Ronald Rahaman

Then isn't that the fault of the MPI implementation? We get MPI_TAG_UB from mpi_get_attr. If MPI_TAG_UB doesn't accommodate the maximum ranks on the machine, then it's not the right implementation...

In the Nek drivers, I see some cases where `gsl::index` might not be the best choice. In some loops, we're ultimately using the loop counter to index a Fortran array....

Here are some suggestions based on the Core Guidelines and a look at what's happening in ENRICO: # Flagging Signedness Mismatches of Container Indices The Core Guidelines recommend using signed...

@paulromano and I had some Slack discussions, and I wanted to summarize here. What we settled on was: 1. Always declare indices as `gsl::index` 2. Use `gsl::narrow` when passing indices...

Greetings! No, you have to use one of the two already-implemented neutronics solvers: OpenMC (which distributed with ENRICO) or SHIFT (which is not). If you'd like to couple to another...

Hello! ENRICO can't, in general, be run with an arbitrary version of OpenMC. This is because the OpenMC API is under active development. We update ENRICO's OpenMC submodule from upstream...

It’s not feasible to statically link everything into the ENRICO binaries because one of its third-party single-physics solvers is configured to use dynamic linking. I recommend setting CMAKE_INSTALL_PREFIX to something...

Not all our 3rd-party libraries can be configured to be built with fPIC. The rpath in the installed ENRICO binaries is set to the install path of its libraries, including...

One of our dependent CFD libraries -- nekRS and its backend OCCA -- have many header in the ENRICO installation. A nekRS case has a number of case-specific source files...

Are you able to successfully install via something like this? * I'm inferring the MPI compiler paths from the CMake command you quoted above. * I excluded these all the...