Abstract:Software Pipelining is a classic and important loop-optimization for VLIW processors. It improves instruction-level parallelism by overlapping multiple iterations of a loop and executing them in parallel. Typically, it is implemented using heuristics. In this paper, we present an optimal software pipeliner based on a Satisfiability Modulo Theories (SMT) Solver. We show that our approach significantly outperforms heuristic algorithms and hand-optimization. Furthermore, we show how the solver can be used to give feedback to programmers and processor designers on why a software pipelined schedule of a certain initiation interval is not feasible.
Submission history
From: Jan-Willem Roorda [view email]
[v1]
Thu, 29 Jan 2026 15:19:12 UTC (220 KB)
[v2]
Fri, 30 Jan 2026 14:08:20 UTC (220 KB)