Mapping i32.add and local.get directly to x86-64 instructions is precisely what V8's Liftoff baseline compiler does to get Wasm running fast within milliseconds of a page load, before TurboFan's more optimizing tier kicks in for hot functions. Wasmtime's Cranelift backend performs the same register-allocation and jump-patching steps you're documenting here, just with far more sophisticated optimization passes.