* Make particle IDs valid in BoundaryScraping

* Add CI test

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Fix automated test

* Remove test in scraping with filtering

This is because the scraping diagnostic does not output all particles in that case (particle filtering is on).

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

…m solution (ECP-WarpX#4889)

* Use amrex logic for device synchronization in tiny profiler instead of custom solution

* improve comment

* remove useless comment

* fix missing include

* fix bug

* fix bug

* Add None particle boundary condition

* Use None particle boundary in CI test langmuir/inputs_rz

* Update BC for picmi input files

…WarpX#4947)

This failed if `AMREX_USE_OMP` is undefined.

…arpX#4951)

* picmi: allow different `random_fraction` or `uniform_stride` for different species

* fix codeQL issue

* use dictionary to specify particle diagnostic stride or random fraction

* use 1 as stride and random fraction default for unspecified species

* use species rather than species.name as key for random fraction dict

* update docstring

* AMReX: Weekly Update

* pyAMReX: Weekly Update

* CMake: `WarpX_PSATD` -> `WarpX_FFT`

Rename `WarpX_PSATD` to `WarpX_FFT` in the build configuration
options. We now use FFTs for more than the PSATD solver since
the introduction of the IGF solver for electrostatic
simulations.

* Change Define: `WARPX_USE_PSATD`

* GNUMake: `USE_PSATD` -> `USE_FFT`

* Temporary backwards compatibility

We keep the flag `WarpX_PSATD` as an alias around for a few
releases before we drop it.

* changed  2

* add assert to check fft compilation

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Update Source/ablastr/fields/PoissonSolver.H

Co-authored-by: Axel Huebl <[email protected]>

* Update Source/ablastr/fields/PoissonSolver.H

Co-authored-by: Axel Huebl <[email protected]>

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Axel Huebl <[email protected]>

… (JFNK) for the nonlinear solver. (ECP-WarpX#4736)

* created Source/FieldSolver/WarXImplicitFieldsEM.cpp file to store functions used with the implicit solvers. Only has ComputeRHSE() and ComputeRHSB() functions thus far.

* OneStep_ImplicitPicard routine now uses ComputeRHSE and ComputeRHSB functions in place of EvolveE and EvolveB functions.

* created Source/Evolve/WarpXImplicitOps.cpp file to store common functions used by implicit solvers. Moved several functions from WarpXOneStepImplicitPicard.cpp to here.

* added NonlinearSolver:: enum. Options are picard and newton. Set from input file using algo.nonlinear_solver = picard, for example. Default is picard. It is not used yet.

* changed EvolveScheme::ImplicitPicard and SemiImplicitPicard to ThetaImplicit and SemiImplicit, respectively. This affects parsing from input file. algo.evolve_scheme = implicit_picard ==> theta_implicit. Have not updated Docs yet.

* NonlinearSolver ==> NonlinearSolverType

* intermediate commit. Added an ImplicitSolversEM class. WarpX owns an instance of this class. OneStepImplicitEM is now done here. This class owns all things pertaining to the implicit integrator. Also added a NonlinerSolvers class, but it is not used yet.

* cleaning up ImplicitSolverEM::OneStep().

* more refactoring of ImplicitSovlerEM class.

* WarpXFieldVec ==> WarpXSolverVec

* removed depricated functions. WarpXSolverVec has zero ghost.

* ImplicitSolverEM::OneStep now looks exactly like Picard solver. Next step is to move it there.

* ImplicitSolverEM::OneStep() now uses the Picard solver object to solve the nonlinear equation.

* changed where implicit solver parameters are printed.

* refactoring of WarpXImplicitOps.cpp

* added NewtonSolver.H file. Doesn't work yet.

* adding more functionality to WarpXSovlerVec class.

* added JacobianFunctionJFNK.H file in NonlinarSolvers/ folder. It contains all of the necessary functions required by the linear operator template parameter for AMReX_GMRES.

* dotMask object used for dot product from Linear Function class now lives in the implicit solver class. This ensures that 1 and only 1 instance of this object will be defined. dotProduct and norm can now be called through the implicit sovler class.

* moved temporary linear_function and GMRES testing lines out of Picard::Define() and into Newton::Define()

* intermediate commit. JFNK almost ready.

* small refactoring of PreRHSOp() and PostUpdateState() functions.

* cleaning things up.

* Newton solver runs. GMRES runs. Next step is to do Particle-suppressed (PS) part.

* minor clean up.

* fixed typo in convergence message for Newton and Picard solvers.

* changed how PostUpdateState() is used in implicit solver. Now parsing Picard particle solver parameters in ImplicitSolverEM class. Using a new formula for the epsilon used in the finite-difference Jacobian action calculation that is suitable for large absolute norms of the solution vector.

* Picard method for particle is now being used. PS-JFNK works.

* moved WarpXImplicitOps.cpp from Source/Evolve/ to Source/FieldSolvers/ImplicitSolvers/

* minor cleanup. PostUpdateState() ==> UpdateWarpXState().

* Moved the particle convergence check into its own function.

* added increment function to WarpXSolverVec class.

* removed some commented out lines in JacobianFunctionJFNK.H

* removed a_tol condition for print message when maximum iterations reached in nonlinear solvers. Newton solver iter is base zero now.

* cleaned up picard method for self-consistent particle update. added ablastr warning message for particles that don't converge after the maximum number of iterations.

* fixed small bug in PicardSolver related to absolute tolerance check for convegence. Default absolute tolerance values are set to zero.

* the mask used to compute the dot product of a WarpXSolverVec is now owned by the WarpXSolverVec class rather then the time solver class. It is a static member to avoid multiple definitions.

* defined accessors for Efield_fp and Bfield_fp Vectors owned by WarpX. ImplicitSolver is no longer a friend class of WarpX. Small tidy for PicardSolver.H.

* SemiImplicitEM and ThetaImplicitEM are now their own independent derived classes from the base ImplicitSolver class.

* added algorithm descriptions to the top of the SemiImplicitEM.cpp and ThetaImplicitEM.cpp files.

* updated appropriate files in Examples and Regression folders to reflect new changes.

* updating docs.

* JacobianFunctionJFNK.H ==> JacobianFunctionMF.H

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* need to call clear on the static vector m_dotMask owned by the WarpXSovlerVec class to avoid malloc_consolidate(): unaligned fastbin chunk detected error messages at simulation finish.

* moved WarpXSovlerVec class from ../ImplicitSolvers/Utils/ to ../ImplicitSolvers/. Utils directory is deleted.

* ImplicitPushXP: GPU Support for Convergence Test

* cleaning up.

* Atomic: Fix type of `1`

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* bug fix.

* Update Source/FieldSolver/ImplicitSolvers/WarpXSolverVec.H

Co-authored-by: Revathi  Jambunathan <[email protected]>

* Remove Zero

* adding Copyright lines to top of files. Removed commented code.

* Prevent calling the implicit solver if not initialized

* More robust if conditions

* set implicit verbose to warpx verbose

* Update Source/FieldSolver/ImplicitSolvers/ThetaImplicitEM.cpp

* Simplify call to updates of E

* changed benchmarks_json file names as needed.

* using warpx.verbose

* clang-tidying

* changed header names.

* clang-tyding

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* more clang-tyding

* clang tidy again

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* UpdateElectricFiled ==> SetElectricFieldAndApplyBCs

* clang tidy

* more clang tidy

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* prohibiting copy and move constructors for solver classes.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fixed bug in move constructor in WarpXSovlerVec.

* slight refactoring of ThetaImplicitEM class.

* reducing divisions in Picard method for particles.

* small cosmetic changes to implicit solvers.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* removed commented out code.

* updating Docs and adding briefs.

* Fix HIP compilation

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix const

* fixed indent. updated comments.

* New Python test: Particle-Boundary interaction  (ECP-WarpX#4729)

* enable the diagnostic of ParticleScraping in Python

* Update picmi.py

* Update picmi.py

* new test

* python update

* modification of the script

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Update PICMI_inputs_rz.py

* json

* update

* Update PICMI_inputs_rz.py

* Update particle_boundary_interaction.json

* Update PICMI_inputs_rz.py

* Update PICMI_inputs_rz.py

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Update PICMI_inputs_rz.py

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Fix hanging script in parallel

* Make the test executable

* Update analysis script

* Update particle_containers.py

* Update PICMI_inputs_rz.py

* Update analysis.py

* Update analysis.py

* Update particle_containers.py

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Remi Lehe <[email protected]>
Co-authored-by: Axel Huebl <[email protected]>

* Adding normal components to regular boundary buffer (ECP-WarpX#4742)

* first draft

* adding normal only

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Update ParticleBoundaryBuffer.cpp

* Update ParticleBoundaryBuffer.cpp

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

* Add function to set domain boundary potentials from Python (ECP-WarpX#4740)

* add function to set domain boundary potentials from Python

* switch function arguments to form `potential_[lo/hi]_[x/y/z]` and add to docs

* clean up `ablastr/fields` (ECP-WarpX#4753)

* move PoissonInterpCPtoFP to Interpolate.H

* concatenate nested namespaces

* Split clang-tidy CI test into 4 to improve performances (ECP-WarpX#4747)

* split clang-tidy checks to improve performances

* rename folders and tests

* fix concurrency

* Simplify

---------

Co-authored-by: Axel Huebl <[email protected]>

* Replace links to learn git (ECP-WarpX#4758)

* Replace links to learn git

* Bugfix in `fields.py` for GPU run without `cupy` (ECP-WarpX#4750)

* Bugfix in `fields.py` for GPU run without `cupy`

* apply suggestion from code review

* Release 24.03 (ECP-WarpX#4759)

* AMReX: 24.03

* pyAMReX: 24.03

* WarpX: 24.03

* Implement stair-case Yee solver with EB in RZ geometry (ECP-WarpX#2707)

* Allow compilation with RZ EB

* Do not push cells for RZ Yee solver, when covered with EB

* Fix compilation errors

* Fix additional compilation errors

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Fix additional compilation errors

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Add automated test

* Add automated test

* Fix path in tests

* Enable parser in RZ

* Update example script

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Clean-up PR

* Initialize EB quantities

* Modified EM field initialization in 2D with EB

* Typo fix

* Typo fix

* Ignoring unused variables correctly

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Correct condition for updating E

* Correct update of B

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Update B push

* Update input script

* Revert "Update input script"

This reverts commit 5087485.

* Update initialization

* Updated test

* Move test to a different folder

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Add test for WarpX-test.ini

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Fix path for tests

* Update test description

* Update test metadata

* Add checksum file

* Revert changes

* Revert changes

* Change lx to lr

* Revert "Change lx to lr"

This reverts commit be3039a.

* Change lx to lr

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: lgiacome <[email protected]>

* AMReX/pyAMReX/PICSAR: Weekly Update (ECP-WarpX#4763)

* AMReX: Weekly Update

* pyAMReX: Weekly Update

* clean up (ECP-WarpX#4761)

* Fix compilation

* updating some function names to contain Implicit.

* fixed bug that caused segfault on GMRES restart.

* parsing GMRES restart length from input file.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Update field accessor

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* removed +, -, and * operators from WarpXSolverVec class. These operators encourage inefficient vector arithmetic.

* fix/workaround to field accessor issue.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* changing implicit-scheme related header files in WarpX.H and WarpX.cpp

* WarpX::max_particle_iterations ==> WarpX::max_particle_its_in_implicit_scheme and WarpX::particle_tolerance ==> WarpX:particle_tol_in_implicit_scheme

* updating docs.

* updating docs again.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* adding comments about the template parameters Vec and Ops used in the NonlinearSolver class.

* adding comments. addressing comments from PR.

* Ensure that laser particles can converge in the implicit solver

* Add braces to make clang-tidy happy

* moving nonlinear solver parameters to base ImplicitSolver class

* mirrors not to be used with implicit schemes.

* moved more to base implicit solver class. adding comments.

* removed some WarpXSolverVec functions. Updated comments.

* clang tidy complains when removing default copy constructor in WarpXSolverVec.H

* amrex::ExecOnFinalize(WarpXSolverVec::clearDotMask)

* WarpXSolverVec (const WarpXSolverVec&) = delete

* updating briefs for nonlinear solvers.

* adding loop over levels.

* static cast amrex::Vector.size() to int

* updating docs for nonlinear solvers.

* adding gmres.restart_length to docs.

* fixed typos in docs.

* Removed PreRHSOp() call from nonlinear solvers.

* clang tidy.

* Prohibit = operator for WarpXSolverVec. Using Copy() instead.

* Document PICMI function `LoadInitialField`

* updating comments in WarpXImplicitOps.cpp

* moved static member m_dotMask definition to the header file with inline added to the declaration.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fixed indent.

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Axel Huebl <[email protected]>
Co-authored-by: Debojyoti Ghosh <[email protected]>
Co-authored-by: Revathi  Jambunathan <[email protected]>
Co-authored-by: Remi Lehe <[email protected]>
Co-authored-by: Weiqun Zhang <[email protected]>
Co-authored-by: Eya D <[email protected]>
Co-authored-by: Roelof Groenewald <[email protected]>
Co-authored-by: Arianna Formenti <[email protected]>
Co-authored-by: Luca Fedeli <[email protected]>
Co-authored-by: lgiacome <[email protected]>

…ing (ECP-WarpX#4969)

* add upper bound for weight parameter on particle merging

* increase readability, fix clang-tidy test

* rework from ignoring particles above some maximum weight to instead resample particles such that they stay under a target weight

* avoid implicit capture of this, cleanup

* remove leftover code

Co-authored-by: Roelof Groenewald <[email protected]>

* clarify docstring

* clarifying comment

---------

Co-authored-by: Avigdor Veksler <[email protected]>
Co-authored-by: Roelof Groenewald <[email protected]>

* add initialization of extra attribs to picmi

* avoid new `Bucket`

* update CI checksum

* update ionization test analysis to always include runtime attribute

* Update Examples Further, Fix CI

* update docstring; update checksum values

* allow user defined attributes to be added to `ParticleDiagnostic` output

* more CI fixes

* revert adding `z_orig` to inputs_2d_bf_rt

---------

Co-authored-by: Axel Huebl <[email protected]>

* Add temperature diagnostic

* Add to CI test

* Add input to picmi

* Add comment on the method

* const cleanup

* Fix particle array declaration

* Update benchmark for CI

* Clean up BeamRelavent diagnostics, using get_particle_position

* In Field diagnostics, simplify calculation of dV

* Clean up of ParticleExtrema

* In WarpXOpenPMD, removed duplicate particle coordinate transformation with RZ

* Added more amrex prefixes in ParticleExtrema

* Fix const

* Small cleanup in WarpXOpenPMD

* Clean up calculation of dV

* Cleanup call to CellSize

* Fix in FieldEnergy

* AMReX: 24.06

* pyAMReX: 24.06

* WarpX: 24.06

The constructor of FieldProbe might be called earlier than the
WarpX class parameter init. That could lead to relying on an
uninitialized particle shape static. Use the parser instead, similar
to our general efforts to reduce the static members stored in the
WarpX class.

* CMake: heFFTe

* Doc: Perlmutter heFFTe

* Update Spack Dev Envs

* Finalize After Testing

Co-authored-by: Alfred Mishi <[email protected]>

---------

Co-authored-by: Alfred Mishi <[email protected]>

Resetting after a recent change. Might need more investigation.

* Initial clean up of current deposition

* Fix of current deposition

* Further cleanup of current deposition

* Cleanup SharedDepositionUtils

* Clean up in charge deposition

* Clean up field gather plus other fixes

* Change LowerCorner to return XDim3, add InvCellSize

* Change variable name in Lower and UpperCorner to avoid confusion

* Add const

* Update RepellingParticles CI test

* Update Python_ionization CI test

* Update ionization_boost and ionization_lab CI tests

* Update PEC_particle CI test

* Update BTD_rz CI test

Generalize text post ECP.
(Various funding sources from many people.)

…pX#4998)

* Doc: Resonant excitation of plasma waves in a plasma channel

WarpX used in the paper by A. J. Ross et al., PRR 6 (2024)

* Nguyen, B et a. IPAC24

* Sort by Pub Year

* Update PASC24 in other sections

* add warning instead of error

* Apply suggestions from code review

Co-authored-by: Roelof Groenewald <[email protected]>

---------

Co-authored-by: Roelof Groenewald <[email protected]>

* AMReX: Weekly Update

* pyAMReX: Weekly Update

Need to ignore time macros.

Update our HankelTransform logic calling into BLAS++ by using the latest APIs for `blas::Queue`.
Pass the active AMReX GPU stream.

For BLAS++ and LAPACK++, switch the documented HPC dependency install
logic from using the latest commit on `master` to the latest release
`2024.05.31`. Using tags instead of living branches increases
robustness and reproducibility with regards to future changes.

* AMReX: Weekly Update

* pyAMReX: Weekly Update

* obvious bug fix in ElasticCollisionPerez

* updating benchmark json files for collisionISO and collisionXYZ

* Add random seed to Examples/Tests/collision/inputs_3d

* Update benchmark values

* Update benchmark again

* simplify VelocityProperties.cpp

* fix issues

* touch particle tiles before OMP loop

* touch tiles before OMP loop

…pX#5024)

* Update default behavior for gathering with direct deposition

* Modify condition under which to use galerkin interpolation

* Update condition

* Update benchmarks

---------

Co-authored-by: Edoardo Zoni <[email protected]>

* AMReX: 24.07

* pyAMReX: 24.07

* WarpX: 24.07

…WarpX#5028)

While working on another PR, I noticed that the CI test `Langmuir_fluid_2D` was compiled in debug mode.

The test was added originally in ECP-WarpX#3991 and it might be that the debug build type was left over unintentionally.

In general, I think we avoid running CI tests in debug mode, in order to keep the runtime of the tests as low as possible (the current runtime of this test in debug mode is around 80 seconds). 

However, we might have changed policy in the last months and I might not be up-to-date, so feel free to let me know and close this PR without merging if it is not relevant.

* Update documentation

* Allocate fields

* Always parse filename

* Add option "FromFile" in external particles fields

* Create two separate examples

* Update documentation

* Allocate aux array

* Update aux to take into account copy

* Add external fields from file to auxiliary data

* Update tests

* Remove debugging Print statements

* If averaged PSATD is used, we copy Efield_avg_fp to Efield_aux, otherwise copy Efield_fp.

* Do not access cells that have invalid data when summing external fields

* Update path of checkpoint files

* Update comments

* Update number of components

* Fix clang tidy error

* Make the code compatible with momentum-conserving gather

* Add Python interface and tests

* Use PICMI version that defined LoadAppliedField

* Update file names

* Update PICMI standard version

* Allocate dedicated MultiFabs for external particle fields

* Some clean-up

* Performance optimization: do not use external fields in gather kernel

* Allow external particle fields and external grid fields to be simultaneously defined

* External fields are incompatible with the moving window

* Support load balancing

* Update Source/Initialization/WarpXInitData.cpp

* Update Source/Initialization/WarpXInitData.cpp

---------

Co-authored-by: Remi Lehe <[email protected]>

* Turn on particle comparison

* New value for checksum

…CP-WarpX#5037)

…arpX#5038)

…rk for particles of varying weight. (ECP-WarpX#5040)

…X#5043)

* CI: remove unused params, check particles in `collisionXYZ`

* Update benchmark

---------

Co-authored-by: Remi Lehe <[email protected]>

…ons (ECP-WarpX#5045)

* shuffling the full list of particles for intra-species binary collisions

* fixing merge issue with collision json files.

* moved checksum to before error assert for intraspecies DD fusion regression test analysis script.

* updating benchmark values for DD_fusion_3D_intraspecies.

* adding fixed random seed to DD_3D_intraspecies fusion regression test.

* putting checksum back after assert.

* adjusting json.

* updating json.

Update TOSS3 install scripts.

Use the correct ABLASTR (not WarpX) CMake options and compiler
defines. E.g., for ImpactX, we only control ABLASTR, not WarpX.

* AMReX: Weekly Update

* pyAMReX: Weekly Update

Funny typo.

`pybind11::lto` is only defined if `CMAKE_INTERPROCEDURAL_OPTIMIZATION`
is not set in `pybind11Common.cmake`. Package managers like Spack
use the latter.

…-WarpX#5060)

Sadly, the TOSS4 transition was not completed before the IBM contract ran out and Lassen stays with TOSS3. This simplifies the documentation again.

* moved density and temperature calc from each pair to once for each cell.
* pulled out n12 calc to cell level.
* fixed dV issue for cyl geom.

* AMReX: Weekly Update

* pyAMReX: Weekly Update

…w moves. (ECP-WarpX#5082)

* Recompute the macroscopic properties everytime the moving window moves.

* Minor cleanup

* Separate allocation and initialization

* update profile script for Frontier

* satisfy dependency for module load

* change module versions and manually install adios2

* fix bug

* fix bug

* update instructions

* update scripts

* remove line for LibEnsemble

* PICMI (Bucket): NumPy 2.0 Compatibility

`repr` returns `'np.float64(value)'` instead of `'value'` in
NumPy 2.0

* Use fstrings
* Remove outdated comment
* `requirements.txt`: Relax `numpy` version

Version 1 and 2 support.

* Remove outdated comment

---------

Co-authored-by: David Grote <[email protected]>

…5091)

* using corrected weighted-particle Coulomb collision method.

* adding CI test for weighted Coulomb collisions.

* using n12 in UpdateMomentumPerezElastic

* updating Checksum

* fixing type issue.

* updating checksums

* checksum

* AMReX: Weekly Update

* pyAMReX: Weekly Update

* Increase Clang-Tidy `timeout-minutes`

* Cast long long to int when using IParser

---------

Co-authored-by: Edoardo Zoni <[email protected]>

…CP-WarpX#4922)

* Removal of asserttion which prevented from usung the averaged PSATD algorithms with PML BC

* Clean-up

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Fixed to arr_aux(j,k,l) = fine when ES solve is used

* Removed temporary print statements

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Clean-up

* Added CI test ElectrostaticSphereEB_RZ_MR_lev_1 to check the fields on the level=1

* Updated becnmarks for ElectrostaticSphereLabFrame_MR_emass_10

* Imported regular expression (re) in the analysis script.

* Fixed typo

* United two CI tests for different levels of MR in one test.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Updated CI test ElectrostaticSphereEB_RZ_MR and the corresponding analysis script with  smaller MR patch.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Clean-up

* Do deepcopy for lev>0 and collocated grid

* Fix bugs to resolve failure of CI tests

* Preserve plotfile output, update benchmark file

* Working on  CI test

* Update benchmark of `ElectrostaticSphereEB_RZ_MR`

* Initialize with value all `aux`, `cax` fields

* Remove changes related to averaged Galilean PSATD with PML

* Remove style changes (e.g., changes to empty lines)

* Replace `MFIter`/`ParallelFor` loop with simple copy

* Apply suggestions from code review

* Revert part of the code to its previous, equivalent state

* Removed DeepCopy & no need for ghost cells in temp phi_cp.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Update Source/ablastr/fields/PoissonSolver.H

* Add inline comments

* Update analysis script

* Use `TilingIfNotGPU`, `growntilebox` in E loop

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edoardo Zoni <[email protected]>
Co-authored-by: Edoardo Zoni <[email protected]>
Co-authored-by: Remi Lehe <[email protected]>
Co-authored-by: Weiqun Zhang <[email protected]>

Add agreed upon HPSF boilerplate text in README and manual.
Update acknowledegments.

* add: profile, dependencies, and sbatch scripts for OSC pitzer

* add: pitzer documentation page

* fix: add pitzer name to manual sidebar

* chore: update pitzer dependencies installation script filename

* fix formatting issues in doc

* separate cpu and gpu scripts

* add boost support

* add separate instructions for CPU and GPU dependencies

* rename venv name

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* add header to dependencies intalling scripts

* chores for doc

* fix: dependencies installing scripts typo

* fix typos

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* chore: Update batch script and profile file names for Pitzer CPU and V100

* remove gpu allocation in CPU profile

* Update Docs/source/install/hpc/pitzer.rst

Co-authored-by: Axel Huebl <[email protected]>

* Update Docs/source/install/hpc.rst

Co-authored-by: Axel Huebl <[email protected]>

* Update Docs/source/install/hpc/pitzer.rst

Co-authored-by: Axel Huebl <[email protected]>

* use built-in cuda-aware MPI

* Remove repetition of source profile

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Axel Huebl <[email protected]>

We cannot use global variables from WarpX in the ABLASTR library,
otherwise it will not build on its own or in ImpactX.

* Implement PICMI interface for implicit solvers

* Add CI test

* Add ExplicitEvolveScheme

* Add evolve scheme classes to PICMI documentation

* Made evolve_scheme argument to Simulation

* Small fix, removing unused variable

* Apply suggestions from code review

---------

Co-authored-by: Remi Lehe <[email protected]>

* use external field multifabs consistently for different external field types

* make external field MF vectors length equal to `nlevs_max`

* avoid adding external grid fields multiple times

* load parse ext grid fields after restart

* rename `LoadExternalFieldsFromFile` to `LoadExternalFields`

* Update inline comment on why external fields are added to ES solutions

Co-authored-by: Remi Lehe <[email protected]>

---------

Co-authored-by: Remi Lehe <[email protected]>

…#5114)

This test uncoveres a bug of an invalid pc passed early
to I/O routines.

We will add it back with the fix.

* Debugging CI tests

* Run all Cartesian 3D tests

* Clear cache after python installation commands

* Update Regression/WarpX-tests.ini

* `run_test.sh`: Detailed `df -h` Prints

* Update checksum

* Cleanup

---------

Co-authored-by: Remi Lehe <[email protected]>
Co-authored-by: Axel Huebl <[email protected]>

…40807