Andreas Noack

Looked a bit into this and it appears that it is actually a bug in R that was fixed relatively recently, see https://bugs.r-project.org/show_bug.cgi?id=18711. So we should upgrade `Rmath-julia` and `Rmath_jll`....

Once https://github.com/JuliaPackaging/Yggdrasil/pull/9413 has been merged, this issue should be resolved. It would be good to add a test case, though. @Martin-exp-z2 would you be able to prepare a PR with...

I can confirm that this now works ```julia (StatsFuns) pkg> up Updating registry at `~/.julia/registries/General.toml` Installed Rmath ───── v0.8.0 Installed Rmath_jll ─ v0.5.1+0 Downloaded artifact: Rmath Updating `~/.julia/dev/StatsFuns/Project.toml` [79098fc4] ↑...

> In future, it might be a good idea to write all the computation in Julia itself We are working towards that, see https://github.com/JuliaStats/StatsFuns.jl/issues/37, but indeed it takes time to...

I just realized that a new version of StatsFuns is needed here because of the minor version bump of Rmath. Once https://github.com/JuliaRegistries/General/pull/115053 is available, you should be able to upgrade.

It would be great to get this one in. From the Actions warnings, it looks like the current action versions are outdated.

I think this is good. Are there other breaking changes that we can come up with before releasing this?

In some of the algorithms for random number generation, for efficiency reasons, some numbers should only to be calculated once given the parameter values. In C you could use static...

The Rosenbrock tests are failing on master, see https://github.com/SciML/OrdinaryDiffEq.jl/actions/runs/20029704426/job/57435465939, so it isn't easy to assess the impact of this PR. I ran the allocation tests for Rosenbrock locally and they...

I believe the definition in ForwardDiff should just be an optimization, and that going through the one in GenericLinearAlgebra should still work. The problem is with the NaNs, though. Because...