Feedback from 11/7/18 virtual form that needs to be addressed by 11/19/18
[x] need to incorporate MPI_Comm_create_from_group from ticket mpi-forum/mpi-standard#44 into mpi-forum/mpi-standard#55
[x] need to incorporate MPI_File_open_from_group from ticket mpi-forum/mpi-standard#44 into mpi-forum/mpi-standard#55
[x] need to incorporate MPI_Win_create_from_group from ticket mpi-forum/mpi-standard#46 into mpi-forum/mpi-standard#55
[x] add a simplified example for initializing using mpi://world (Rolf wants to see concurrency in at least one example, multiple library, etc.)
[x] modify the current more complex example to use a user-specified pset name (Rolf wants to see concurrency in at least one example. Concurrency here could be both with multiple threads in each process creating sessions/comms, as well as the multiple libraries example)
[x] add wording about sessions isolation concept and limitations of mixing objects derived from different sessions in, for example MPI_Waitall etc.
[x] lines 36-37 on page 425 Seesion vs Session
[x] line 20 on page 390 - is turned should be returned
[x] rework the NULL terminated language for MPI_Session_get_nth_pset. This doesn't comport with Fortran. Use wording like that used for MPI_info_get
Feedback from 11/7/18 virtual form that needs to be addressed by 11/19/18
MPI_Comm_create_from_group
from ticket mpi-forum/mpi-standard#44 into mpi-forum/mpi-standard#55MPI_File_open_from_group
from ticket mpi-forum/mpi-standard#44 into mpi-forum/mpi-standard#55MPI_Win_create_from_group
from ticket mpi-forum/mpi-standard#46 into mpi-forum/mpi-standard#55mpi://world
(Rolf wants to see concurrency in at least one example, multiple library, etc.)MPI_Waitall
etc.MPI_Session_get_nth_pset
. This doesn't comport with Fortran. Use wording like that used forMPI_info_get