Microarchitecture Design and Benchmarking of Custom SHA-3 Instruction for RISC-V

• URL: http://arxiv.org/abs/2508.20653v1
• Date: Thu, 28 Aug 2025 10:57:13 GMT
• Title: Microarchitecture Design and Benchmarking of Custom SHA-3 Instruction for RISC-V
• Abstract summary: Hardware-assisted cryptographic instructions, such as Intel's AES-NI and ARM's custom instructions for encryption workloads, have demonstrated substantial performance improvements.<n>However, efficient SHA-3 acceleration remains an open problem due to its distinct permutation-based structure and memory access patterns.<n>This study investigates the architectural challenges of embedding a SHA-3 permutation operation as a custom instruction within a general-purpose processor.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Integrating cryptographic accelerators into modern CPU architectures presents unique microarchitectural challenges, particularly when extending instruction sets with complex and multistage operations. Hardware-assisted cryptographic instructions, such as Intel's AES-NI and ARM's custom instructions for encryption workloads, have demonstrated substantial performance improvements. However, efficient SHA-3 acceleration remains an open problem due to its distinct permutation-based structure and memory access patterns. Existing solutions primarily rely on standalone coprocessors or software optimizations, often avoiding the complexities of direct microarchitectural integration. This study investigates the architectural challenges of embedding a SHA-3 permutation operation as a custom instruction within a general-purpose processor, focusing on pipelined simultaneous execution, storage utilization, and hardware cost. In this paper, we investigated and prototyped a SHA-3 custom instruction for the RISC-V CPU architecture. Using cycle-accurate GEM5 simulations and FPGA prototyping, our results demonstrate performance improvements of up to 8.02x for RISC-V optimized SHA-3 software workloads and up to 46.31x for Keccak-specific software workloads, with only a 15.09% increase in registers and a 11.51% increase in LUT utilization. These findings provide critical insights into the feasibility and impact of SHA-3 acceleration at the microarchitectural level, highlighting practical design considerations for future cryptographic instruction set extensions.

Related papers

• SABLE: Minimalist Instruction-Level Authenticated Encryption for Constrained Confidential Computing [3.0128599957899858]
  We present a design exploration of a RISC-V processor architecture that enables instruction-level authenticated encryption of programs.<n>SABLE is agnostic to the underlying micro-architecture, remaining compatible with the standard RISC-V tool.<n>We implement and evaluate the designs using ASCON-128a on a Xilinx Artix-7 FPGA with the open-source NEORV32 system on chip.
  arXiv  Detail & Related papers  (2026-07-18T11:12:28Z)
• AutoVecCoder: Teaching LLMs to Generate Explicitly Vectorized Code [69.84692701646769]
  AutoVecCoder is a novel framework designed to empower Large Language Models with the capability of automated explicit vectorization.<n>VecPrompt, an automated data synthesis pipeline, injects domain-specific intrinsic knowledge.<n>VecRL, a reinforcement learning framework, aligns code generation with execution efficiency.<n>AutoVecCoder-8B trained by this framework achieves state-of-the-art performance on the SSE and AVX subsets of SimdBench.
  arXiv  Detail & Related papers  (2026-05-18T07:33:15Z)
• CryptRISC: A Secure RISC-V Processor for High-Performance Cryptography with Power Side-Channel Protection [1.5866931449827322]
  CryptRISC is the first RISC-V processor that combines cryptographic acceleration with hardware-level power side-channel resistance.<n>Masking is a widely used countermeasure, yet software-based techniques often introduce significant performance overhead and implementation complexity.
  arXiv  Detail & Related papers  (2026-02-23T19:13:35Z)
• Magellan: Autonomous Discovery of Novel Compiler Optimization Heuristics with AlphaEvolve [44.73800369169414]
  We present Magellan, an agentic framework that evolves the compiler pass itself by executable C++ decision logic.<n>M Magellan couples an LLM coding agent with evolutionary search and autotuning in a closed loop of generation, evaluation on user-provided macro-benchmarks, and refinement.<n>We report preliminary results on XLA problems, demonstrating portability beyond LLVM with reduced engineering effort.
  arXiv  Detail & Related papers  (2026-01-28T22:34:56Z)
• KScaNN: Scalable Approximate Nearest Neighbor Search on Kunpeng [46.35664429179457]
  A naive port of existing x86 ANNS algorithms to ARM platforms results in a substantial performance deficit.<n>We introduce KScaNN, a novel ANNS algorithm co-designed for the Kunpeng 920 ARM architecture.
  arXiv  Detail & Related papers  (2025-11-05T09:01:32Z)
• Eliminating Multi-GPU Performance Taxes: A Systems Approach to Efficient Distributed LLMs [61.953548065938385]
  We introduce the ''Three Taxes'' (Bulk Synchronous, Inter- Kernel Data Locality, and Kernel Launch Overhead) as an analytical framework.<n>We propose moving beyond the rigid BSP model to address key inefficiencies in distributed GPU execution.<n>We observe a 10-20% speedup in end-to-end latency over BSP-based approaches.
  arXiv  Detail & Related papers  (2025-11-04T01:15:44Z)
• Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation [129.45368843861917]
  We introduce the Gated Memory Unit (GMU), a simple yet effective mechanism for efficient memory sharing across layers.<n>We apply it to create SambaY, a decoder-hybrid-decoder architecture that incorporates GMUs to share memory readout states from a Samba-based self-decoder.
  arXiv  Detail & Related papers  (2025-07-09T07:27:00Z)
• EAQuant: Enhancing Post-Training Quantization for MoE Models via Expert-Aware Optimization [46.40666108181214]
  Mixture-of-Experts (MoE) models have emerged as a cornerstone of large-scale deep learning.<n>MoE models have inherent complexities that challenge conventional quantization techniques.<n>We propose EAQuant, a novel PTQ framework tailored for MoE architectures.
  arXiv  Detail & Related papers  (2025-06-16T10:18:50Z)
• Design and Implementation of a RISC-V SoC with Custom DSP Accelerators for Edge Computing [0.0]
  We examine the RV32I base instruction set with extensions for multiplication (M) and atomic operations (A)<n>Our results demonstrate RISC-V's advantages in embedded systems and its scalability for custom accelerators.
  arXiv  Detail & Related papers  (2025-06-07T07:17:40Z)
• gECC: A GPU-based high-throughput framework for Elliptic Curve Cryptography [15.39096542261856]
  Elliptic Curve Cryptography (ECC) is an encryption method that provides security comparable to traditional techniques like Rivest-Shamir-Adleman (RSA)<n> ECC is still hindered by the significant performance overhead associated with elliptic curve (EC) operations.<n>This paper presents gECC, a versatile framework for ECC optimized for GPU architectures.
  arXiv  Detail & Related papers  (2024-12-22T01:50:50Z)
• Mixed-precision Neural Networks on RISC-V Cores: ISA extensions for Multi-Pumped Soft SIMD Operations [5.847997723738113]
  Modern embedded microprocessors provide very limited support for mixed-precision NNs. We introduce a hardware-software co-design framework that enables cooperative hardware design, mixed-precision quantization, ISA extensions and inference. Our framework can achieve, on average, 15x energy reduction for less than 1% accuracy loss and outperforms the ISA-agnostic state-of-the-art RISC-V cores.
  arXiv  Detail & Related papers  (2024-07-19T12:54:04Z)
• A 3D Memristor Architecture for In-Memory Computing Demonstrated with SHA3 [1.024113475677323]
  Security is a growing problem that needs hardware support. Memristors provide an alternative technology for hardware-supported security implementation. This paper presents a specific technique that utilizes the benefits of hybrid CMOS-memristors technology demonstrated with SHA3 over implementations that use only memristor technology.
  arXiv  Detail & Related papers  (2024-02-14T19:43:18Z)
• Using the Abstract Computer Architecture Description Language to Model AI Hardware Accelerators [77.89070422157178]
  Manufacturers of AI-integrated products face a critical challenge: selecting an accelerator that aligns with their product's performance requirements. The Abstract Computer Architecture Description Language (ACADL) is a concise formalization of computer architecture block diagrams. In this paper, we demonstrate how to use the ACADL to model AI hardware accelerators, use their ACADL description to map DNNs onto them, and explain the timing simulation semantics to gather performance results.
  arXiv  Detail & Related papers  (2024-01-30T19:27:16Z)
• SOCI^+: An Enhanced Toolkit for Secure OutsourcedComputation on Integers [50.608828039206365]
  We propose SOCI+ which significantly improves the performance of SOCI. SOCI+ employs a novel (2, 2)-threshold Paillier cryptosystem with fast encryption and decryption as its cryptographic primitive. Compared with SOCI, our experimental evaluation shows that SOCI+ is up to 5.4 times more efficient in computation and 40% less in communication overhead.
  arXiv  Detail & Related papers  (2023-09-27T05:19:32Z)
• MAPLE: Microprocessor A Priori for Latency Estimation [81.91509153539566]
  Modern deep neural networks must demonstrate state-of-the-art accuracy while exhibiting low latency and energy consumption. Measuring the latency of every evaluated architecture adds a significant amount of time to the NAS process. We propose Microprocessor A Priori for Estimation Estimation MAPLE that does not rely on transfer learning or domain adaptation.
  arXiv  Detail & Related papers  (2021-11-30T03:52:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.