GCDS: A Compiler Strategy for Trading Code Size Against Performance in Embedded Applications
Abstract
In this paper we present GDS, a new approach for code optimization within the context of embedded systems. GDS applies several transformation sequences on each piece of code and chooses a posteriori the best trade-off between code size and performance, given the particular constraints. The proposed implementation relies on a modular framework that makes easy to handle differentcompilation strategies on the same piece of code.