[go-nuts] Re: [generics] I think it's better have higher perfomance than lower compiling time

2020-06-27 Thread Viktor Kojouharov
I seem to recall that the compilation time of C (and C++) programs is directly proportional to the number of includes your program has. I'm sure that monomorphisation in C++ exacts some kind of cost to the compilation time, but perhaps it's not as much as people seem to think. On Friday, June

Re: [go-nuts] Re: [generics] I think it's better have higher perfomance than lower compiling time

2020-06-22 Thread Jesper Louis Andersen
On Fri, Jun 19, 2020 at 8:13 PM wrote: > We have seen performance decreases in some language versions, but we have > also seen performance increases. Efficient execution is a project goal, but > efficient compilation is as well, so I would not recommend hoping for or > expecting that performance

Re: [go-nuts] Re: [generics] I think it's better have higher perfomance than lower compiling time

2020-06-21 Thread David Skinner
Tyler Compton is correct. In the early 90s, the plethora of C++ optimization switches resulted in obscure memory errors which resulted in major software houses shipping bad products, declaring bankruptcy, and disappearing from the face of the earth like dinosaurs. I am totally opposed to optimizati

Re: [go-nuts] Re: [generics] I think it's better have higher perfomance than lower compiling time

2020-06-21 Thread 'Axel Wagner' via golang-nuts
Why not just build your release-build with gccgo? It often already gives better performance than gc, AIUI. If enough of the people who really care about it would focus their energy on it, that could probably get even more pronounced. There's no reason that gc needs to grow optimization knobs, if yo

Re: [go-nuts] Re: [generics] I think it's better have higher perfomance than lower compiling time

2020-06-20 Thread Tyler Compton
The idea of having development and release compilation modes sounds appealing at first, though I'm a bit skeptical about how it would turn out. This would mean that, during development, you wouldn't have a full understanding of the performance characteristics of your program. Admittedly, that's not

[go-nuts] Re: [generics] I think it's better have higher perfomance than lower compiling time

2020-06-19 Thread David Skinner
I remember going on a two-hour lunch break at a Cajun restaurant and returning to my office to discover that my precompiled C++ headers were still compiling. I really love that Go compiles quickly and links quickly. The day comes when code needs to ship, final compile, and final quality control

[go-nuts] Re: [generics] I think it's better have higher perfomance than lower compiling time

2020-06-19 Thread howardcshaw
I think in general this is not the attitude of the Go developers, nor the Go community - fast compilation was an explicit design goal of the original creation of Go (as a reaction to slow compile times with C++), and has remained a concern. >From the FAQ, https://golang.org/doc/faq#creating_a_n