array constraints
ConceptArray constraints are a way to model reads and writes on arrays in constraint solving without expanding each update into element-wise constraints over the entire array. In the cited STCS work, classical SSA-style encodings are described as impractical for large memories, so the authors introduce specialized array read/write constraints and an alternative two-step strategy based on choosing equality relationships among access indices. The number of possible aliasing relationships among n indices is characterized by the nth Bell number.
WIKI
Array constraints model how array accesses interact inside a constraint system, especially when the array is large enough that element-wise expansion is impractical.
Motivation
Prior approaches cited in the source handle array manipulations with constructions equivalent to SSA form. In that style, each assignment to an array element is translated into element constraints of the underlying constraint logic programming language. For an array of size n, a single assignment introduces 2n - 1 element constraints. The source notes that this becomes infeasible for memory-like arrays addressed by wide index registers, such as the 24-bit addressed memory used in the ST processor case study. [C1]
NEIGHBORHOOD
No graph connections found for this entity yet. It may appear in future ingestion runs.
explore full graph →