BUG: checkpointing produces error

[colinjcotter]

Describe the bug

I get an error message when I try to save functions to a checkpoint file.

Steps to Reproduce

from firedrake import *
mesh = IcosahedralSphereMesh(radius=1.0,
                                refinement_level=4,
                                degree=1, name="mesh")
V = FunctionSpace(mesh, "CG", 1)
u = Function(V, name="u").assign(1.0)

with CheckpointFile('test.h5', 'w') as afile:
    afile.save_function(u)

Expected behavior

Error message

(firedrake) journal todo +cjc1@cjc1-HP-Z840-Workstation:~/repositories/sw_timestepping$ python test_chk.py 
Traceback (most recent call last):
  File "/home/cjc1/repositories/sw_timestepping/test_chk.py", line 9, in <module>
    afile.save_mesh(mesh)
  File "petsc4py/PETSc/Log.pyx", line 188, in petsc4py.PETSc.Log.EventDecorator.decorator.wrapped_func
  File "petsc4py/PETSc/Log.pyx", line 189, in petsc4py.PETSc.Log.EventDecorator.decorator.wrapped_func
  File "/home/cjc1/firedrake/src/firedrake/firedrake/checkpointing.py", line 638, in save_mesh
    self._save_function_topology(mesh._coordinates)
  File "petsc4py/PETSc/Log.pyx", line 188, in petsc4py.PETSc.Log.EventDecorator.decorator.wrapped_func
  File "petsc4py/PETSc/Log.pyx", line 189, in petsc4py.PETSc.Log.EventDecorator.decorator.wrapped_func
  File "/home/cjc1/firedrake/src/firedrake/firedrake/checkpointing.py", line 971, in _save_function_topology
    self._save_function_space_topology(tV)
  File "petsc4py/PETSc/Log.pyx", line 188, in petsc4py.PETSc.Log.EventDecorator.decorator.wrapped_func
  File "petsc4py/PETSc/Log.pyx", line 189, in petsc4py.PETSc.Log.EventDecorator.decorator.wrapped_func
  File "/home/cjc1/firedrake/src/firedrake/firedrake/checkpointing.py", line 866, in _save_function_space_topology
    self._save_mesh_topology(tmesh)
  File "petsc4py/PETSc/Log.pyx", line 188, in petsc4py.PETSc.Log.EventDecorator.decorator.wrapped_func
  File "petsc4py/PETSc/Log.pyx", line 189, in petsc4py.PETSc.Log.EventDecorator.decorator.wrapped_func
  File "/home/cjc1/firedrake/src/firedrake/firedrake/checkpointing.py", line 703, in _save_mesh_topology
    raise ValueError(f"Mesh ({tmesh_name}) already exists in {self.filename}, but the global number of DMPlex points is inconsistent: {order_array_size1} ({self.filename}) != {order_array_size} ({tmesh_name})")
ValueError: Mesh (firedrake_default_topology) already exists in test.h5, but the global number of DMPlex points is inconsistent: 9 (test.h5) != 15362 (firedrake_default_topology)

Environment:

• Linux 24.04

• Python 3.12.3 Firedrake Configuration: package_manager: True minimal_petsc: False mpicc: None mpicxx: None mpif90: None mpiexec: None disable_ssh: False honour_petsc_dir: False with_parmetis: False slepc: False packages: [] honour_pythonpath: False opencascade: False torch: False jax: False petsc_int_type: int32 cache_dir: /home/cjc1/firedrake/.cache complex: False remove_build_files: False with_blas: None netgen: False Additions: None Environment: PYTHONPATH: None PETSC_ARCH: None PETSC_DIR: None Status of components:

  |Package |Branch |Revision |Modified |

  |FInAT |master |ba0e9bb |False | |PyOP2 |master |31471a60 |False | |fiat |master |13131da |False | |firedrake |master |4e6cba30b |False | |h5py |firedrake |db2ab02f |False | |libsupermesh |master |b80408e |False | |loopy |main |d9876d83 |False | |petsc |firedrake |1c1d65e4821|False | |pyadjoint |master |b9574fb |False | |pytest-mpi |main |f4a9692 |False | |tsfc |master |0ec9234 |False | |ufl |master |c1a8afb1 |False |

Additional Info

[connorjward]

Are you using an up-to-date Firedrake branch? I think @ksagiyam fixed this a short while ago.

[ksagiyam]

https://github.com/firedrakeproject/firedrake/pull/3792/files

It seems you managed to update your Firedrake after our PETSc fork was updated, but before the matching Firedrake PR went in.