That link admits that the compiler performance was lower than needed in Go1.18 and Go1.19, because of generics, even when compiling Go code without generics. I can confirm this based on my own open source projects written in Go [1].
But the time spent on designing, implementing and then optimizing the generics is lost. This time could be spent on more valuable things for Go ecosystem.
Generics consistently showed up as one of the most desired features (if not the most desired) by working Go developers in the previous developer surveys, so I think it makes sense that the Go team felt the ecosystem saw much value in it relative to other features and worth the time.
Unfortunately, the Go team was misguided by vocal minority who was using the "missing generics" argument as an excuse why they do not switch to Go from their favorite programming languages. The majority of working Go developers were happy with Go, so they didn't take part in the debates and surveys about generics.
The irony is that vocal minority still do not use Go, since they have other excuses now - "bad error handling", "missing functional features", etc. But if these harmful features will be implemented in Go, the vocal minority will find another reason why they do not use Go.
[1] https://github.com/valyala/