Simd Intrinsics

Teaches Claude to read compiler vectorization reports, write SSE2/AVX2/NEON intrinsics, and debug why loops aren't auto-vectorizing. Covers the full workflow from checking GCC's fopt-info-vec output to manually writing intrinsics when the compiler can't optimize your code. Includes runtime CPU feature detection with __builtin_cpu_supports, common auto-vectorization blockers like pointer aliasing, and concrete intrinsics examples for x86 and ARM. The guidance on when to use intrinsics versus letting the compiler handle it is honest: try auto-vectorization first, only drop to intrinsics for hand-tuned shuffles or when the compiler fails. Useful if you're profiling hot loops and need to squeeze out performance without rewriting everything in assembly.

skills.sh installs: 701
GitHub stars: 279
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add mohitmishra786/low-level-dev-skills --skill simd-intrinsics --agent claude-code

Installs into .claude/skills of the current project.

Files

SIMD Intrinsics

Purpose

Guide agents through SIMD: reading auto-vectorization output, writing SSE2/AVX2/NEON intrinsics, runtime CPU feature detection, and choosing between compiler auto-vectorization and manual intrinsics.

Triggers

"How do I check if my loop is being auto-vectorized?"
"How do I write SSE2/AVX2 intrinsics?"
"Auto-vectorization failed — how do I fix it?"
"How do I check for CPU features at runtime?"
"Should I use intrinsics or let the compiler vectorize?"
"How do I write NEON intrinsics for ARM?"

Workflow

1. Check auto-vectorization

# GCC: show vectorization info
gcc -O2 -march=native -fopt-info-vec src/hot.c -o hot

# Verbose: show missed + successful
gcc -O2 -march=native -fopt-info-vec-missed -fopt-info-vec-optimized src/hot.c

# Clang: vectorization remarks
clang -O2 -march=native \
    -Rpass=loop-vectorize \
    -Rpass-missed=loop-vectorize \
    -Rpass-analysis=loop-vectorize \
    src/hot.c -o hot

# Example missed message:
# hot.c:15:5: remark: loop not vectorized: value that could not be identified as
# reduction is used outside the loop [-Rpass-missed=loop-vectorize]

Common auto-vectorization blockers:

SKILL.md on GitHub