Apparently performances of libsodium.js appear to be better than argon2-browser

[evilaliv3]

Apparently performances of libsodium.js appear to be better than argon2-browser, even if both the two are wasm implementations.

@antelle @jedisct1 do you know the reasons?

At GlobaLeaks we look forward to use libsodium in the browser, but at the moment we have the requirement of just running argon2. For this reason we were hoping to run with argon2-browser to have smaller footprint but apparently 10x slower than the libsodium.js implementation.

Thank you if you could provide any hint.

For compatibility reasons we are not evaluating to use the SIMD implementation.

[msmannan00]

starting to analysis the cases, most probably if both are using wasm and are eventually doing same thing the performance difference should not be there, will review it out