Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper presents a graph-based methodology for Engineering Change Orders (ECOs) in Google's High-Level Synthesis (HLS) tool, XLS. They use Graph Edit Distance (GED) to identify structural differences between original and revised intermediate representations (IRs), converting these differences into patch operations. The method enforces XLS IR constraints and preserves semantic correctness while maintaining the original pipeline registers, demonstrating high structural reuse and schedule preservation.
Late-stage design changes in HLS don't have to be a nightmare: this graph-based ECO approach achieves high structural reuse and schedule preservation, promising faster iteration cycles.
High-level synthesis (HLS) tools offer limited support for Engineering Change Orders (ECOs), making late-stage design modifications challenging and costly. This paper introduces a graph-based ECO methodology tailored for Google XLS. A Graph Edit Distance (GED) algorithm is used to detect structural differences between original and revised intermediate representations (IRs), which are then transformed into patch operations. A patch application mechanism is developed to enforce XLS IR constraints while preserving semantic correctness, together with a schedule constraining scheme that maintains the original pipeline registers. Experiments across several XLS designs demonstrate high structural reuse ratios, effective schedule preservation, and full functional correctness, highlighting the practicality of the approach for production HLS flows.