Synthesis of SI Circuits from Burst-Mode Specifications

Abstract

In this paper, we present a new workflow that is based on the conversion of Extended Burst-Mode (XBM) specifications to Signal Transition Graphs (STGs). While XBMs offer a simple design entry to specify asynchronous circuits, they cannot be synthesised into speed-independent (SI) circuits, due to the 'burst mode' timing assumption inherent in the model. Furthermore, XBM synthesis tools are no longer supported, and there are no dedicated tools for formal verification of XBMs. Our approach addresses these issues, by granting the XBMs access to sophisticated synthesis and verification tools available for STGs, as well as the possibility to synthesise SI circuits. Experimental results show that our translation only linearly increases the model size and that our workflow achieves a much improved synthesis success rate, with a 33% average reduction in the literal count.

Burst Automaton: Framework for Speed-Independent Synthesis Using Burst-Mode Specifications

“…An example of this translation is shown in Fig. 15. One can see that this translation is small for practical BAs.…”

“…aà nd a´) with the associated consistency requirements are refinements of the original BA model. Thus, this design flow based on the conversion from XBMs to STGs via BAs, which builds on [15], has been implemented in WORKCRAFT and is shown in Fig. 8.…”

“…In Section III-D, it was described that the XBM's conditionals and "don't cares" can be seen as auxiliary components that are separate from the BA, but can be optionally attached to the BA to create an XBM-like model. In [15], the translation of these XBM components is covered. There, the translation method shows that conditionals are translated into elementary cycles with so-called lock places, while "don't cares" are translated into explicitly delayed STG transitions where they are connected from the first burst they appear in and connected to the burst they terminate in.…”

“…To incorporate these XBM components into our three proposed translations from BA to STG, one can independently translate both conditionals and "don't cares" following the same translation method in [15] and then attach them to the resulting STG from one of our proposed translation, such that they are connected to the transitions and/or places that corresponds with their interacting burst.…”

“…BA relaxes the [X]BM's well-formedness properties and its inherent BM timing assumption to enable a more flexible approach for modelling, and aims to establish the interoperability between many different models including XBMs and STGs, allowing users to easily switch between these models or composing models that are created in different formalisms. We also build on [15] and propose three translations from BMs to STGs, preserving either strong bisimulation, weak bisimulation, or the language respectively, with the latter translation being linear in the XBM's size. The tool support for BAs has been integrated into the WORKCRAFT framework, providing a design flow for BAs and [X]BMs that includes interoperability with FSMs and STGs, formal verification, and synthesis into SI or QDI circuits.…”