From Tool Invocation to Source-Mechanism Exploration: Protected White-Box DSE for Open-Source EDA

2026-07-13

AI Summary

This work addresses the limitations of traditional black-box design space exploration (DSE), which is constrained to tuning only the publicly exposed parameters of EDA tools without insight into their internal optimization mechanisms. The paper proposes ReviewDSE, a novel framework that enables protected white-box DSE at the source-code level of open-source EDA tools such as OpenROAD. By constructing mechanism-level evidence, generating origin-aware branches, guiding search through teacher-review feedback, and validating the entire workflow, ReviewDSE advances DSE from mere tool invocation to deep mechanistic exploration and supports reusable warm-start strategies. Evaluated on nine target designs, ReviewDSE reduces post-placement half-perimeter wirelength (HPWL) by 1.78% on average under a 2× runtime budget—substantially outperforming black-box DSE (0.38%). Runtime-aware selection maintains a 1.68% improvement with only 1.11× overhead and effectively resolves hard routing legality violations.

Abstract

Open-source EDA tools allow design-space exploration (DSE) to move beyond public knobs and into bounded source-level mechanisms inside staged optimizers. We present ReviewDSE, a protected white-box DSE framework that explores such mechanisms for a target design. ReviewDSE evaluates complete source candidates under a protected evaluator and records reusable search knowledge as reviewed mechanism-level evidence. It first constructs method evidence and source-start branches from calibration designs, then uses these fixed warm-start products to initialize target-case exploration under Teacher review and full-flow validation. We instantiate ReviewDSE on OpenROAD detailed placement as a representative staged open-source EDA optimizer. Across nine target tasks, ReviewDSE reduces final post-DPL half-perimeter wirelength (HPWL) by 1.78\% on average under a 2$\times$ runtime gate, compared with 0.38\% for public-knob black-box DSE. A runtime-aware ReviewDSE selection retains a 1.68\% reduction at 1.11$\times$ runtime, and full-flow review exposes stage-composability failures while source-mechanism exploration repairs hard cut-row legality failures.

Problem

• design-space exploration
• open-source EDA
• source-level mechanisms
• white-box DSE
• staged optimizers

Innovation

• white-box DSE
• source-mechanism exploration
• open-source EDA
• design-space exploration
• ReviewDSE