Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 1711.02333

Critique of "Asynchronous Logic Implementation Based on Factorized DIMS"

P Balasubramanian
View original source ↗Read source document ↗Export metadata ↓

Abstract

This paper comments on "Asynchronous Logic Implementation Based on Factorized DIMS" [Journal of Circuits, Systems, and Computers, vol. 26, no. 5, 1750087: 1-9, May 2017] with respect to two main problematic issues: i) the gate orphan problem implicit in the factorized DIMS approach discussed in the referenced article which affects its strong-indication, and ii) how the enumeration of product terms to represent the synthesis cost is skewed in the referenced article because the logic expression contains sum of products and also product of sums. It is observed that the referenced article has not provided a general logic synthesis algorithm excepting only an example illustration involving a 3-input AND logic function. The absence of a general logic synthesis algorithm would make it difficult to reproduce the research described in the referenced article. Moreover, the example illustration in the referenced article describes an unsafe logic decomposition which is not suitable for the multi-level synthesis of strong-indication asynchronous circuits. Further, a logic synthesis method which safely decomposes the DIMS solution to synthesize multi-level strong-indication asynchronous circuits is available in the existing literature, which was neither cited nor taken up for comparison in the referenced article, which is another drawback. Subsequently, it is concluded that the referenced article has not advanced existing knowledge in the field but on the contrary, has caused confusions. Hence, in the interest of readers, this paper additionally highlights some important and relevant literature which provide valuable information about robust asynchronous circuit synthesis techniques which employ delay-insensitive codes for data representation and processing and the 4-phase return-to-zero handshake protocol for data communication.

Explore related subjects

cs.AR

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

P Balasubramanian. 2018-09-23. Critique of "Asynchronous Logic Implementation Based on Factorized DIMS". https://arxiv.org/abs/1711.02333

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

HLSmith: An Expert-Guided Agentic Framework for C/C++-to-HLS Translation

Application-specific FPGA accelerators offer substantial performance and energy-efficiency gains across many application domains, but developing them is costly, often requiring months of specialized effort. Even with high-level synthesis (HLS), designers still need extensive hardware expertise to build high-performance accelerators. Although large language models (LLMs) have demonstrated strong software-generation capabilities, even frontier models lack the hardware intuition and procedural knowledge needed to reliably translate baseline C/C++ programs into high-performance HLS designs: they struggle to identify effective architectures, follow the optimization processes used by HLS experts, and apply hardware transformations consistently across diverse kernels. We present HLSmith, an expert-guided framework for translating C/C++ programs into optimized HLS accelerators. HLSmith combines three components: an HLS optimization expertise library that encodes guarded transformation recipes, their applicability and prerequisite conditions, and unsafe cases to avoid; a staged, feedback-driven orchestration flow modeled on expert HLS development practice that guides agents through synthesis, bottleneck analysis, and optimization; and a tool-grounded model-adaptation pipeline that converts optimization trajectories from commercial frontier models into training data for fine-tuning open-weight LLMs. We evaluate HLSmith on PolyBench against ChatHLS, a leading prior agent-orchestration framework for HLS accelerator development. HLSmith achieves a geometric mean speedup of 4.24x over ChatHLS while producing functionally correct designs, in both software and RTL simulation, for every benchmark, compared with ChatHLS's 57% valid-design rate. It further reaches speedups of up to 252x and 138x with commercial frontier models and open-weight models, respectively.

cs.AR↗
arXiv2026

Accelerating the Mitigation of LLM Inference Nondeterminism Across GPU Architectures

Large language model (LLM) outputs are expected to be reproducible under greedy decoding, yet in practice the same model, prompt, and software stack produce different outputs on different GPUs. The root cause is floating-point non-associativity combined with hardware-dependent kernel selection. Inference frameworks select different matrix-multiplication kernels on each architecture, with different parallel reduction orders and unspecified tensor-core arithmetic, and the resulting rounding differences can flip output tokens. Existing solutions have imperfect cross-architecture reproducibility and incur a significant performance penalty. We present a solution employing a set of fixed-configuration fused-upcast GEMM kernels that load 16-bit weights from memory, upcast them to FP32 in registers, and accumulate with IEEE-754 arithmetic in a reduction order that is a pure function of the problem shape and is therefore independent of the device, its SM count, or kernel scheduling. By fixing the floating-point reduction order as a function of problem shape alone, every GPU runs the same operation sequence, so cross-architecture reproducibility of the linear layers reduces to correct IEEE-754 arithmetic rather than to rounding differences staying below a tie-flip threshold. We confirm our solution's linear-layer outputs are bitwise identical across NVIDIA Ampere, Ada, and Hopper GPUs, while running $1.17$ to $3.1\times$ faster end-to-end than the state-of-the-art solution and cutting weight-memory traffic in half.

cs.AR↗
arXiv2026

Decoupling Logical Masks from GPU Execution for Dynamic Block-Sparse Attention

Attention computation makes inference expensive in video diffusion transformers (vDiTs), which generate videos through iterative denoising. Block-sparse attention (BSA) reduces this cost by computing only blocks selected by a logical mask, which specifies attention interactions to compute. However, coupling logical block geometry to execution choices limits adaptation to varying masks and graphics processing units (GPUs), while runtime kernel specialization can incur preparation overhead that outweighs execution time savings. We present Tessera, a specialized runtime for dynamic BSA that decouples logical masks from GPU execution while preserving specified attention interactions. Its physical mapping layer retains, combines, or subdivides logical attention blocks into physical tiles suited to different attention mask shapes and GPU architectures. Its task organization layer groups and schedules tiles within GPU tasks to reuse data, expose parallelism, and overlap data movement with computation. Finally, profile-guided regime selection enables low- overhead execution plan selection through a lookup table constructed from offline profiling. We implement Tessera with specialized CUDA kernels supporting four NVIDIA GPU generations. Evaluated on 2,315 real attention masks and industrial video diffusion models, Tessera achieves up to 6.79x BSA request speedup over baseline systems in the evaluated video diffusion models.

cs.AR↗