Accelerating Control Flow on CGRAs via Speculative Iteration Execution

Coarse-Grained Reconfigurable Arrays (CGRAs) offer a promising architecture for accelerating general-purpose, compute-intensive tasks. However, handling control flow within these tasks remains a challenge for CGRAs. Current methods for handling control flow in CGRAs execute condition operations before selecting branch paths, which adds extra execution time. This article proposes a CGRA architecture that decouples the control flow condition and path selection within an iteration through speculative iteration execution (SIE), where the condition is predicted before the start of the current iteration. Compared to existing methods, the SIE CGRA achieves a geometric mean speedup of $1.31\times$ over Partial Predication, $1.17 \times$ over Dynamic-II Pipeline and $1.12\times$ over Dual-Issue Single-Execution.