Hacker News Reader

Comment by shereadsthenews | Hacker News Reader

The focus on Go compiler speed is kinda weird to me. Other languages sometimes offer flexible tradeoffs to developers. You can build and test C++ programs fairly quickly if all the optimizations are disabled, or you can spend a long time building a release binary with optimization, load-testing, profile-guided recompilation, link-time optimization, and post-link optimization. But Go only offers the first thing: a fast compiler with almost no optimizations.

View on HN

C++ with all optimizations disabled is still slow. g++ takes 2.8 seconds to compile a simple helloworld on my machine - or .14 seconds the second time after caches are warm... Python2 can build and run helloworld in .028 seconds, or .011 once caches are warm. Of course the C++ hello world runs in .002 seconds once build (caches don't seem to make a difference but that might be because it was still in the cache from building)

What version of gcc? What OS? I am showing ~0.4 seconds on a machine that is 3-4 years old. [PS: I think even 0.4 seconds is too slow]

4.8.4 Note that only the cached numbers are repeatable. I'm running in a VM, which is likely to be a factor. (IT... the machine exists to run the one VM)

I empathize very strongly with you. At a past gig I was in an almost similar position.

C++'s templates (code!) in headers can cause a lot of pain on the compilation front. A simple #include <boost> can add many seconds to your compile.

Recently at work I was able to use C++ templates to automate a lot of repeated boilerplate code I was responsible for. The reason was to minimize the amount of (error-prone!) rote work that had to be done as the project would continue to scale.

It worked out great, but I’ve noticed that the compilation time of translation units that use the templates have been growing linearly with features added. It’s not a lot in the big scheme of things, but it made me wish I could reason more about C++ compile times when I use its various features.

You can make use of libraries and external templates introduced in C++11.

Well that's easy. Python doesn't compile anything. The old adage, 'the secret to writing fast code is writing code that does less' is true.

Sure it does — its just not compiling ahead of time; or rather, compiling + running is behind a single command

And the compilation is depressingly literal because optimization is so difficult with Pythons level of dynamics.

Python2 is an interpreter. If you want to do a proper comparisasion you need to compare with a C++ interpreter.

Or get a Python2 compiler to native code.

My comparison is valid. Engineering is about tradeoffs. When considering python vs C++ cost of compiling is on factor, and cost of the run time is another. There are many other factors that you need to consider when deciding what language to use. The ability to compile python to native code is one as well.

It is not, because it is not the same execution model that is being compared, it is apples vs pears.

The Go compiler contains lots of optimizations, especially considering its age.

Can you share some more details about what those optimizations are? Can you point us to a link?

https://quasilyte.dev/blog/post/go_ssa_rules/

is maybe a good starting point.

You could try compiling with and without `-gcflags '-N -l'`

`-N` disables optimization and `-l` disables in-lining.

I also found this[0] page which lists some optimizations done by the compiler.

[0] https://github.com/golang/go/wiki/CompilerOptimizations

edit: You can see what the compiler emits as assembly code with `-gcflags '-S'` so you can compare the optimized vs unoptimized assembly.

go is indeed slow, cd benchmarkgames website

Reply on news.ycombinator.com