PETScMatrix.array() getting transposed when copying values into a new petsc4py matrix

Dear all,

in the following MWE I am trying to assign a DOLFIN PETScMatrix to a custom petsc4py matrix.

import petsc4py
petsc4py.init()
from petsc4py import PETSc
from dolfin import *

mesh = UnitIntervalMesh(4)
V = FunctionSpace(mesh, "CG", 1)

u = TrialFunction(V)
v = TestFunction(V)

a = u*v.dx(0)*dx
A = assemble(a)
A = A.array()
print(A)

B = PETSc.Mat().createDense(A.shape)
B.setUp()
B[:, :] = A
B.assemble()
B.view()

When running with the former swig backend copy of A into B is done successfuly

[[ 0.5  0.5  0.   0.   0. ]
 [-0.5  0.   0.5  0.   0. ]
 [ 0.  -0.5  0.   0.5  0. ]
 [ 0.   0.  -0.5  0.   0.5]
 [ 0.   0.   0.  -0.5 -0.5]]
Mat Object: 1 MPI processes
  type: seqdense
5.0000000000000000e-01 5.0000000000000000e-01 0.0000000000000000e+00 0.0000000000000000e+00 0.0000000000000000e+00 
-5.0000000000000000e-01 0.0000000000000000e+00 5.0000000000000000e-01 0.0000000000000000e+00 0.0000000000000000e+00 
0.0000000000000000e+00 -5.0000000000000000e-01 0.0000000000000000e+00 5.0000000000000000e-01 0.0000000000000000e+00 
0.0000000000000000e+00 0.0000000000000000e+00 -5.0000000000000000e-01 0.0000000000000000e+00 5.0000000000000000e-01 
0.0000000000000000e+00 0.0000000000000000e+00 0.0000000000000000e+00 -5.0000000000000000e-01 -5.0000000000000000e-01

When running with current master instead I get that B = A^T

[[ 0.5  0.5  0.   0.   0. ]
 [-0.5  0.   0.5  0.   0. ]
 [ 0.  -0.5  0.   0.5  0. ]
 [ 0.   0.  -0.5  0.   0.5]
 [ 0.   0.   0.  -0.5 -0.5]]
Mat Object: 1 MPI processes
  type: seqdense
5.0000000000000000e-01 -5.0000000000000000e-01 0.0000000000000000e+00 0.0000000000000000e+00 0.0000000000000000e+00 
5.0000000000000000e-01 0.0000000000000000e+00 -5.0000000000000000e-01 0.0000000000000000e+00 0.0000000000000000e+00 
0.0000000000000000e+00 5.0000000000000000e-01 0.0000000000000000e+00 -5.0000000000000000e-01 0.0000000000000000e+00 
0.0000000000000000e+00 0.0000000000000000e+00 5.0000000000000000e-01 0.0000000000000000e+00 -5.0000000000000000e-01 
0.0000000000000000e+00 0.0000000000000000e+00 0.0000000000000000e+00 5.0000000000000000e-01 -5.0000000000000000e-01

What's going on here? I used the same petsc and petsc4py in both cases.

Thanks,

Francesco

Comments (1)