Why WebAssembly isn't truly a stack machine: A technical analysis
By
signa11
Front-window bakery material. Catches the eye, delivers the goods.
Summary
The article challenges the common characterization of WebAssembly (Wasm) as a stack machine. The author, after hand-writing Wasm code, discovered a fundamental difference between Wasm and traditional stack-based languages. While Wasm uses a stack for instruction operands, it differs from classic stack machines in key ways that make the "stack machine" label misleading. The article explores the distinction between register-based and stack-based architectures, arguing that Wasm's design has important structural differences from other stack-based languages.
Key quotes
· 3 pulledEveryone knows Wasm is a stack machine. Wikipedia says so, the official Wasm design specification says so, you get it. I thought so too.
That is, until I started writing Wasm code – not compiling for Wasm, but writing the instructions by hand. And I found out that there exists a major difference between Wasm and all other stack-based languages, that makes this claim misleading.
What is a stack machine, even?
You might also wanna read
Tokyo AI Event Addresses Engineering Challenges of Production-Grade AI Systems
This article covers a Tokyo AI (TAI) community event focused on the engineering and operational challenges of transitioning AI from experime
luma.com·46m ago
Copyparty: A Lightweight File Server That Runs as a Single Python Script
Copyparty is a lightweight, full-featured file server that runs as a single Python script, making it extremely easy to set up without needin
hackaday.com·54m ago
DeepSeek-V4: Hybrid Sparse-Attention Architecture Enables Efficient Million-Token Context Inference
DeepSeek-V4 introduces a hybrid sparse-attention architecture combined with on-policy distillation across domain specialists, enabling 1M-to
artgor.medium.com·59m ago
How to Deploy Self-Hosted OpenClaw AI Agent on a Debian VPS
This article provides a step-by-step guide for deploying OpenClaw, an open-source personal AI agent, on a Debian VPS server. It covers the s
blog.radwebhosting.com·2h ago
Linux kernel OOM killer discussion: protecting critical processes from memory pressure (2004)
A Linux kernel mailing list email thread from 2004 where Thomas Habets proposes a sysctl mechanism to protect specific processes (like xlock
lwn.net·3h ago
Netflix engineer's open-source tool cuts AI token usage by up to 90%
Netflix senior engineer Tejas Chopra created software called "Project Headroom" that prunes redundant tokens from AI agent instructions befo
theregister.com·4h ago