Electrical computers and digital processing systems: processing – Processing control – Context preserving (e.g. – context swapping – checkpointing,...
Reexamination Certificate
1997-05-06
2001-01-16
Kim, Kenneth S. (Department: 2783)
Electrical computers and digital processing systems: processing
Processing control
Context preserving (e.g., context swapping, checkpointing,...
C709S241000, C711S203000
Reexamination Certificate
active
06175916
ABSTRACT:
TECHNICAL FIELD
This invention relates to operating systems and to methods of calling and executing system functions.
BACKGROUND OF THE INVENTION
Modern operating systems are capable of concurrently executing a plurality of different processes. Generally, an individual process is an executable program, defined by code and data. It has a private virtual address space, which is a set of virtual memory addresses that the process can use. In addition, a process has one or more execution threads.
Operating systems use virtual memory to isolate different processes from each other to a certain degree, and to thereby prevent one process from interfering with another. With virtual memory, a process is assigned its own, private virtual address space, which is generally not available to other processes. Through its virtual memory, a process has a logical view of memory that does not correspond to the actual layout of physical memory. Each time a process uses a virtual memory address, a virtual memory system translates it into a physical address using a virtual-to-physical address mapping contained in some type of look-up structure and address mapping database. The virtual-to-physical address mappings are unique for each address space. The virtual memory system prevents processes from directly accessing the virtual memory used by other processes or by the operating system.
As mentioned above, a process has one or more execution threads. Generally, an execution thread comprises a sequence of processor instructions that execute in a single processor context. The particular elements of a thread's context vary depending on the microprocessor being used. For purposes of the discussion herein, however, a thread's context always includes its private memory stack or stacks. Therefore, by definition, a single thread always uses the same private memory stack. Any time the processor context changes to a different memory stack, the processor is said to be executing a different thread. In many cases, a thread's context also includes the volatile contents of a set of processor registers.
A processor is capable of executing only one thread at a time. However, a multitasking operating system allows users to run multiple programs and processors, while appearing to execute all of them at the same time. It achieves this in the following way:
1. It runs a thread until the thread's execution is interrupted or until the thread must wait for a resource to become available.
2. It saves the thread's context.
3. It loads another thread's context.
4. It repeats this sequence as long as there are threads waiting to execute.
It is common for a computer processor and associated operating system to allow two different levels of resource availability. One level, referred to as a non-privileged mode or user mode, is used by application programs and other so-called “user” processes or programs. At this level, an execution thread is prevented by the operating system and by the computer processor from performing certain security-critical operations. The thread is also prevented from directly accessing many system resources. The purpose of the non-privileged execution mode is to isolate a user process as much as possible so that it cannot interfere with other user processes or with operating system functions. While a user process may itself crash, it should not be able to crash other programs or the operating system.
The other level of execution is referred to as privileged mode, system mode, or kernel mode. Critical operating system components are implemented in kernel mode—kernel-mode components are responsible for things like virtual memory management, responding to interrupts and exceptions, scheduling execution threads, synchronizing the activities of multiple processors, and other critical or sensitive functions. Such components, which execute from system mode, are generally referred to collectively as “the kernel.” The kernel is responsible for supervising the virtual memory system in most computer systems.
A user-mode execution thread gains access to the operating system by calling system services or functions. However, a conventional program call or jump instruction to system code is not allowed because the operating system is located in a protected virtual address space that cannot be directly accessed by user-mode threads. Rather, the calling user-mode thread utilizes a special processor instruction, typically called a trap instruction. A trap instruction causes an automatic jump to a privileged-mode trap handler. One or more arguments are usually provided by the calling process, indicating which system function is desired. These arguments are passed in processor registers or on a memory stack. The trap handler suspends the calling thread and schedules a new thread for execution of the system function. When the system function has completed, the operating system reschedules the calling thread.
The Windows® NT operating system is an example of an operating system that implements multiple threads. The Windows® NT operating system is described in
Inside Windows NT
, by Helen Custer, published by Microsoft Press, 1993, which is hereby incorporated by reference. As described in this book at page 230, a trap instruction results in a switch to privileged-mode execution and a jump to the kernel. The kernel copies the caller's arguments from the thread's user-mode stack to another stack that is referred to as the thread's kernel-mode stack. The kernel then executes the system service. Thus, a context change is made (changing stacks) in response to a trap instruction; in accordance with the terminology adopted herein, a different thread (using a different stack) is initiated within the kernel and is used for executing the kernel itself and the called system function.
As noted above, arguments can be passed to a system function on the memory stack of the calling thread. In addition, it is desirable to be able to pass pointer arguments to a system function so that data can be specified by reference. Accordingly, the kernel and system functions must be able to access the virtual address spaces of calling processes.
In the Windows® NT operating system, this is accomplished as illustrated in FIG.
1
. The operating system utilizes a large virtual address space
20
of 4 GB. Each application has its own user address space
22
, which is limited to the lower 2 GB of the overall 4GB space. This lower 2 GB is accessible to both user-mode and kernel-mode threads. The upper 2 GB of the overall space forms a system address space
24
. It accessible only to kernel-mode threads. Components of the operating system are located in system address space
24
. Although the mapping of user address space
22
changes according to which user-mode process is currently executing, the mapping of system address space
24
remains constant.
The Windows 95® operating system is described in
Inside Window
95, by Adrian King, published by Microsoft Press, 1994. This book is also incorporated by reference. The Windows 95® operating system utilizes a slightly different memory arrangement, as shown in FIG.
2
. The lowest 1 MB of the virtual address space, referenced by numeral
26
, is used for the currently executing MS-DOS process. Each such process also has a valid memory map within the 2 GB to 3 GB region, referenced by numeral
28
. When a particular process is active, its memory map in region
28
is identical to the memory map of region
26
. This mapping allows the operating system to address the memory of any MS-DOS process (by accessing an address in the 2 GB to 3 GB virtual memory region), regardless of whether the MS-DOS process is active. Pointers supplied to the operating system by the MS-DOS process, which reference region
26
, are translated by the operating system so that they refer to the corresponding location in the 2 GB to 3 GB range
28
.
The upper GB
30
of virtual memory is reserved for privileged-mode system components. The 4 MB to 2 GB region
32
is used as private or user virtual
Fenwick Thomas
Ginsberg Michael
Kim Kenneth S.
Lee & Hayes PLLC
Microsoft Corporation
LandOfFree
Common-thread inter-process function calls invoked by jumps... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Common-thread inter-process function calls invoked by jumps..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Common-thread inter-process function calls invoked by jumps... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2518977