Re: [petsc-users] question about arbitrary eigenvector selection in SLEPC

[Giacomo Mulas]

On Thu, 29 May 2014, Matthew Knepley wrote:

There might be an easier way to do this:
  PetscScalar val = 0.0, gval;
  VecGetOwnershipRange(xr, &low, &high);
  if ((myindex >= low) && (myindex < high)) {
    VecGetArray(localx1,&a);
    val = a[myindex-low];
    VecRestoreArray(localx1, &a);
  }
  MPI_Allreduce(&val, &gval, 1, MPIU_SCALAR, MPI_SUM, PETSC_COMM_WORLD);

Now everyone has the value at myindex.

brilliant, why didn't I think of this? Only, I guess you were
copying/pasting and some variable names slipped, namely localx instead of
xr.  Should it be

  PetscScalar val = 0.0, gval;
  PetscScalar *a;

  VecGetOwnershipRange(xr, &low, &high);
  if ((myindex >= low) && (myindex < high)) {
    VecGetArray(xr,&a);
    val = a[myindex-low]; 
    VecRestoreArray(xr, &a);
  }
  MPI_Allreduce(&val, &gval, 1, MPIU_SCALAR, MPI_SUM, PETSC_COMM_WORLD);
  *rr = gval*gval;
  *ri = 0;

?

Thanks!
Giacomo

--
_________________________________________________________________

Giacomo Mulas <gmu...@oa-cagliari.inaf.it>
_________________________________________________________________

INAF - Osservatorio Astronomico di Cagliari
via della scienza 5 - 09047 Selargius (CA)

tel.   +39 070 71180244
mob. : +39 329  6603810
_________________________________________________________________

"When the storms are raging around you, stay right where you are"
                         (Freddy Mercury)
_________________________________________________________________