Electrical computers and digital data processing systems: input/ – Input/output data processing – Peripheral adapting
Reexamination Certificate
1998-05-29
2001-12-18
Kim, Matthew (Department: 2186)
Electrical computers and digital data processing systems: input/
Input/output data processing
Peripheral adapting
C710S065000, C711S165000, C345S205000
Reexamination Certificate
active
06332172
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to embedded systems. In particular, the present invention relates to a system and method for utilizing virtual memory in an embedded computer system.
BACKGROUND OF THE INVENTION
An embedded computer system typically refers to a computer which is physically embedded within a larger system and whose primary purpose is to maintain some property or relationship between the other components of the system in order to achieve the overall system objective. Embedded computers are now used in a wide variety of systems, such as aircraft, automobiles, appliances, weapons, medical devices, and computer peripheral devices.
Embedded software presents a unique problem and requires a different type of development strategy than other types of software, such as data processing or transaction systems where the computer is at the center of the application. In the computer-centralized system, peripheral equipment with which the computer interacts, such as input, storage, and output devices, is there to serve the needs of the computer and not vice versa. In this type of system, the behavior of the other components of the system are usually known and often designed or chosen with the needs of the computer as the guiding feature. In the embedded system, the computer is typically used to service the needs of the other components; thus, its behavior and design is usually severely constrained by the external process being controlled. Furthermore, the knowledge about the behavior of the physical processes may only be partially known and is often continuous and stochastic and therefore difficult to incorporate into the usually discrete and deterministic computer software model. Instead of having the freedom to select external devices that satisfy the requirements of the computer, the other system components usually dictate the requirements for the embedded computer. Furthermore, the order, timing, and required handling of input events by the computer is typically completely controlled by the other system components, rather than by the software designer. Events that occur in large numbers in a short time or simultaneously must be handled by the computer software in ways that will satisfy the needs and requirements of the larger system. Software requirements for embedded systems are typically allocated during the system engineering process.
Errors must also typically be handled differently in embedded systems than in ordinary computer systems. In most other computer systems, providing information that an error has occurred and discontinuing the processing of the erroneous transaction is satisfactory and perhaps even desirable. A human can then intervene to analyze the error and determine the appropriate recovery procedure. Although the computer system needs to provide corruption procedures (e.g. for erroneous entries in an electronic data base), the decision to make the correction can be handled externally and often off-line. In embedded systems, however, errors and failures must be dealt with immediately, and often the detection and recovery from errors must be automated. The computer must be robust (must continue to operate in a specified manner), even though other components of the system may fail. Also, the other components must typically be made robust in the face of computer errors and failures. Finally, embedded computer software must typically provide facilities to detect and recover from its own errors or, at the very least, to fail gracefully in a way that minimizes damage to the overall system.
FIG. 1
is a block diagram of a conventional embedded system
100
. The embedded system
100
is shown to include a central processing unit (CPU)
102
coupled to a random access memory (RAM)
104
, a FLASH memory
106
, and a read-only only memory (ROM)
108
. CPU
102
can read and write from RAM
104
and FLASH Memory
106
, in addition to reading data from ROM
108
.
FIG. 2
is an illustration of how a conventional “run-from-RAM” embedded application is built. An image
200
of RAM
104
(of
FIG. 1
) is built along with an application. Image
200
is typically a picture of what RAM
104
will include. Image
200
typically includes code (text) and data. Image
200
is then compressed. Compressed RAM Image
202
is then stored in FLASH memory
106
(of FIG.
1
). FLASH memory
106
typically includes an extractor which is utilized in conjunction with the compressed image. An extractor itself is a small executable image which can read compressed RAM image and write uncompressed RAM image to the RAM. During run-time, the compressed RAM Image
202
is then decompressed into RAM
It is often highly desirable to restrict RAM, ROM, and FLASH memory as much as possible in an embedded computer system to reduce cost. The amount of code and data is limited by the size of the RAM in the embedded system.
It would be desirable to optimize the available RAM such that more computer code and data can be utilized with existing amount of RAM. Alternatively, it would be desirable to be able to utilize the same amount of computer code and data with less RAM. The present invention addresses such a need.
SUMMARY OF THE INVENTION
The present invention provides a system and method for reducing memory requirements in a low resource computer system. Although examples herein are described in terms of embedded systems, the present invention is equally applicable to other low resource computer systems, such as palmtops, and laptops. Accordingly, the term embedded system is herein meant to include palmtops and laptops, as well as traditional embedded systems. Memory requirements, such as RAM requirements, can be reduced by combining virtual memory with a secondary memory with statically compressed contents. According to embodiments of the present invention, executable image of memory, such as the image of RAM, is compressed at image production time. The compressed image is then stored in a non-volatile memory, such as FLASH memory. At run-time, when a request identifying a virtual address is received, it is determined whether a physical address in the physical memory, such as RAM, is associated with that virtual address. If there is no physical memory associated with that virtual address, then a physical address is obtained to be associated with that virtual address. The obtained physical address can be either an available physical address (having no current association with a virtual address), or a physical address which is made available (by severing a current association with a virtual address). Data related to the requested virtual address may be extracted from the compressed image stored in the non-volatile memory, decompressed, and stored in RAM at the location of the obtained physical address. The physical address may then be mapped in order to be associated with the requested virtual address.
A method according to an embodiment of the present invention for compressing data for use in an embedded computer system is presented. The method comprising the steps of providing a virtual region of an image of memory, the virtual region including at least one pageable page; providing a physical region of the image of memory, the physical region including at least one page; and compressing a pageable page by referring to the at least one non-pageable page.
In another aspect of the present invention, a system according to an embodiment of the present invention for compressing data for use in an embedded computer system is presented. The system comprises a means for providing a pageable region of an image of memory, the pageable region including at least one pageable page; a means for providing a non-pageable region of the image of memory, the non-pageable region including at least one non-pageable page; and a means for compressing a pageable page by referring to the at least one non-pageable page.
A system according to an embodiment for compressing data for use in an embedded computer system is also presented. The system comprising a non-volatile memory which provides a page
Cisco Technology Inc.
Kim Matthew
Vanpelt & Yi LLP
Vital Pierre M.
LandOfFree
Method and system for virtual memory compression in an... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for virtual memory compression in an..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for virtual memory compression in an... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2600246