Kristoffer Carlsson
Kristoffer Carlsson
Yeah, makes sense
> `About to run: goto %4 if not _J2` I don't think it is good to stop at these internal looking statements.
No, JuliaInterpreter ends up asking the Julia compiler to run the ccall: https://github.com/JuliaDebug/JuliaInterpreter.jl/blob/19757b45fa438cf8335601099a21525a88c9bb39/src/optimize.jl#L386.
> I find 40 calls to iolock_begin in base/ and 48 in stdlib/, which means this would be a lot of methods to add to compiled_methods. Any ideas about alternatives?...
On 0.7. ``` julia> @time FastGaussQuadrature.JacobiAsy(101,1., 1.) 28.552633 seconds (61.87 M allocations: 2.746 GiB, 4.82% gc time) ``` Going deeper ``` julia> @time FastGaussQuadrature.asy1(101, 1.0, 1.0, 10) 26.084398 seconds (55.88...
Yes, compile time; just posting an update. Seems the code here is written in a way that interacts badly with the compiler but yeah, indeed an issue with the compiler.
I think the large dotted expressions is a problem for the compiler. A workaround might be to not use the `@.` macro so much and use normal operators between scalars...
I was doing some benchmarking for audit and based on that I think that a pass should take *all* the libraries. The reason for this is to avoid overhead with...
> `-funsafe-math-optimizations` means there are math optimizations that could be unsafe. We don't like unsafe optimizations in the Julia world; we'd rather be correct and a little slower, since the...
Updated to reflect that the problem is the `dlopen` of `libgomp` in CompilerSupportLibraries_jll