Fabian Gittins

The type-instability for your ODE is resolved if the function is declared `isinplace` or not (see the [documentation](https://docs.sciml.ai/DiffEqDocs/stable/basics/problem/#In-place-vs-Out-of-Place-Function-Definition-Forms)): ```julia function main() du0 = [1.0] u0 = [0.0] tspan = (0.0,20.0)...

I have found a way to remove the type-instability. If I change the initial condition passed to `prob2` to be `2.0` rather than `2`, ```julia prob2 = ODEProblem{false}(perturbation, 2.0, rspan,...

The type-instabilities disappear when you: 1. Tell the compiler whether the function `isinplace` or not. So write ```julia prob = ODEProblem{true}(lorenz!, u0, tspan) ``` Then ```julia julia> @code_warntype ODEProblem{true}(lorenz!, u0,...

Thanks for the feedback. I had some time so I put together an [initial implementation](https://github.com/fgittins/SimpleNonlinearSolve.jl) at this algorithm with some simple tests. If it looks reasonable, I can make a...

I have checked this with DiffEqBase v6.151.4 and there is still an error. Maybe this is fixed in an upcoming release? I'll include the error message and my environment below....

Yes. It only works for univariate, scalar functions.