luchr
luchr
If `odesolver≠nothing` then `odesolver` (the IVP-solver) must be a (julia) function with input: `(rhs::Function, t0::Real, T::Real,x0::Vector, opt::AbstractOptionsODE)` and output `(t,x,retcode,stats)`. That's the form of all the ODE-Solvers in `ODEInterface`. Important:...
> So it's essentially the same as the options that you pass in, but with these three fields changed? The last two make sense, but why is the relative tolerance...
BVPSOL is a boundary value problem solver working with the multiple-shooting principle. Here is a picture (the black dashed line is the initial guess given by the user). The shooting...
The fortran solver BVP_M2 can only handle the singularities described in [Shampine, Muir, Xu: A User-Friendly Fortran BVP Solver](http://cs.stmarys.ca/~muir/JNAIAM_Shampine_Muir_Xu2006.pdf). Sorry.
For RETARD the answer is simply: No. RETARD needs at _compile time_ common blocks with data-structures that depend on the number of "back steps". So the number of (maximal) supported...
Can you attach the output of the build-log for ODEInterface. To get the log-file, try to (re-)build the interface with: ```julia using Pkg Pkg.build("ODEInterface") ``` Then there is message showing...
Thank you for your feedback. We have two separate problems here: First problem: Compiling `bvpsol` (the solver written by Deuflhard, Bader, Weimann) with GCC Version 10. Here setting `std=legacy` helps....
In the automatic Windows tests with [AppVeyor](https://ci.appveyor.com/project/luchr/odeinterface-jl/branch/master) I (still) use this `x86_64-7.3.0-posix-seh-rt_v5-rev0\mingw64` version, which works. I'll have to investigate further.
It's difficult to find the problem. `ldd` and `DependencyWalker` all show no problem. I even tried to load the dll-module the way Julia as doing it: https://github.com/JuliaLang/julia/blob/release-1.5/src/dlload.c#L94 ```C #include #include...
I still have no success in narrowing down the problem. I've found similar problems (in the past), e.g. https://sourceforge.net/p/mingw/mailman/message/19787051/ where an error in a def-file of mingw caused the same...