E2ESlack: An End-to-End Graph-Based Framework for Pre-Routing Slack Prediction

TL;DR

E2ESlack introduces a true end-to-end, graph-based framework for pre-routing slack prediction, addressing the missing RAT estimation crucial for TNS/WNS computation. It combines a TimingParser that converts LIB/SDF/DEF inputs into PyTorch DGL graphs, an arrival-time predictor, and a fast RAT estimation module that accounts for clock reconvergence pessimism to produce corrected RAT and Slack values. The method achieves RAT MAE of 0.6454 across circuit endpoints, outperforming adapted TimingPredict and pre-routing STA, while delivering TNS/WNS results comparable to post-routing STA and offering up to 23x runtime savings. This approach enables engineers to estimate timing viability earlier in the design flow, accelerating iteration and reducing cost without routing.

Abstract

Pre-routing slack prediction remains a critical area of research in Electronic Design Automation (EDA). Despite numerous machine learning-based approaches targeting this task, there is still a lack of a truly end-to-end framework that engineers can use to obtain TNS/WNS metrics from raw circuit data at the placement stage. Existing works have demonstrated effectiveness in Arrival Time (AT) prediction but lack a mechanism for Required Arrival Time (RAT) prediction, which is essential for slack prediction and obtaining TNS/WNS metrics. In this work, we propose E2ESlack, an end-to-end graph-based framework for pre-routing slack prediction. The framework includes a TimingParser that supports DEF, SDF and LIB files for feature extraction and graph construction, an arrival time prediction model and a fast RAT estimation module. To the best of our knowledge, this is the first work capable of predicting path-level slacks at the pre-routing stage. We perform extensive experiments and demonstrate that our proposed RAT estimation method outperforms the SOTA ML-based prediction method and also pre-routing STA tool. Additionally, the proposed E2ESlack framework achieves TNS/WNS values comparable to post-routing STA results while saving up to 23x runtime.

Figures (1)

• Figure 1: Framework Architecture. Blue arrows indicate the workflow for label generation, which is only required for training. Black arrows indicate the inference workflow.