something is broken though.

preliminary. in particular solvers cannot be a simple function, they
should handle checkpointing --- especially if Adams-Bashforth particles
are to be reimplemented with the new classes.

I think it's more obvious when we talk about "index of component" and
"index of variable" what is meant, rather than having "FIELD_IDX" and
plain "IDX".

Make it compile, but I would have prefer a class attribute instead of method to decompose template of field

Berenger Bramas authored 8 years ago and Cristian Lalescu committed 8 years ago

Add a sampling method -- update to be generic against the field -- need to update NSVEparticles::do_stats

Use OpenMP in the new particles system -- parallelize adams_bashforth with a classic for - parallelize computation with a task based scheme

Make it works when the number of processes is not a factor of the size of the problem -- FFTW will give a partition of size 0 to several processes at the end in such case

Remove warnings -- split the accuracy (float/double) between field and particles, use double by default for the particles

Call the spline using the position of the particle inside the cell (normalize to 0-1) -- pass an array of box width to the constructor

Add an interface builder (with runtime values for spline mode and interpolation size) -- templatize the real number types (splines still rely on double)

Update all -- dims idx are now z-y-x -- field wrapper use a memory leading array for data access -- debug rhs particles extension when re distribute the particles