Maybe we should have a separate section at the bottom of the changes file for internal changes to PETSc that do not affect the user API.
Barry > On Sep 19, 2020, at 2:40 PM, Satish Balay via petsc-dev > <petsc-dev@mcs.anl.gov> wrote: > > https://gitlab.com/petsc/petsc/commit/b7b8f77a1c31acd4cca9a56e02609d1dffe2a4b3 > > > include/petsc/private/f90impl.h is not really user facing - so perhaps this > shouldn't have been mentioned in the changes file. > > Satish > > On Sat, 19 Sep 2020, Victor Eijkhout wrote: > >> "The F90Array routines now take a MPI_Datatype argument instead of a >> PetscDataType. This should not affect user code.” >> >> What does that refer to? >> >> https://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Mat/MatDenseGetArrayF90.html >> >> has no MPI_Datatype arguments. >> >> V. >>
