cs.ARNov 9, 2026

To SIMD or Not to SIMD: Comparing RVV vs. CGRA SoCs for Non-Linear Functions in AI

Authors: Sameer Deshmukh, Naoya Mizuki, Takumi Honda

Organizations: Fujitsu Research of Japan, Fujitsu Limited, Kawasaki, Japan

Abstract

Non-linear functions are an important class of operations in modern large language models. However, chip vendors building for AI have so far relied on traditional vector or scalar processing units for non-linear functions. The performance of these units has not kept pace with matrix multiplication units (MMUs), which can be found in both high-performance and edge-computing devices from almost every vendor. Reconfigurable units such as FPGAs and CGRAs have been proposed in the literature, but have not been systematically compared with traditional units backed by MMUs. In this paper, we build two SoCs: one using an RVV vector processor and the other using a latency-insensitive (elastic) CGRA, both coupled with an MMU and driven by a single-core CPU. We then propose mappings for common feed-forward networks on our SoCs and find that the CGRA reaches a maximum speedup in clock cycles of 7.12x over RVV on the activation stage of simple point-wise non-linear functions like SiLU, whereas the RVV reaches a maximum speedup of 9.8x over the CGRA on complex non-linear functions such as softmax\mathrm{softmax}. Over the enclosing feed-forward block the corresponding maxima are 3.16x and 4.24x.

Figures & tables

Explore similar work

CardsList
  1. GRAU: A Generic Runtime-Reconfigurable Activation-Requantization Unit for Quantized Neural Network Accelerators

    Feb 25, 2026Yuhao Liu, Salim Ullah, Akash KumarNeural Network Hardware AccelerationNeural Network Activation Functions

  2. MIVE: A Minimalist Integer Vector Engine for Softmax LayerNorm and RMSNorm Acceleration

    Jun 16, 2026Kosmas Alexandridis, Giorgos DimitrakopoulosAI Accelerator InferenceInference Acceleration

  3. Fast & Faithful Function Vectors

    Jun 3, 2026Minh An Pham, Anton Segeler, Thomas Wiegand +4Task VectorsLLM Interpretability