arXiv: 2604.22234 · PDF

Authors: Taizun Jafri, Vidya A. Chhabria

Primary category: cs.AR · all: cs.AR

Matched keywords: large language model, llm, agent, agentic, rag

TL;DR

GR-Evolve uses an agentic LLM to iteratively evolve global router source code, specializing EDA algorithms per-design via QoR feedback within OpenROAD, achieving up to 8.72% post-detailed-routing wirelength reduction over baselines.

Key Ideas

  • Introduces “design-adaptive EDA tooling”: algorithms themselves adapt to each design, not just hyperparameters.
  • Uses LLM-driven code evolution on global router source code.
  • Closes the loop with QoR-driven feedback from OpenROAD toolchain.
  • Equips the LLM with persistent contextual knowledge about open-source routers.

Approach

GR-Evolve is a code evolution framework wrapping an agentic LLM around an open-source global router. The LLM iteratively edits the router’s source code; each candidate is compiled and evaluated through an integrated OpenROAD QoR pipeline. Persistent context about router internals grounds the LLM, and QoR metrics (notably post-detailed-routing wirelength) steer subsequent mutations.

Experiments

  • Seven benchmark designs spanning three technology nodes.
  • Baselines: existing open-source global routers (unspecified in abstract, likely within OpenROAD).
  • Metric: post-detailed-routing wirelength as primary QoR signal.

Results

Up to 8.72% post-detailed-routing wirelength reduction over baseline routers. Abstract reports only this headline; no runtime, congestion, DRC, or per-design breakdown given here — claim is plausible but narrowly characterized.

Why It Matters

Shifts EDA automation from hyperparameter tuning to algorithm synthesis, suggesting LLMs can specialize tool internals per-design. For AI-infra and EDA practitioners, it hints at a new layer of optimization above traditional autotuning, potentially reducing manual heuristic engineering in routing, placement, and beyond.

Connections to Prior Work

  • LLM-driven code evolution: FunSearch, AlphaEvolve, Eureka.
  • Learning-based EDA optimization and autotuning: e.g., CircuitNet, ML-guided placement/routing.
  • OpenROAD ecosystem and open-source global routers (FastRoute, CU-GR lineage).
  • Agentic LLM coding frameworks applied to systems code.

Open Questions

  • How stable and reproducible are evolved routers across unseen designs?
  • Compute cost per design — is iterative LLM evolution practical at tapeout scale?
  • Do gains hold on congestion, DRC violations, timing, and runtime, not just wirelength?
  • Does evolution produce generalizable algorithmic insights, or one-off overfits?
  • Which LLM, how many iterations, and what’s the failure/regression rate of generated code?
  • Can the paradigm extend to placement, CTS, or detailed routing?

Figures

Figure 1: Figure 1 (extracted from PDF)

Figure 1

Figure 2: Figure 2 (extracted from PDF)

Figure 2

Figure 3: Figure 3 (extracted from PDF)

Figure 3

Original abstract

Modern ASIC design is becoming increasingly complex, driving up design costs while limiting productivity gains from existing EDA tools. Despite decades of progress, current tools rely on fixed heuristics and offer limited control via tool hyperparameters, requiring extensive manual tuning to achieve an acceptable quality of results (QoR). While prior work has explored learning-based optimization and design-specific hyperparameter tuning, these approaches operate within the constraints of static tool algorithm implementations and do not adapt the underlying algorithms to individual designs. To address this limitation, we introduce the concept of design-adaptive EDA tooling, in which the internal algorithms of EDA tools are automatically specialized to the characteristics of a given design. We instantiate this paradigm through GR-Evolve, a code evolution framework that leverages an agentic large language model (LLM) to iteratively modify global routing source code using QoR-driven feedback. The framework equips the LLM with persistent contextual knowledge of open-source global routers along with an integrated toolchain for QoR evaluation within the OpenROAD infrastructure. We evaluate GR-Evolve across seven benchmark designs across three technology nodes and demonstrate up to 8.72% reduction in post-detailed-routing wirelength over existing baseline routers, highlighting the potential of LLM-driven EDA code evolution for design-adaptive global routing.