and second kernel object instantiations in the kernel mode memory space.
(a) allocating a first region of kernel mode memory space for the first user and a second region of kernel mode memory space for the second user;
(b) forming a first instantiation of a computer kernel object associated with the first user of the computer system;
(c) forming a second instantiation of the computer kernel object associated with the second user of the computer system;
(d) storing the first kernel object instantiation in the first kernel mode memory region and the second kernel object instantiation in the second kernel mode memory region; and (e) accessing a data structure associated with each user of the computer system for obtaining a mapping between a virtual memory address that is common for each instantiation of the computer kernel object and a physical memory address of the first and second kernel object instantiations in the kernel mode memory space.
22. The method of claim 21 comprising the step of accessing a page directory to obtain a location of the data structure associated with the first user when executing at least one process associated with the first user.
23. The method of claim 21 further comprising the step of determining from the data structure whether the physical memory address associated with the first kernel object instantiation can be accessed by a plurality of other kernel object instantiations.
24. The method of claim 21 comprising the step of dynamically allocating the first and second regions of kernel mode memory space in response to the occurrence of a predetermined event.
Applications Claiming Priority (5)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US3468997P | 1997-01-07 | 1997-01-07 | |
| US60/034,689 | 1997-01-07 | ||
| US08/813,819 US5913230A (en) | 1997-01-07 | 1997-03-06 | Object and method for providing efficient multi-user access to shared operating system kernal code using instancing |
| US08/813,819 | 1997-03-06 | ||
| PCT/US1998/000105 WO1998030956A1 (en) | 1997-01-07 | 1998-01-07 | Object and method for providing efficient multi-user access to shared operating system kernal code using instancing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2275970A1 CA2275970A1 (en) | 1998-07-16 |
| CA2275970C true CA2275970C (en) | 2002-04-30 |
Family
ID=26711255
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002275970A Expired - Lifetime CA2275970C (en) | 1997-01-07 | 1998-01-07 | Object and method for providing efficient multi-user access to shared operating system kernal code using instancing |
Country Status (10)
| Country | Link |
|---|---|
| US (2) | US5913230A (en) |
| EP (1) | EP0951675B1 (en) |
| JP (1) | JP2001508212A (en) |
| KR (1) | KR20000069891A (en) |
| AU (1) | AU734996B2 (en) |
| CA (1) | CA2275970C (en) |
| DE (1) | DE69819686T2 (en) |
| ES (1) | ES2210714T3 (en) |
| IL (1) | IL130730A0 (en) |
| WO (1) | WO1998030956A1 (en) |
Families Citing this family (46)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6804766B1 (en) * | 1997-11-12 | 2004-10-12 | Hewlett-Packard Development Company, L.P. | Method for managing pages of a designated memory object according to selected memory management policies |
| US6032240A (en) * | 1997-11-12 | 2000-02-29 | Intergraph Corporation | Bypassing a nonpaged pool controller when accessing a remainder portion of a random access memory |
| US6367072B1 (en) * | 1998-03-12 | 2002-04-02 | Applica Systems, Inc. | Apparatus and method for identifying and modifying computer operating system components |
| US6222529B1 (en) | 1999-05-05 | 2001-04-24 | Shareware, Inc. | Method and apparatus for providing multiple sessions on a single user operating system |
| US6345351B1 (en) * | 1999-11-12 | 2002-02-05 | Telefonaktiebolaget Lm Ericsson(Publ) | Maintenance of speculative state of parallel executed jobs in an information processing system |
| US20020030843A1 (en) * | 2000-02-02 | 2002-03-14 | Tuli Raja Singh | Portable high speed internet access device |
| US7356570B1 (en) | 2000-08-29 | 2008-04-08 | Raja Tuli | Portable high speed communication device |
| US7023572B2 (en) * | 2000-02-02 | 2006-04-04 | Raja Singh Tuli | Portable high speed internet access device |
| US20020115477A1 (en) * | 2001-02-13 | 2002-08-22 | Raja Singh | Portable high speed internet access device with scrolling |
| US7068381B1 (en) | 2000-02-02 | 2006-06-27 | Raja Tuli | Portable high speed internet access device |
| US6633314B1 (en) | 2000-02-02 | 2003-10-14 | Raja Tuli | Portable high speed internet device integrating cellular telephone and palm top computer |
| US7289244B2 (en) | 2000-02-02 | 2007-10-30 | Raja Singh Tuli | Portable high speed internet access device |
| US6941382B1 (en) | 2000-02-07 | 2005-09-06 | Raja Tuli | Portable high speed internet or desktop device |
| US6874009B1 (en) | 2000-02-16 | 2005-03-29 | Raja Tuli | Portable high speed internet device with user fees |
| US20020029285A1 (en) * | 2000-05-26 | 2002-03-07 | Henry Collins | Adapting graphical data, processing activity to changing network conditions |
| US7117239B1 (en) | 2000-07-28 | 2006-10-03 | Axeda Corporation | Reporting the state of an apparatus to a remote computer |
| US7069192B1 (en) * | 2000-08-25 | 2006-06-27 | Hewlett-Packard Company | CAD system |
| US7035912B2 (en) | 2000-08-28 | 2006-04-25 | Abaco.P.R., Inc. | Method and apparatus allowing a limited client device to use the full resources of a networked server |
| US7185014B1 (en) | 2000-09-22 | 2007-02-27 | Axeda Corporation | Retrieving data from a server |
| US8108543B2 (en) | 2000-09-22 | 2012-01-31 | Axeda Corporation | Retrieving data from a server |
| US6842777B1 (en) | 2000-10-03 | 2005-01-11 | Raja Singh Tuli | Methods and apparatuses for simultaneous access by multiple remote devices |
| US7191211B2 (en) * | 2000-10-03 | 2007-03-13 | Raja Tuli | Portable high speed internet access device priority protocol |
| US6915327B1 (en) | 2000-10-30 | 2005-07-05 | Raja Singh Tuli | Portable high speed communication device peripheral connectivity |
| US6928461B2 (en) | 2001-01-24 | 2005-08-09 | Raja Singh Tuli | Portable high speed internet access device with encryption |
| US7254601B2 (en) | 2001-12-20 | 2007-08-07 | Questra Corporation | Method and apparatus for managing intelligent assets in a distributed environment |
| US7178149B2 (en) | 2002-04-17 | 2007-02-13 | Axeda Corporation | XML scripting of soap commands |
| US8176428B2 (en) * | 2002-12-03 | 2012-05-08 | Datawind Net Access Corporation | Portable internet access device back page cache |
| US7966418B2 (en) | 2003-02-21 | 2011-06-21 | Axeda Corporation | Establishing a virtual tunnel between two computer programs |
| US7450128B2 (en) * | 2004-11-15 | 2008-11-11 | Hewlett-Packard Development Company, L.P. | Systems and methods of providing image copy and modify commands to a receiver with an associated display |
| US8171169B2 (en) * | 2005-03-14 | 2012-05-01 | Citrix Systems, Inc. | Method and apparatus for updating a graphical display in a distributed processing environment |
| US8423673B2 (en) * | 2005-03-14 | 2013-04-16 | Citrix Systems, Inc. | Method and apparatus for updating a graphical display in a distributed processing environment using compression |
| US7817849B2 (en) * | 2005-08-18 | 2010-10-19 | Hewlett-Packard Development Company, L.P. | Method and apparatus for graphical data compression |
| US8370479B2 (en) | 2006-10-03 | 2013-02-05 | Axeda Acquisition Corporation | System and method for dynamically grouping devices based on present device conditions |
| US8065397B2 (en) | 2006-12-26 | 2011-11-22 | Axeda Acquisition Corporation | Managing configurations of distributed devices |
| US20080313647A1 (en) * | 2007-06-18 | 2008-12-18 | Microsoft Corporation | Thread virtualization techniques |
| KR100928866B1 (en) * | 2007-09-10 | 2009-11-30 | 주식회사 안철수연구소 | Apparatus and method for running an application in a virtual environment |
| US20090322771A1 (en) * | 2008-03-13 | 2009-12-31 | Eldad Eilam | Method for providing multiple users with private access to a computer |
| KR20110054415A (en) * | 2009-11-17 | 2011-05-25 | 삼성전자주식회사 | Method and apparatus for displaying screen |
| US8634302B2 (en) | 2010-07-30 | 2014-01-21 | Alcatel Lucent | Apparatus for multi-cell support in a network |
| US8504744B2 (en) * | 2010-10-28 | 2013-08-06 | Alcatel Lucent | Lock-less buffer management scheme for telecommunication network applications |
| US8737417B2 (en) | 2010-11-12 | 2014-05-27 | Alcatel Lucent | Lock-less and zero copy messaging scheme for telecommunication network applications |
| US8730790B2 (en) | 2010-11-19 | 2014-05-20 | Alcatel Lucent | Method and system for cell recovery in telecommunication networks |
| US8861434B2 (en) | 2010-11-29 | 2014-10-14 | Alcatel Lucent | Method and system for improved multi-cell support on a single modem board |
| US9357482B2 (en) | 2011-07-13 | 2016-05-31 | Alcatel Lucent | Method and system for dynamic power control for base stations |
| US9864638B2 (en) * | 2012-06-22 | 2018-01-09 | Intel Corporation | Techniques for accessing a graphical processing unit memory by an application |
| CN105681291B (en) * | 2016-01-07 | 2019-07-09 | 锐达互动科技股份有限公司 | A kind of realization multi-client uniform authentication method and system |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5129084A (en) * | 1989-06-29 | 1992-07-07 | Digital Equipment Corporation | Object container transfer system and method in an object based computer operating system |
| US5774642A (en) * | 1990-08-09 | 1998-06-30 | Bull S.A. | Architecture for dynamic service processor exchange providing multitasking environment where multiple processors have access to a system configuration table |
| US5339411A (en) * | 1990-12-21 | 1994-08-16 | Pitney Bowes Inc. | Method for managing allocation of memory space |
| US5504814A (en) * | 1991-07-10 | 1996-04-02 | Hughes Aircraft Company | Efficient security kernel for the 80960 extended architecture |
| US5519867A (en) * | 1993-07-19 | 1996-05-21 | Taligent, Inc. | Object-oriented multitasking system |
| US5729710A (en) * | 1994-06-22 | 1998-03-17 | International Business Machines Corporation | Method and apparatus for management of mapped and unmapped regions of memory in a microkernel data processing system |
| US5771383A (en) * | 1994-12-27 | 1998-06-23 | International Business Machines Corp. | Shared memory support method and apparatus for a microkernel data processing system |
| US5752031A (en) * | 1995-04-24 | 1998-05-12 | Microsoft Corporation | Queue object for controlling concurrency in a computer system |
-
1997
- 1997-03-06 US US08/813,819 patent/US5913230A/en not_active Expired - Lifetime
-
1998
- 1998-01-07 IL IL13073098A patent/IL130730A0/en unknown
- 1998-01-07 JP JP53100898A patent/JP2001508212A/en active Pending
- 1998-01-07 KR KR1019997006084A patent/KR20000069891A/en not_active Application Discontinuation
- 1998-01-07 DE DE69819686T patent/DE69819686T2/en not_active Expired - Lifetime
- 1998-01-07 AU AU59073/98A patent/AU734996B2/en not_active Expired
- 1998-01-07 ES ES98902395T patent/ES2210714T3/en not_active Expired - Lifetime
- 1998-01-07 WO PCT/US1998/000105 patent/WO1998030956A1/en active IP Right Grant
- 1998-01-07 CA CA002275970A patent/CA2275970C/en not_active Expired - Lifetime
- 1998-01-07 EP EP98902395A patent/EP0951675B1/en not_active Expired - Lifetime
-
1999
- 1999-01-25 US US09/236,928 patent/US6023749A/en not_active Expired - Lifetime
Also Published As
| Publication number | Publication date |
|---|---|
| JP2001508212A (en) | 2001-06-19 |
| EP0951675B1 (en) | 2003-11-12 |
| AU5907398A (en) | 1998-08-03 |
| CA2275970A1 (en) | 1998-07-16 |
| AU734996B2 (en) | 2001-06-28 |
| US5913230A (en) | 1999-06-15 |
| WO1998030956A1 (en) | 1998-07-16 |
| ES2210714T3 (en) | 2004-07-01 |
| IL130730A0 (en) | 2000-06-01 |
| DE69819686T2 (en) | 2004-09-16 |
| US6023749A (en) | 2000-02-08 |
| EP0951675A1 (en) | 1999-10-27 |
| DE69819686D1 (en) | 2003-12-18 |
| KR20000069891A (en) | 2000-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2275970C (en) | Object and method for providing efficient multi-user access to shared operating system kernal code using instancing | |
| JP4160255B2 (en) | Application programming interface that controls the allocation of physical memory in a virtual storage system by an application program | |
| US5539899A (en) | System and method for handling a segmented program in a memory for a multitasking data processing system utilizing paged virtual storage | |
| US4742450A (en) | Method to share copy on write segment for mapped files | |
| US5581765A (en) | System for combining a global object identifier with a local object address in a single object pointer | |
| US5893166A (en) | Addressing method and system for sharing a large memory address space using a system space global memory section | |
| US4761737A (en) | Method to automatically increase the segment size of unix files in a page segmented virtual memory data processing system | |
| US5835775A (en) | Method and apparatus for executing a family generic processor specific application | |
| US6134602A (en) | Application programming interface enabling application programs to group code and data to control allocation of physical memory in a virtual memory system | |
| JP4718683B2 (en) | Method and system for restoring physical memory state when focus is changed between application programs in a computer | |
| US5873127A (en) | Universal PTE backlinks for page table accesses | |
| US5835961A (en) | System for non-current page table structure access | |
| US5897660A (en) | Method for managing free physical pages that reduces trashing to improve system performance | |
| US5860144A (en) | Addressing method and system for providing access of a very large size physical memory buffer to a number of processes | |
| JPS623359A (en) | Virtual storage system | |
| GB2322209A (en) | Multi-tasking computer system has shared address space among multiple virtual address spaces | |
| WO1998014878A1 (en) | A method of obtaining a buffer of contiguous memory and building a page table | |
| EP0892351A1 (en) | Large memory allocation method and apparatus | |
| US5873120A (en) | Variable split virtual address space allocation with multi-system compatibility | |
| US5875487A (en) | System and method for providing efficient shared memory in a virtual memory system | |
| EP0403124A2 (en) | Overlay swapping | |
| US5940869A (en) | System and method for providing shared memory using shared virtual segment identification in a computer system | |
| US6600493B1 (en) | Allocating memory based on memory device organization | |
| WO1990012363A1 (en) | Addressing technique for transparently extending data processing system address space | |
| CN109783145B (en) | Method for creating multi-image-based multifunctional embedded system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| MKEX | Expiry |
Effective date: 20180108 |