Nevkontakte

I created the new [generics](https://github.com/gopherjs/gopherjs/tree/generics) branch for this work and enabled the `typeparam` test suite from the Go repository in it. This should give us a decent idea for the...

Since I haven't _really_ started coding yet, I can't point at anything specific 😅 My general plan is to [look at CI](https://app.circleci.com/pipelines/github/gopherjs/gopherjs?branch=generics), pick a failing test case and try to...

Good question, I assumed it should be more or less neutral, but the GoroutineSwitching benchmark shows a significant degradation: ``` $ benchstat /tmp/old.txt /tmp/new.txt name old time/op new time/op delta...

After some investigation I was able to make this change more or less performance-neutral, and even push size reduction a bit further. Here are some benchmarking results: ``` Env: Node.js...

Overall it's worth noting that the like with every synthetic benchmark, the real-world difference is likely to be much, much smaller. `GoroutineSwitching` benchmark was designed to cause goroutines block on...

As a curious side note, gopherjs on Node 12 performs in this benchmark only two times slower than native go: ``` name old time/op new time/op delta GoroutineSwitching 157ns ±...

Thanks @paralin, any additional testing is always welcome, although I'm fairly confident in this PR and was going to merge it as soon as I address a couple of minor...

I can confirm this can still be reproduces in GopherJS 1.17.

To me this behavior seems fairly reasonable. So I guess we only need to update documentation to declare this behavior as intended?

I wonder how relevant this still is... It seems like in the modern day most information is passed through REST (or similar) APIs, which would be unmarshaled by gopherjs using...