Porting architecture specific intrinsics
Introduction
Code Migration to Arm
Use sse2neon to port code to Arm
Use SIMD Everywhere to port code to Arm
Find intrinsics in large code bases
Next Steps

  Log an issue

  Fork and edit

  Discuss on Discord

About this Learning Path

Skill level:  Advanced
Reading time:  30 min
Last updated:  03 Aug 2026
Skill level:
  Advanced
Reading time:
  30 min
Last updated:
  03 Aug 2026
Author:Jason Andrews, Arm
Arm IP:Neoverse Cortex-A
Tags: Performance and Architecture Linux Neon SVE Intrinsics Runbook
Author:
  • Jason Andrews, Arm
Arm IP:
Neoverse Cortex-A
Tags:
Performance and Architecture Linux Neon SVE Intrinsics Runbook

Who is this for?

This is an advanced topic for software developers interested in porting architecture specific intrinsics to Arm processors.

What will you learn?

Upon completion of this Learning Path, you will be able to:

  • Describe what intrinsics are and how to find them in code.
  • Evaluate options and use header-only libraries to port architecture-specific intrinsics to Arm.

Prerequisites

Before starting, you will need the following:

  • Some understanding of SIMD concepts.
  • An Arm based machine or cloud instance running Ubuntu Linux.
  • Optionally, an x86_64 machine also running Ubuntu.

Summary

AI-assisted

This summary was drafted with an approved AI-assisted workflow and reviewed by Arm contributors before publication. Human technical review remains part of the process so the final page reflects engineering rigor, accuracy, and Arm editorial standards.

Close
?
You’ll port x64 SIMD code to Arm by identifying architecture-specific intrinsics and mapping them to Arm equivalents. You’ll use sse2neon for SSE or MMX code, or SIMD Everywhere (SIMDe) when a project also contains AVX. You’ll adjust headers, add the required shim or macros, rebuild on Arm, and use Porting Advisor for Graviton to locate intrinsics in larger repositories.

Frequently asked questions

AI-assisted

These FAQs were drafted with an approved AI-assisted workflow and reviewed by Arm contributors before publication. Human technical review remains part of the process so the final page reflects engineering rigor, accuracy, and Arm editorial standards.

Close
?
How do I know if my code uses architecture-specific intrinsics?
Look for function-like calls from SIMD headers tied to a specific instruction set architecture (ISA), such as SSE, MMX, or AVX on x64. These compiler-provided functions aren’t standard library APIs.
Which option should I use if my codebase mixes SSE and AVX?
Use SIMD Everywhere (SIMDe). SIMDe supports SSE, MMX, and AVX, while sse2neon targets SSE and MMX.
What changes do I need when applying `sse2neon`?
Adjust the SSE-specific headers for Arm, include sse2neon.h, and update the compiler flags for the Arm target. Then, rebuild on an Arm-based platform.
What do I configure when using SIMDe?
Select the SIMDe header that matches your intrinsics, define the macro shown in the project documentation, and rebuild. SIMDe then maps the intrinsics to portable Arm implementations.
How can I scan a large repository for intrinsics before compiling?
Use Porting Advisor for Graviton to assess portability and locate intrinsics across the codebase. Install and run the tool, then review its report before editing the sources.
Next