Robin Freyler

Thank you for those benchmarks! They are really insightful. Seems like there is some room for improvement for wasmi. Your toywasm looks very strong! :)

@yamt thanks to your memory consumption benchmarks I took a better look at `wasmi`'s internal bytecode and was able to make it more space efficient while keeping the original performance...

> noted for the next run of the benchmark. looking forward :) > (probably not too distant future, as i want to see how toywasm regressed by the recent simd...

> having said that, these "do more work per instruction" style instructions can be rather friendly to interpreters like toywasm because it can hide instruction-fetch-parse overhead. At Parity we even...

> in toywasm, the value stack cell size depends on build-time configurations. ah, that's very interesting and perfect for research about what cell size is the best for which use...

What just crossed my mind about cell sizes and SIMD support is the following: Maybe it is practical and efficient to have 2 different stacks, e.g. one stack with 64-bit...

Awesome work @yamt and thanks a ton for those benchmarks! 🚀 I am especially fond of the fact that there is nearly no difference between `toywasm (SIMD)` and `toywasm (no...

Ah I thought your were simply installing `wasmi` via `cargo install wasmi_cli`. The Cargo bug that not all optimizations are properly applied mostly affects certain binaries installed via `cargo install`....

If you want to benchmark `wasmi` with full optimizations and build it from sources you can build it via: ``` cargo build --profile bench ``` So instead of `--release` you...

Hi @yamt , thanks a lot for updating me about this! The new `wasmi` results look much more as I would expect, being roughly twice as slow as Wasm3. It...