check-parity

Automate aarch64 JIT parity checks against the generic CPU reference.

438|75|Updated Jan 11, 2015
One-click install
npx skills add https://github.com/sebastianbiallas/pearpc --skill check-parity
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: check-parity
Source: https://github.com/sebastianbiallas/pearpc/tree/main/.claude/skills/check-parity
Command: npx skills add https://github.com/sebastianbiallas/pearpc --skill check-parity

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill automates the validation that the aarch64 JIT interpreter produces behavior identical to the generic CPU reference, across supported opcodes.

Core Features & Use Cases

  • Parity validation across opcode implementations to ensure identical results between JIT and interpreter.
  • Incremental checks with optional opcode-name argument to inspect a single opcode.
  • Audit-ready workflow: includes a stub audit step to catch no-op stubs prior to parity checks.

Quick Start

Run the parity check workflow on your target build to verify that the aarch64 JIT matches the generic CPU behavior.

Frequently Asked Questions about check-parity

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify aarch64 JIT parity with a generic CPU reference?

To verify aarch64 JIT parity, run an automated parity check workflow that cross-checks the JIT interpreter against the generic CPU reference across supported opcodes to ensure identical behavior and exception handling.

Can I check interpreter parity for a single specific opcode?

Yes, you can perform an incremental parity check by providing a specific opcode-name argument to the parity script, allowing you to inspect and validate that individual opcode independently rather than testing all functions.

What does a stub audit do before running JIT parity checks?

A stub audit runs prior to parity checks to catch no-op stubs in the aarch64 JIT implementation, ensuring that incomplete opcode implementations are identified before cross-checking against the generic CPU reference.

Why does my aarch64 JIT produce different exceptions than the generic CPU?

Exceptions differ when the aarch64 JIT lacks parity with the generic CPU reference, requiring an automated parity script to cross-check opcode handling and verify correct exception behavior across implementations.

Does the parity check workflow require any external dependencies?

No, the parity check workflow operates without external dependencies, using an internal parity script and stub audit to validate aarch64 JIT interpreter behavior against the generic CPU reference.