CHiPReP—A Compiler for the HiPReP High-Performance Reconfigurable Processor

This article presents CHiPReP, a C compiler for the HiPReP processor, which is a high-performance Coarse-Grained Reconfigurable Array employing Floating-Point Units. CHiPReP is an extension of the LLVM and CCF compiler frameworks. Its main contributions are (i) a Splitting Algorithm for Data Dependence Graphs, which distributes the computations of a C loop to Address-Generator Units and Processing Elements; (ii) a novel instruction clustering and scheduling heuristic; and (iii) an integrated placement, pipeline balancing and routing optimization method based on Simulated Annealing. The compiler was verified and analyzed using a cycle-accurate HiPReP simulation model.

Similarity-Aware Architecture/Compiler Co-Designed Context-Reduction Framework for Modulo-Scheduled CGRA

Modulo-scheduled coarse-grained reconfigurable array (CGRA) processors have shown their potential for exploiting loop-level parallelism at high energy efficiency. However, these CGRAs need frequent reconfiguration during their execution, which makes them suffer from large area and power overhead for context memory and context-fetching. To tackle this challenge, this paper uses an architecture/compiler co-designed method for context reduction. From an architecture perspective, we carefully partition the context into several subsections and only fetch the subsections that are different to the former context word whenever fetching the new context. We package each different subsection with an opcode and index value to formulate a context-fetching primitive (CFP) and explore the hardware design space by providing the centralized and distributed CFP-fetching CGRA to support this CFP-based context-fetching scheme. From the software side, we develop a similarity-aware tuning algorithm and integrate it into state-of-the-art modulo scheduling and memory access conflict optimization algorithms. The whole compilation flow can efficiently improve the similarities between contexts in each PE for the purpose of reducing both context-fetching latency and context footprint. Experimental results show that our HW/SW co-designed framework can improve the area efficiency and energy efficiency to at most 34% and 21% higher with only 2% performance overhead.

M-CUBE

Millimeter-wave (mmWave) technologies represent a cornerstone for emerging wireless network infrastructure, and for RF sensing systems in security, health, and automotive domains. However, the lack of an experimental platform has been impeding research in this field. In this article, we propose to fill the gap with M3 (M-Cube), the first mmWave massive MIMO software radio. M3 features a fully reconfigurable array of phased arrays, with up to 8 RF chains and 256 antenna elements. Despite the orders of magnitude larger antenna arrays, its cost is orders of magnitude lower, even when compared with state-of-the-art single RF chain mmWave software radios. Case studies have demonstrated the usefulness of M3 design for research in mmWave massive MIMO communication and sensing.

Partially activated reconfigurable arrays to guide acoustic waves

Recent studies have exemplified the potential for curved origami-inspired acoustic arrays to focus waves. Yet, reconfigurable structures that adopt curvatures are often difficult to translate to practice due to mechanical deformation of the facets that inhibit straightforward folding. In addition, not all tessellations that curve upon folding are also flat-foldable, which is a key advantage of portability inherent to many origami-inspired structures. This research introduces a new concept of partially activated reconfigurable acoustic arrays as a means to mitigate these drawbacks. Here, tessellations are studied where a subset of the facet surfaces are considered to radiate acoustic waves. The analytical results reveal focusing behaviors in such arrays that are otherwise not manifest for the array when fully activated. The focused waves are more intense in amplitude and space for partially activated arrays than fully activated counterparts. These trends are verified by experiment and are also found to be applicable to multiple reconfigurable array geometries. The results encourage broader study of the design space accessible in reconfigurable arrays to capitalize on all of the functionality afforded by origami-inspired wave guiding structures.