That's a great analysis!
We have some room for improvement there 
-
-
Replying to @syrusakbary
The benchmark can be reproduced with dist-build/wasm32-wasi.sh --bench (included in the libsodium source code). I was quite surprised to see very different results for WASM/LLVM and WAVM.
2 replies 0 retweets 1 like -
Replying to @jedisct1
Syrus Akbary Retweeted Wasmer
We just fixed one of the issues that was causing the slowdown. The speed with the LLVM backend is now a bit closer to native
https://twitter.com/wasmerio/status/1187168208756756481 …
PS: there is still some room for improvement... we are working on it!Syrus Akbary added,
1 reply 0 retweets 2 likes
Is there a massive difference in compile time between LLVM with the most expensive optimizations disabled and Cranelift?
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
It's release time!
We improved the runtime speed (LLVM backend) in Wasmer 0.9.0 by 1.5~2x