SIMD Everywhere Optimization from ARM NEON to RISC-V Vector Extensions

TL;DR

This paper presents an enhanced version of the open source tool SIMDe to automate the migration of ARM NEON SIMD code to RISC-V Vector Extensions, significantly improving performance in real-world libraries.

Contribution

The authors extend SIMDe to support ARM NEON to RISC-V Vector Extensions conversion, enabling automated, efficient migration of SIMD code with customized function conversions.

Findings

Achieved speedup of 1.51x to 5.13x in XNNPACK library using enhanced SIMDe.

Developed strategies for type conversion considering vector length agnostic architectures.

Customized function conversions lead to significant performance improvements.

Abstract

Many libraries, such as OpenCV, FFmpeg, XNNPACK, and Eigen, utilize Arm or x86 SIMD Intrinsics to optimize programs for performance. With the emergence of RISC-V Vector Extensions (RVV), there is a need to migrate these performance legacy codes for RVV. Currently, the migration of NEON code to RVV code requires manual rewriting, which is a time-consuming and error-prone process. In this work, we use the open source tool, "SIMD Everywhere" (SIMDe), to automate the migration. Our primary task is to enhance SIMDe to enable the conversion of ARM NEON Intrinsics types and functions to their corresponding RVV Intrinsics types and functions. For type conversion, we devise strategies to convert Neon Intrinsics types to RVV Intrinsics by considering the vector length agnostic (vla) architectures. With function conversions, we analyze commonly used conversion methods in SIMDe and develop customized conversions for each function based on the results of RVV code generations. In our experiments with Google XNNPACK library, our enhanced SIMDe achieves speedup ranging from 1.51x to 5.13x compared to the original SIMDe, which does not utilize customized RVV implementations for the conversions.