The Phantom of PCIe: Constraining Generative Artificial Intelligences for Practical Peripherals Trace Synthesizing

TL;DR

This paper introduces Phantom, a framework that combines generative AI with a constraint-enforcing filter to produce realistic PCIe TLP traces, overcoming hallucination issues for device simulation.

Contribution

Phantom systematically addresses AI hallucinations in TLP synthesis by enforcing PCIe constraints, enabling high-fidelity, large-scale trace generation for practical device testing.

Findings

Phantom significantly outperforms existing models with up to 1000× improvement in task metrics.

It achieves up to 2.19× better FID scores compared to backbone-only methods.

The prototype implementation is open-source for community use.

Abstract

Peripheral Component Interconnect Express (PCIe) is the de facto interconnect standard for high-speed peripherals and CPUs. The development of PCIe devices for emerging applications requires realistic Transaction Layer Packet (TLP) traces that accurately simulate device-CPU interactions. While generative AI offers a promising avenue for synthesizing complex TLP sequences, it is prone to a critical challenge inherent in all generation tasks: hallucination. Naively applying these models often produces traces that violate fundamental PCIe protocol rules, such as ordering and causality, rendering them unusable for device simulation. To resolve this, our work introduces a methodology to bridge the gap between generative AI and high-fidelity device simulation. This paper presents Phantom, a framework that systematically addresses AI-generated hallucinations in TLP synthesis. Phantom achieves this by coupling a generative backbone with a novel post-processing filter that enforces PCIe-specific constraints, effectively eliminating invalid TLP sequences. We validate Phantom's effectiveness by synthesizing TLP traces for an actual PCIe network interface card. Experimental results show that Phantom produces practical, large-scale TLP traces, significantly outperforming existing models, with improvements of up to 1000$\times$ in task-specific metrics and up to 2.19$\times$ in Fr\'echet Inception Distance (FID) compared to backbone-only methods. The prototype implementation has been made open-source.