Cache persistence analysis

Abstract

To compute a worst-case execution time (WCET) estimate for a program, the architectural effects of the underlying hardware must be modeled. For modern processors this results in the need for a cache and pipeline analysis. The timing-relevant result of the cache analysis is the categorization of the accesses to cached memory. Categorizations that are obtainable by the well-known must and may cache analysis are always-hit, always-miss and not-classified. The cache persistence analysis tries to provide additional information for the not-classified case to limit the number of misses. There exists a cache persistence analysis by Ferdinand and Wilhelm based on abstract interpretation computing these classifications. In this paper, we present a correctness issue with this analysis and a novel analysis that fixes it. For fully timing compositional architectures the persistence information is straightforward to use. We will focus on the application of the persistence analysis for state-of-the-art architectures that show timing anomalies. Such architectures do not allow to quantify the costs of a single cache hit or miss in isolation. To make the usage of the persistence information feasible, we integrate the novel persistence analysis together with a novel path analysis approach into the industrially used WCET analyzer aiT.