H. Wenske. Design and implementation of fast local ipc for the L4 microkernel. Studienarbeit, System Architecture Group, University Karlsruhe, Germany, 2002. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Identifying and Modeling Components in the SawMill Operating System - Wagner (Correct)
....may help in evaluating the system. These tools provide more functionality and prevent inconsistencies. Furthermore, information about the dynamic behavior of generated operating systems would be interesting. This information may be used to implement kernel speci c optimizations like LIPC [LW01, Wen02] It may also be used during runtime with runtime aspect weaving [BH01] for optimizations, or to add garbage collection servers to the available set of SawMill s system servers. Appendix A Implementation Details A.1 Code Elements and Corresponding Node and Arc Types This Section details, ....
H. Wenske. Design and implementation of fast local ipc for the L4 microkernel. Studienarbeit, System Architecture Group, University Karlsruhe, Germany, 2002.
Using Platform-Specific Optimizations in Stub-Code Generation - Haeberlen (2002) (Correct)
....temporary mapping. 9 3 TARGET PLATFORM 3.2 The Intel Pentium II Processor Recent work [28] has introduced an optimization for intra domain IPC which does not require a switch to kernel mode in certain cases. With this optimization, the bestcase overhead can be reduced to approximately 23 cycles [35]. The Pistachio kernel The Pistachio kernel implements the V4 kernel API [33] The most important difference to the X0 API is that this version does not support memory messages; instead, every thread has 64 message registers (MRs) which can be transferred to the receiver s MRs partly or entirely ....
....an optimization for frequent intra domain communication; it allows the callee to run in the context of the caller for some time, thereby saving the cost for a context switch. The principle is part of the V4 interface and has also been implemented in an experimental variant of the Hazelnut kernel [35]. Special system calls (local IPC, lipc) are available in Hazelnut and Pistachio for switching lazily from the sender to the receiver. My experiments show that using this primitive can boost RPC performance by up to 900 ; however, it can also add overhead when used improperly, e.g. for nonlocal ....
[Article contains additional citation context not shown here]
Horst Wenske. Design and implementation of fast local IPC for the L4 microkernel. Study thesis, University of Karlsruhe, Jul 2002. 38
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