Heinz N. Gies
Heinz N. Gies
Ja, we probably want something resembling the upstream solution that does only once include a conditional and after that memorizes the choice.
Yes very good idea 👍 that way applications who want to make sure there is 0 overhead on the hot path can call it on startup :), good thinking!
That generally looks cool, that said I'm a bit concerned about the change in naming for functions. It will make SIMD code a lot harder to share or keep in...
I'm curious about that, what kind of CPU are you trying to compile it for. Does `i686` even support sse4.2?
Oh, never thought about that :D it's a neat idea. That sadly makes some sense on some of the commands missing :( we have the same issue with aarch64, as...
This isn't quite possible simd-json does in situ parsing, so the slice needs to be mutable. The possible solution I see here would be to add support for NDJSON to...
Just sketching something here. An API that would be really nice would be something like (non valid rust syntax but pseude code! ``` fn iter_lines(r: Read) -> Iter; for items...
We've done quite a bit on this, I'd love to get some feedback of what's missing from users and not just us.
Heya, Sorry for the late reply. I'm terrible with GitHub notifications :sweat_smile: `target-cpu=native` means that rust figures out the most advanced instruction set that the CPU supports. This is nice...
To my knowledge, that should include all instruction sets for Zen 2 the compiler understands, but I've never cross compiled between architectures. But yes AFIAK that's what you describe/aim for...