E. C. Cooper, and R. P. Draves, "C threads", Technical report, Computer Science Department, Carnegie Mellon University, CMU-CS-88-154, March, 1987.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....However, in the Mach kernel, it is often difficult to analyze the runtime behaviour of the time critical activities due to lack of real time scheduling and synchronization facilities. For instance, Mach provides synchronization facility among threads using a conventional monitor based mechanism[5]. Threads can enter a critical section in FIFO order, however, this may cause a a priority inversion problem among threads. If many lower priority threads are already waiting on a mutex variable, a higher priority thread which may have a tighter deadline must wait for their completion. Thus, the ....

....Like Real Time Mach, Amoeba can support a set of single board computers without having local disks, however, it does not provide any safe mechanisms for creating a periodic thread and avoiding priority inversion problems. The POSIX Thread proposal[7] is very similar to Mach s C Thread package[5]. However, it also does not distinguish between real time threads and non real time threads. Like RealTime Mach, it can dynamically select the thread scheduling policy. A POSIX thread also contains the thread attributes such as inherit priority, scheduling priority, scheduling policy, and minimum ....

E. C. Cooper, and R. P. Draves, "C threads", Technical report, Computer Science Department, Carnegie Mellon University, CMU-CS-88-154, March, 1987.

....built on top of SML NJ s continuations should be efficient relative to the rest of the ML computation, thread applications may not perform as well overall under SML NJ as under stack based systems. The heap based approach has another significant advantage over systems, such as C threads [12], that require a stack to be allocated explicitly for each thread. Since SML NJ threads do not use a stack, no stack space needs to be reserved for them. Consequently, we support the use of 1 Many useful scheduling policies would require minor changes to the signature; for example, priority ....

E. C. Cooper and R. P. Draves. C Threads. Technical Report CMU-CS-88-154, Computer Science Department, Carnegie Mellon University, June 1988.

....operations built on top of SML NJ s continuations should be ecient relative to the rest of the ML computation, thread applications may not perform as well overall under SML NJ as under stack based systems. The heap based approach has another signi cant advantage over systems, such as C threads [12], that require a stack to be allocated explicitly for each thread. Since SML NJ threads do not use a stack, no stack space needs to be reserved for them. Consequently, we support the use of 1 Many useful scheduling policies would require minor changes to the signature; for example, priority ....

E. C. Cooper and R. P. Draves. C Threads. Technical Report CMU-CS-88-154, Computer Science Department, Carnegie Mellon University, June 1988.

....: 4 4. 2 Synchronization Cost Analysis : 5 5 Conclusion and Future Work 8 ii 1 Introduction Mach provides synchronization facility among threads using a conventional monitor based mechanism[4]. However, in real time applications, it is often encountered a a priority inversion problem among synchronizing activities. For instance, if there are many lower priority threads are already waiting on a mutex variable, then a higher priority thread which may have a tighter deadline cannot skip ....

E. C. Cooper, and R. P. Draves, "C threads", Technical report, Computer Science Department, Carnegie Mellon University, CMU-CS-88-154, March, 1987.

....are exported separately, allowing both synchronous and asynchronous styles of I O. The external memory object protocol allows a user to map the frame buffer for a graphics device directly into its address space. 3.1.6. User Multiprocessing A user level multithreading package, the C Thread library [3], eases the use of multiple threads within an address space. It exports mutual exclusion mutex locks and condition variables for synchronization via condition wait and condition signal operations. 3.2. Unix Server The bulk of Unix services are provided by the Unix Server. It is implemented as a ....

Cooper, E. and Draves, R. C Threads. Technical Report CMU-CS-88-154, Computer Science Department, Carnegie Mellon University, June, 1988.

....thread abstractions. Mach tasks are passive entities consisting of an address space and communication facilities. Mach threads are kernel level and are the units of scheduling. A C threads library that implements light weight user level thread abstractions on top of Mach threads is also provided [33]. C threads are multiplexed onto Mach threads, and user mode context switching is implemented. The C threads library also provides mutual exclusion locks and condition variables for synchronisation. A number of scheduling classes can be defined for threads, and scheduling policies can be ....

E. Cooper and R. Draves. C Threads. Technical Report CMU-CS-88-154, Computer Science Department, Carnegie Mellon University, 1988.

....article refers to these entities as multiroutines; they may be thought of as multiprocessor generalizations of coroutines. Special cases of multiroutines include coroutines (only 1 VP) and the common programming notion of multiple threads (1 VP per multiroutine) as found the Mach Cthreads library[7]. Multiroutines and Virtual Processors can be identified in almost any parallel programming language implementation or environment. One example is that an Ada 10 runtime on Unix would use Unix processes as its virtual processors and Ada tasks as its multiroutines. Another example is that the ....

E. C. Cooper and R. P. Draves, "C Threads," Tech. Rep. CMU-CS-88-154, Computer Science Department, Carnegie Mellon University, June 1988.

....have demonstrated that multiple threads are useful, but multiple address spaces are unnecessary [6, 14, 16] As a result, we designed the CAB to provide a single physical address space with a runtime system that supports multiple threads. The threads package is based on Mach C threads [9]: it provides mutex locks to enforce critical regions, and condition variables for synchronization. Preemption and priorities were added so that system threads (e.g. protocol threads) can be scheduled quickly, even in the presence of long running application threads [8] The SPARC CPU has seven ....

Eric C. Cooper and Richard P. Draves. C Threads. Tech. Rept. CMU-CS-88-154, Computer Science Department, Carnegie Mellon University, June, 1988.

....model ffl distinguishes between real time and non real time threads, ffl assumes explicit timing constraints for each real time thread, and ffl provides a priority inheritance protocol to avoid unbounded priority inversion. The POSIX Thread proposal[11] is very similar to Mach s C Thread package[5] and it also does not distinguish between real time threads and non real time threads. This poses a problem of identifying the type of threads that can pinned down its memory objects. However, it can dynamically select the thread scheduling policy and a thread also contains the thread attributes ....

E. C. Cooper, and R. P. Draves, "C threads ", Technical report, Computer Science Department, Carnegie Mellon University, CMU-CS-88-154, March, 1987.

....of DTS, a system which allows a user to distribute a program over a network of loosely coupled workstations. As described above, this can lead to significant improvements in program execution time. DTS uses Parallel Virtual Machine (PVM) 8] as the underlying message passing system and C Threads [3] for parallel execution on a single node. It basically offers the usual thread functions fork and join, here extended to distributed computing in a SPMD (single program over multiple data streams) 4] programming environment. In contrast to the client server model, there is only one executable, ....

E.C. Cooper and R.P. Draves, C Threads, Technical Report, Computer Science Department, Carnegie Mellon University, Pittsburgh, PA 15213, July 1987.

....2) we provide explicit timing constraints for each real time thread, 3) a scheduling policy for a real time application can be selected, and 4) various locking protocols are provided to avoid unbounded priority inversion. The POSIX Thread proposal[16] is very similar to Mach s C Thread package[7] and it also does not distinguish between real time threads and non real time threads. This poses a problem of identifying the type of threads that has a hard deadline or can pinned down its memory objects. However, it can dynamically select the thread scheduling policy and a thread also contains ....

E. C. Cooper, and R. P. Draves, "C threads", Technical report, Computer Science Department, Carnegie Mellon University, CMU-CS-88-154, March, 1987.

....messages with the CAB on behalf of the application. The CAB runtime system manages hardware devices such as timers and DMA controllers, supports multiprogramming (the threads package) and manages data buffers (the mailbox module) The threads package, derived from the Mach C Threads package [9], supports lightweight threads in a single address space. Threads provide a low cost, flexible method of sharing the CABCPU between concurrent activities, which is important for communication protocol implementation. Mailboxes provide flexible and efficient management of buffer space in the CAB ....

Eric C. Cooper and Richard P. Draves. C threads. Technical Report CMU-CS-88-154, Carnegie-Mellon University, Computer Science Department, June 1988. 10 of 10

....for programming applications. 6.1 The CAB Kernel Candidate run time systems for the CAB range from a minimal single task system to a complete UNIX implementation. To provide the required efficiency and flexibility, we built the CAB kernel around lightweight processes similar to Mach threads [8]. Threads support multitasking so the CAB can execute multiple activities concurrently in a time shared fashion, but, since threads have little state associated with them, the cost of context switching is low. Thread switching takes between 10 and 15 microseconds; almost all of this time is spent ....

Eric C. Cooper and Richard P. Draves. C Threads. Technical Report CMU--CS--88--154, Computer Science Department, Carnegie Mellon University, June 1988.

....collection) and a powerful module facility. The Standard ML of New Jersey (SML NJ) implementation [3, 7] supports type safe, first class 1 UNIX is a trademark of AT T Bell Laboratories. Hoare s monitors [22] Mesa processes [25] Modula 2 threads [9, 10, 33] CMA [17] C threads [15] Phi Phi Phi Wand s coroutines [38] POSIX threads [24] Ramsey s and Reppy s coroutines [28, 30] H H H Phi Phi Phi SML threads Figure 1: Genealogy of thread interfaces continuations with callcc and throw, and provides asynchronous exception handling facilities in the form of signal ....

....waiting, and go back to sleep if necessary. At DEC SRC, the researchers who implemented processes and monitors in Mesa went on to add threads to Modula 2 for the Firefly multiprocessor workstation. The DEC SRC thread design [9, 10, 33] is the ancestor of various thread interfaces for UNIX and C [15, 17, 24], as well as the current work. This similarity in ancestry has an important practical implication: there is a large, common subset of functionality in all of these packages. For example, Birrell s excellent tutorial on programming with threads [10] is directly applicable to the SML threads ....

[Article contains additional citation context not shown here]

Eric C. Cooper and Richard P. Draves. C Threads. Technical Report CMU-CS-88-154, Computer Science Department, Carnegie Mellon University, June 1988.

....the past few years, and, second, to describe the 3DFBT as an example implementation of a vision system following this approach. Because of space limitations, many important technical details must be omitted. 2 Real Time Task Directed Vision Our approach, real time task directed robot vision [6, 9, 5], is based on the principle that a robot must interact with its environment in performing its assigned tasks. This approach has similarities to the active vision paradigm (see, for example, 1, 2, 3, 11] but is wider in its scope. By real time , we mean that the outputs of visual processing ....

James Cooper and Leslie Kitchen. A proposal for real-time, task-directed robot vision. Technical Report 89/10, Computer Science Department, The University of Western Australia, Nedlands, W.A. 6009, Australia, June 1989.

No context found.

COOPER, E. AND DRAVES , R. C Threads. Technical report CMU-CS-88-154, Carnegie Mellon University. Computer Science Department, Pittsburgh, PA (June 1988).

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC