sundials-openmpi-5.2.0-bp156.3.4

- Update to version 5.2.0 (See
  https://computing.llnl.gov/projects/sundials/release-history for
  details):
  - Fixed a bug in how ARKode interfaces with a user-supplied,
    iterative, unscaled linear solver.
  - Fixed a bug in how ARKode interfaces with a user-supplied,
    iterative, unscaled linear solver.
  - Fixed a similar bug in how ARKode interfaces with scaled
    linear solvers when solving problems with non-identity mass
    matrices.
  - Fixed a memory leak in CVODES and IDAS from not deallocating
    the atolSmin0 and atolQSmin0 arrays.
  - Fixed a bug where a non-default value for the maximum allowed
    growth factor after the first step would be ignored.
  - Functions were added to each of the time integration packages
    to enable or disable the scaling applied to linear system
    solutions with matrix-based linear solvers to account for
    lagged matrix information.
  - Added two new functions, ARKStepSetMinReduction() and
    ERKStepSetMinReduction() to change the minimum allowed step
    size reduction factor after an error test failure.
  - Added a new SUNMatrix implementation, SUNMATRIX_CUSPARSE, that
    interfaces to the sparse matrix implementation from the NVIDIA
    cuSPARSE library.
  - Added a new "stiff" interpolation module to ARKode, based on
    Lagrange polynomial interpolation, that is accessible to each
    of the ARKStep, ERKStep and MRIStep time-stepping modules.

- Remove duplicated definitions.
- Remove bogus undefines of suffix and mpi_flavor for "serial"
  flavour. The former causes builds to fail for openSUSE >= 1550
  using rpm >= 4.15.

- Run spec-cleaner for minor cleanups.

- Update to version 5.0.0:
  * Two new NVector implementations created to support flexible
    partitioning of solution data among different processing
    elements (e.g., CPU + GPU) or for multi-physics problems that
    couple distinct MPI-based simulations together:
    NVECTOR_MANYVECTOR, amd NVECTOR_MPIMANYVECTOR,
  * An additional NVector implementation, NVECTOR_MPIPLUSX, has
    been created to support the MPI+X paradigm where X is a type
    of on-node parallelism (e.g., OpenMP, CUDA),
  * One new required NVector operation, N_VGetLength, and ten new
    optional vector operations have been added to the NVector API,
  * Two new SUNLinearSolver implementations,
    SUNLINEARSOLVER_SUPERLUDIST which interfaces with the
    SuperLU_DIST distributed, sparse, linear solver library, and
    the SUNLINEARSOLVER_CUSOLVERSP_BATCHQR, which interfaces to
    the cuSOLVE sparse batched QR linear solver,
  * A new SUNNonlinearSolver implementation,
    SUNNONLINSOL_PETSCSNES, which provides an interface to the
    PETSc SNES API,
  * New Fortran 2003 interface modules that provide Fortran users
    access to most of the SUNDIALS C API including ARKode,
    CVODE(S), IDA(S), and KINSOL,
  * Support for using explicit, implicit, or IMEX methods as the
    fast integrator with the MRIStep time-stepper in ARKode,
  * Several other minor changes and bug fixes: see
    https://computing.llnl.gov/projects/sundials/release-history.
- Merge all nvec solver libraries into a single shared lib
  package: %{shlib_nvec}.

- Enable multibuild with serial, openmpi1, openmpi2, openmpi3,
  and mvapich2 flavours.

- Follow shared library packaging policy and split out multiple
  versioned shlib packages. The main shared lib %{shlib_main}
  contains the common shared objects, while each individual solver
  gets its own shared lib package.
- Add blas-devel and lapack-devel BuildRequires; enable blas and
  lapack (does not work with 64 bits) during cmake.
- Enable pthread.

- Update to version 4.1.0:
  * An additional N_Vector implementation was added for Tpetra
    vector from Trilinos library to facilitate interoperability
    between SUNDIALS and Trilinos. This implementation is
    accompanied by additions to user documentation and SUNDIALS
    examples.
  * A bug was fixed where a nonlinear solver object could be freed
    twice in some use cases.
  * The EXAMPLES_ENABLE_RAJA CMake option has been removed. The
    option EXAMPLES_ENABLE_CUDA enables all examples that use CUDA
    including the RAJA examples with a CUDA back end (if the RAJA
    NVECTOR is enabled).
  * The implementation header files (e.g. arkode_impl.h) are no
    longer installed. This means users who are directly
    manipulating package memory structures will need to update
    their code to use the package’s public API.
  * Python is no longer required to run make test and make
    test_install.
  * Fixed a bug in ARKodeButcherTable_Write when printing a
    Butcher table without an embedding.
- Changes between previously packaged version (2.5.0) through
  version 4.0.2:
  https://computation.llnl.gov/projects/sundials/release-history.
- Switch to cmake based build in keeping with upstream.
- Drop devel-static package since application no longer builds
  static libraries anyway.
- Only build one (serial) version for now.
- Update Source and URL tags.
- Remove NOTICE and LICENSE files from includedir; package them
  properly as doc.

- fix mistyping in spec file which broke build where mvapich2
  is not available

- set --with-mpi-libs in configure in order to fix mpi library
  linking

- enable parallel build for openmpi and mvapich2

- remove not applied patch0 from files list

- spec files re-formating
- version 2.5.0
  * Many bugfixes and new features
  * See https://computation.llnl.gov/casc/sundials/download/
  whatsnew.html for a complete list of changes