AMReX/PICSAR: Weekly Update

[ax3l]

Weekly update to latest AMReX.
Weekly update to latest PICSAR (no changes).

./Tools/Release/updateAMReX.py
./Tools/Release/updatePICSAR.py

Notes

• regressions from Rework handling of roundoff domain extent AMReX-Codes/amrex#3199 Fix periodic boundary bug in #3199 AMReX-Codes/amrex#3243
• need fix from Fix Roundoff Domain AMReX-Codes/amrex#3247

[ax3l]

a few checks that raise checksum differences:

• hard_edged_quadrupoles (3D)

Benchmark: [lev=0,Ex] 9.882421125782868e-06
Plotfile : [lev=0,Ex] 9.882421146615367e-06
Benchmark: [lev=0,Ey] 1.044026102498192e-05
Plotfile : [lev=0,Ey] 1.044026104671425e-05
Benchmark: [lev=0,Ez] 1.003739698556440e-05
Plotfile : [lev=0,Ez] 1.003739697324731e-05

• pml_psatd_dive_divb_cleaning (3D)

Benchmark: [lev=0,Bx] 1.482006333945839e-07
Plotfile : [lev=0,Bx] 1.240175732139287e-06
Benchmark: [lev=0,By] 1.482051585061588e-07
Plotfile : [lev=0,By] 1.219308776050087e-06
Benchmark: [lev=0,Bz] 1.495470407235984e-07
Plotfile : [lev=0,Bz] 1.229402125805917e-06
Benchmark: [lev=0,Ex] 1.178979338985698e+01
Plotfile : [lev=0,Ex] 6.142609646226940e+02
Benchmark: [lev=0,Ey] 1.178688476447710e+01
Plotfile : [lev=0,Ey] 6.224773198394323e+02
Benchmark: [lev=0,Ez] 1.177011081446856e+01
Plotfile : [lev=0,Ez] 6.135902829786862e+02
Benchmark: [lev=0,rho] 4.903696256562049e-05
Plotfile : [lev=0,rho] 4.903696525896805e-05

• background_mcc_dp_psp (2D)

Benchmark: [electrons,particle_momentum_z] 2.810005821001351e-19
Plotfile : [electrons,particle_momentum_z] 2.810005842534688e-19
Benchmark: [electrons,particle_position_y] 9.356698625905589e+02
Plotfile : [electrons,particle_position_y] 9.356698666088197e+02

• reduced_diags_single_precision (SP)
  also very small stuff...

[ax3l]

@WeiqunZhang I don't understand the relatively large changes in pml_psatd_dive_divb_cleaning... This test does not even have particles.

[RemiLehe]

@ax3l This test does have particles, I think, since the laser is emitted by an antenna made of macroparticles with a prescribed motion.

[ax3l]

Oh I forgot again, ok that makes sense (including the rho).

Should we print the fields before and after to see it still makes sense? There is a significantly larger change in the fields for this one, concerning me the antenna might have changed now or the box might be too small of the initialized laser in the test.

[ax3l]

Updating once more to include AMReX-Codes/amrex#3248

[ax3l]

That test with the larger change in checksum (pml_psatd_dive_divb_cleaning) has a couple of warnings

* --> [!  ] [Algorithms] [raised once]
*     The chosen current deposition algorithm does not guarantee charge
*     conservation, and no additional current correction algorithm is activated
*     in order to compensate for that. Lack of charge conservation may
*     negatively affect the results of the simulation.
*     @ Raised by: ALL
*
* --> [!  ] [PML] [raised once]
*     Using PSATD together with PML may lead to instabilities if the plasma
*     touches the PML region. It is recommended to leave enough empty space
*     between the plasma boundary and the PML region.
*     @ Raised by: ALL

and might be too small for the sim box.

Don't really know what to make of it.
Untitled.zip

@EZoni as main author of the test, what is the expected output it checks? :) https://github.com/ECP-WarpX/WarpX/commits/development/Examples/Tests/pml/inputs_3d

[ax3l]

In order to go forward, I would merge w/o the pml_psatd_dive_divb_cleaning test being fully understood. We thing the laser antenna is too wide but have not a lot of time to dig deeper into it rn.