[petsc-dev] MatTranspose_MPIAIJ broken

[Jed Brown]

"Mark F. Adams" <mark.adams at columbia.edu> writes:

> madams/petsc> grep PETSC_HAVE_MPI_COMBINER_DUP ./arch-xc30-opt/include/petscconf.h
> #ifndef PETSC_HAVE_MPI_COMBINER_DUP
> #define PETSC_HAVE_MPI_COMBINER_DUP 1

This should be fine.

> I can't reproduce this problem on my Mac with ex56.  This error was on
> Edison at NERSc.  The new Cray.
>
> I'll try rebuilding everything from scratch to double check.

Let me know if this works.  If not, I can help set up a reduced test
case.  It would be helpful to see how control flow goes through
MPIPetsc_Type_compare.