then it doesn't set the toolchain or MPI using the same cmake scripts that were used for building. For the toolchain it does a getToolchain command that will just see what variables have been set. It will thus just default to whatever compiler cmake thinks is the default for that system rather than what toolchain is specified.
In addition to this the architecture path is set before the toolchain/mpi is even determined so how can it select the correct architecture path?
If I have an example in an environment with multiple configurations I can't select the one that want in the standard way. For an example if I
cmake -DOPENCMISS_TOOLCHAIN=xyz -DOPENCMISS_MPI=abc .
then it doesn't set the toolchain or MPI using the same cmake scripts that were used for building. For the toolchain it does a getToolchain command that will just see what variables have been set. It will thus just default to whatever compiler cmake thinks is the default for that system rather than what toolchain is specified.
In addition to this the architecture path is set before the toolchain/mpi is even determined so how can it select the correct architecture path?