Mark Dewing

@ye-luo raised the possibility of removing OOMPI w/o necessarily replacing the use of all the templates in CommOperatorsMPI.h. It looks like OOMPI functions and data structures are only explicitly used...

OOMPI has been removed (#1540 , #1530). What remains - convert rest of real-space code to use new wrapper instead of that in Message/CommOperatorsMPI.h - use fake MPI for non-MPI...

Is there any chance of breaking this into separate PR's that are more reviewable? (That is, are the changes to individual classes separable, or they all linked?) Is there a...

The changes in the Numerics and Particle directories clearly can't depend on SPOSet. And a lot of the changes in QMCDrivers are just minor changes like include files.

It looks like the "new" files have been run through clang-format that is more restrictive than the default one, or been modified in an editor that does some formatting. What...

I think the choice is made here, based on if the twists are special points: https://github.com/QMCPACK/qmcpack/blob/e7fd649ae2fcd569feced8cf853f45d8dc003827/src/QMCWaveFunctions/EinsplineSetBuilderCommon.cpp#L632-L642

The subtle part is the values themselves don't matter, and past the first rotation, the actual values are meaningless. The difference in values between calls to apply the rotation is...

See #4145 (and linked issues and fixes) for previous encounters with cusp correction and atom reordering.

The context is testing for #4901. The cusp correction depends on a discrete parameter (locations where the second derivative of the correction matches the second derivative the basis function). I...

For the overlap matrix, the first row and column are the identity matrix, so the inverse also ends up as the identity matrix in the first row and column, and...