[Debichem-devel] Bug#924032: mpqc3: FTBFS: Could NOT find MPI

[Drew Parsons]

On 2019-03-18 18:27, Andreas Tille wrote:
> Control: tags -1 help
> 
> Hi,
> 
> I stumbled upon bug #924032 which has no response so far.  I tagged
> it help and repost here on Debian Science list since there is usually
> some competence to solve issues like
> 
> -- Detecting Fortran/C Interface
> -- Detecting Fortran/C Interface - Found GLOBAL and MODULE mangling
> -- Verifying Fortran/CXX Compiler Compatibility
> -- Verifying Fortran/CXX Compiler Compatibility - Success
> -- Found PAPI: /usr/lib/x86_64-linux-gnu/libpapi.a
> -- Found MPI_C: /usr/lib/x86_64-linux-gnu/libmpich.so (found version 
> "3.1")
> -- Could NOT find MPI_CXX (missing: MPI_CXX_WORKS)
> -- Found MPI_Fortran: /usr/lib/x86_64-linux-gnu/libmpichfort.so (found
> version "3.1")
> -- Could NOT find MPI (missing: MPI_CXX_FOUND) (found version "3.1")
> CMake Error at external/MPI:36 (message):
>   MPI not found



The bug report does not report the arch, which is important information 
here (please use reportbug!)

The log snippet refers to mpich, but most arches use openmpi by default. 
  So that's a clue.

The full attached log says it's a simple x86_64.  Why then is mpich 
installed?

All the same, it should be possible to use mpich on amd64.

I think the final clue is "Could NOT find MPI_CXX".  It suggests mpicc 
is not being used, which makes these MPI builds simpler.
But the log says mpicc is used, -DMPI_C_COMPILER=mpicc

My guess is the C++ compiler needs to be set: -DMPI_CXX_COMPILER=mpic++  
  (or mpicxx)

Drew