Showing papers by "Claude Castelluccia published in 1995"
23 Aug 1995
TL;DR: This paper describes the design for an optimizer that is incorporated into the Esterel compiler that will improve the performance of application specific protocol implementation when compared to the use of a standard protocol.
Abstract: The Hipparcli project [CasSlb] uses the Esterel language [Dio95] for describing protocol building blocks. These blocks cui be configured to form an application specific protocol implementation. Using application specific knowledge should improve the performance of such ai iniplciiiciitation when compared to the use of a standard protocol. However, the C code generated by tlie currently existing Esterel compiler is not very fast, and moreover has a largc code size lCas94aI. This is a problem the compiler shares with many similar tools that gencratc executable code starting from a protocol defined in a formal description technique (FDT). This paper describes our design for an optimizer that we incorporated into the Esterel compiler. After analyzing tlic code generated by the Esterel compiler, we decided that two optiiiiizatioii techniques nierit further investigation:
8 citations
01 Dec 1995
TL;DR: HIPPCO automatically generated implementations of TCP are compared with the BSD implementation and it is shown that the HIPPCO generated codes requires up to 70% less instructions than its BSD counterpart.
Abstract: Optimisation de code, G=C8n=C8ration automatique de protocoles, ALF} % mots cles en francais \RRabstract{This report presents HIPPCO, an High Performance Protocol Code Optimizer. HIPPCO belongs to the HIPPARCH compiler. HIPPARCH is a tool which proposes to generate automatically from the application communication requirements and the network characteristics an efficient implementation of a customized protocol. HIPPCO is the last stage of this protocol compiler. It takes as input a description of the protocol automaton, optimizes it and generates an implementation in {\it C}. HIPPCO decomposes the protocol automaton in two parts: the common and uncommon path. It then uses this decomposition to apply a set of optimizations toward a good code speed/code size tradeoff. In the first part of this report, the code speed optimizations are described. Those optimizations reduces the number of executed instructions and improves the instruction cache and pipeline behaviors. In the second part, a comparaison of HIPPCO automatically generated implementations of TCP are compared with the BSD implementation. We show that the HIPPCO generated codes requires up to 70% less instructions than its BSD counterpart.
5 citations