This is a decent writeup. And kudos to Figma for releasing IndirectBuffer as open source. Currently lots of dark arts around offloading compute to gpu and asm.j

This is a decent writeup. And kudos to Figma for releasing IndirectBuffer as open source. Currently lots of dark arts around offloading compute to gpu and asm.js, but I think as browsers expose APIs such as SharedArrayBuffer and WebGPU some of the chaos should abate. Am interested in similar gains for high performance browser networking. Some experiments with wasm in linux kernel show safe bypass is a possibility ;) https://github.com/wasmerio/kernel-wasm https://github.com/wasmerio/kernel-wasm