Skip to content

v7.3.0

Latest

Choose a tag to compare

@matnoel matnoel released this 13 Nov 11:54
· 56 commits to main since this release

MPI

  • Elastic, Newmark and Newton solved over the ranks match the serial solve. A solver
    other than petsc raises, as do PhaseField BoundConstrain (lsq_linear is serial)
    and the Solve of a gathered simulation; Plot_Energy needs a gathered one.
  • EASYFEA_MPI=off runs one serial simulation per rank, writing one results file per
    rank.
  • The Newton residual norm over the ranks is the root of the summed squares. Under PETSc
    no orphan diagonal is written: it reads only the owned rows.
  • Mesh element sums (mass, center) count each element once over the ranks; groups store
    isPartitioned and Mesh.center is the rank's.
  • PhaseField energies take their zero shortcut on a reduced value, so a disk Set_Iter
    no longer hangs.
  • print(simu) prints on rank 0 only and never communicates: partitioned results are
    rank 0's, under a warning.
  • use_mpi.md launches with python -m mpi4py, so an uncaught exception aborts every
    rank.

Visualization

  • Vizir._Write_element_solution_file: values constant over each element, written as a
    degree 0 HOSolAt. Every dim-2 group takes the mean of the main-dimension elements
    holding all its nodes, so a boundary face gets its element's value; main-dimension
    groups too on a 3D mesh.
  • Vizir opens a solution file through one context manager writing its header, warp
    vector and end.

API

  • Simu.Get_K_C_M_F returns the assembled matrices, not copies; the time-scheme A is
    summed on the values when C and M share K's pattern.
  • Assembling a complex system raises: b and x are real ndarrays; the CSR column and
    asCsrMatrix are gone.
  • BiLinearForm rows are test functions, columns trial functions.

Development

  • CI: the MPI job installs BLAS and LAPACK and builds petsc and petsc4py with uv,
    cached, so the solves run.
  • ruff format replaces black.

Full Changelog: v7.2.0...v7.3.0