Vladimír Fuka

Thanks for mentioning my little example (should have been updated a long time ago). I do agree that `select type` is a big drawback and I use tye parametric version...

It is certainly possible to have a linked list or an array of container types that contains a `class(*)` item. The user of that list then will be responsible to...

No, you do not have to remember what data is in *that node*. You just have to know what kind of data you expect to encounter in your list. That...

But, I repeat, you do *not* have to remember what kind of data is in which node of the list. You just need to know what data *might* be there...

I am afraid there is no `pow()` function in standard Fortran or in the GCC extensions. I do know how to call C functions from Fortran, but I fear f2py...

I am sorry, but the answer appears to be misunderstanding, not the original code. The original code provided by @coolplay is perfectly legal Fortran. The answer provided by @broukema only...

In this case such argument is just bizzare. But giving that SciPy tries to get rid of all Fortran very hard it perhaps stays in line. On Sat, 10 Feb...

Are there any news about this or some suggestions? I may have time to do some work in this in my new project.

I think the local transposition can still benefit even if memory bound as modern CPU's have more than one memory bus (and we often have two sockets per node). Anyway,...

So the non-threaded portions are in `transpose_chunks` and `transpose_toms513` (or whatever size is applicable). At least for the 3D r2r transform I tried. The time is spent mostly in calls...