Sebastian Pfitzner
Sebastian Pfitzner
Yeah, probably.
@mortenpi wanna fix the docs for this? :)
Yes, it is. Are you using the LS in single-file mode? If so, that'd explain the issue (since we're only searching in the workspace dir). I don't really remember my...
This happens because the LS doesn't realize that `LinearAlgebra` and `SparseArrays` overload the `kron` definition in Base (where only the obviously non-matching `kron(a, b, c, xs...)` is defined).
Apparently it can be, yes. I don't have a machine I can repro this on, but we got reports of a ~3x slowdown. > One gigantic file for all cached...
What Julia version are you on? I only get one method listed (which is the one you're missing, interestingly enough): 
Hm, still can't repro this after clearing my own cache and using the DSP.jl env.
Ah, that explains it then. Local caching and remote caching can lead to different results.
It is, yes. There's no difference between the code except that one doesn't run on your machine :) I suspect that this is an issue with merging the two method...
I feel like we had some recent improvements to cases like this, but should probably try the examples here again.