Application-Level Validation of Accelerator Designs Using a Formal Software/Hardware Interface

TL;DR

This paper introduces 3LA, a formal methodology leveraging Instruction-Level Abstraction (ILA) to enable automated, end-to-end testing of accelerator designs on unmodified applications, reducing development costs and effort.

Contribution

The work presents a novel formal software/hardware interface for accelerators using ILA, enabling automated compiler and simulator development for prototype accelerators.

Findings

Uncovered a previously unknown flaw in a published accelerator design.

Demonstrated successful end-to-end testing of accelerators with modest engineering effort.

Extended existing techniques to support auto-generation of compiler support for accelerators.

Abstract

Ideally, accelerator development should be as easy as software development. Several recent design languages/tools are working toward this goal, but actually testing early designs on real applications end-to-end remains prohibitively difficult due to the costs of building specialized compiler and simulator support. We propose a new first-in-class, mostly automated methodology termed "3LA" to enable end-to-end testing of prototype accelerator designs on unmodified source applications. A key contribution of 3LA is the use of a formal software/hardware interface that specifies an accelerator's operations and their semantics. Specifically, we leverage the Instruction-Level Abstraction (ILA) formal specification for accelerators that has been successfully used thus far for accelerator implementation verification. We show how the ILA for accelerators serves as a software/hardware interface, similar to the Instruction Set Architecture (ISA) for processors, that can be used for automated development of compilers and instruction-level simulators. Another key contribution of this work is to show how ILA-based accelerator semantics enables extending recent work on equality saturation to auto-generate basic compiler support for prototype accelerators in a technique we term "flexible matching." By combining flexible matching with simulators auto-generated from ILA specifications, our approach enables end-to-end evaluation with modest engineering effort. We detail several case studies of 3LA, which uncovered an unknown flaw in a recently published accelerator and facilitated its fix.