Hello,
I ran svn update this morning on dealii directory.
I get a nasty error on compile (gcc-4.1.2, gentoo linux amd64 with
PETSc and metis).
=====base=============debug========== utilities.cc
source/utilities.cc: In function 'unsigned int
dealii::Utilities::System::get_n_mpi_processes(const MPI_Comm&)':
source/utilities.cc:437: error: 'MPI_Comm_size' was not declared in
this scope source/utilities.cc: In function 'unsigned int
dealii::Utilities::System::get_this_mpi_process(const MPI_Comm&)':
source/utilities.cc:446: error: 'MPI_Comm_rank' was not declared in
this scope make[1]: ***
[/home/tyoung/fem/deal.II/lib/base/utilities.g.o] Error 1 make[1]:
Leaving directory `/home/tyoung/fem/deal.II/base' make: *** [baseg]
Error 2
I've traced this back to the function(s):
unsigned int get_n_mpi_processes (
const MPI_Comm &mpi_communicator)
{
int n_jobs=1;
(void) MPI_Comm_size (mpi_communicator, &n_jobs);
return n_jobs;
}
[...]
strangely (perhaps) if I force the compiler to use the alternative:
unsigned int get_n_mpi_processes (const MPI_Comm &)
{
return 1;
}
All is ok and the examples all run (at least for a single processor).
Is this an issue with my gcc version? The notes in utilities.cc seem to
suggest there is some weirdness in interactions between PETSc and some
gcc compilers... Any suggestions on a proper fix for this?
Best,
Toby
--
Toby D. Young - Adiunkt (Assistant Professor)
Department of Computational Science
Institute of Fundamental Technological Research
Polish Academy of Sciences
Room 206, ul. Swietokrzyska 21
00-049 Warszawa, Polska
+48 22 826 12 81 ext. 184
http://rav.ippt.gov.pl/~tyoung
_______________________________________________
