[openasocket]

Yeah, skimming your source it seems most of your loops involve calling some function, and even if that's inlined I believe the Go compiler will put a speculative yield call in there.

I suppose my hypothetical would be an issue if you used a non-Go BLAS implementation, as calling out to C will hog the OS thread. But this is a known issue (e.x. https://www.cockroachlabs.com/blog/the-cost-and-complexity-o...).

[chewxy]

The solution to that is to write a C-batcher. Gorgonia uses that (optional) - https://github.com/chewxy/gorgonia/tree/master/blase

(also it's undergoing major reconstruction/refactoring right now)