Parallel Reduce Overloads for DeviceVector

[ax3l]

Summary

Adds a new header AMReX_GpuParallelReduce.H which GPU-overloaded and support for MPIs with and without GPU-awareness (GPU-pointer support) in a safe manner.

Additional background

Added to prevent common usage bugs like
BLAST-ImpactX/impactx#1532

Checklist

The proposed changes:

• fix a bug or incorrect behavior in AMReX
• add new capabilities to AMReX
• changes answers in the test suite to more than roundoff level
• are likely to significantly affect the results of downstream AMReX users
• include documentation in the code and/or rst files, if appropriate

[ax3l]

/run-hpsf-gitlab-ci

[github-actions]

GitLab CI has started at https://gitlab.spack.io/amrex/amrex/-/pipelines/1664602.

[amrex-gitlab-ci-reporter]

GitLab CI 1664602 finished with status: success. See details at https://gitlab.spack.io/amrex/amrex/-/pipelines/1664602.

[ax3l]

Unrelated test failure in GNU@13 3D Debug Fortran [tests], will restart:

      Start 72: LinearSolvers_NodeTensorLap_3d
72/81 Test #72: LinearSolvers_NodeTensorLap_3d ..................***Failed   15.94 sec
Initializing AMReX (26.06)...
MPI initialized with 2 MPI processes
MPI initialized with thread support level 0
AMReX (26.06) initialized
MLMG: # of AMR levels: 1
      # of MG levels on the coarsest AMR level: 7
MLMG: Initial rhs               = 172.9154691
MLMG: Initial residual (resid0) = 44849.99183
MLMG: Iteration   1 Fine resid/resid0 = 0.02025997706
MLMG: Iteration   2 Fine resid/resid0 = 0.001572036055
MLMG: Iteration   3 Fine resid/resid0 = 0.0001244702566
MLMG: Iteration   4 Fine resid/resid0 = 1.009954306e-05
MLMG: Iteration   5 Fine resid/resid0 = 8.37759707e-07
MLMG: Iteration   6 Fine resid/resid0 = 7.087673307e-08
MLMG: Iteration   7 Fine resid/resid0 = 6.099877431e-09
MLMG: Iteration   8 Fine resid/resid0 = 5.326637399e-10
MLMG: Iteration   9 Fine resid/resid0 = 4.708424764e-11
MLMG: Iteration  10 Fine resid/resid0 = 4.204271835e-12
MLMG: Final Iter. 10 resid, resid/resid0 = 1.885615575e-07, 4.204271835e-12
MLMG: Timers: Solve = 8.163485814 Iter = 7.99826611 Bottom = 0.000278109
Level 0
    max-norm: 0.000282219024
    1-norm  : 4.939185926e-05
amrex::Error::1::VisMF::Write failed !!!
SIGABRT
See Backtrace.1 file for details
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 1 in communicator MPI_COMM_WORLD
with errorcode 6.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------

[ax3l]

/run-hpsf-gitlab-ci

[github-actions]

GitLab CI has started at https://gitlab.spack.io/amrex/amrex/-/pipelines/1665058.

[amrex-gitlab-ci-reporter]

GitLab CI 1665058 finished with status: success. See details at https://gitlab.spack.io/amrex/amrex/-/pipelines/1665058.