Age | Commit message (Collapse) | Author | Files | Lines |
|
Initially we wrote those on open. The problem was that writing this
file depends on an existing directory structure, which is first
created on earlierst on the first `flush()` and in case of extensive
ADIOS buffering actually latest on `close()`.
Since in backtransformed diagnostics (BTD) this flush happens very
late. The directory did not exist and thus the creation of the
helper file early on did fail (silently).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* Weekly Update: AMReX & PICSAR
```
./Tools/Release/updateAMReX.py
./Tools/Release/updatePICSAR.py
```
Updates AMReX to latest `devevelopment` commit
(`0ab63f3f93950b5a90fb091c6eabbf3239886fcd`).
No changes in PICSAR.
* bump amrex again
Co-authored-by: Andrew Myers <atmyers2@gmail.com>
|
|
|
|
* Remove the avg_aux fields
* Update 3D benchmark
* Update benchmark in 2D
* Initialize Bfield_avg_cp
* Fix two whitespaces left out
* Fix two more whitespaces left out
Co-authored-by: Edoardo Zoni <59625522+EZoni@users.noreply.github.com>
|
|
|
|
* Read nghost as Numpy Array From String
* Do Not Use non-tuple Sequence For Multidimensional Indexing
|
|
Avoid writing QED particle attributes as scalars.
Needs testing.
|
|
* add PSATD support when remaking level
* eol
* update_with_rho
* Update WarpXRegrid.cpp
* Upate WarpXRegrid.cpp
* Update WarpXRegrid.cpp
* WIP
* Cleanup
* Remove
* Add function to allocate spectral solver
* EOL
* spaces
* Remove comment
* Update documentation
* eol
* Remove function arguments
* fix
* Move lines into function
* eol
* amrex::Real
* args
|
|
* Start adding centering of current
* Implement arbitrary order centering and split inputs
* No need to define a brand new interpolation function
* Update input file of hybrid CI tests
* Clean up
* Clean up more
* Fix bug and clean up
* Use current centering in two existing CI tests
* Update documentation
* Move Calls To UpdateCurrentNodalToStag Into SyncCurrent
* Add Doxygen For New Function UpdateCurrentNodalToStag
* Add Doxygen For New Functions Used For Stencil Coefficients
* Finite-Order Centering of Currents Not Implemented With MR
|
|
* Added RZ Er, Et, Ez and Br, Bt and Bz
* Answered my own question
* Making Yee and cells domain check for RZ too
* Added relevant coefficients
* Derivative in z requires coefsz and n_coefsz
* Bt computed with UpwardDz
* UpwardDz might require CylindricalYeeAlgorithm.H
* Added mode 0 Bz
* Added higher-order modes for Bt
* Added higher-order modes for Bz
* Fix to EOLs
* Fix typo
* Added cylindrical specific parameters
* Fix error of #endif
* rmin also needed in RZ parameters
* T_Algo needed for RZ -> different initialization in .H
* Fix private / public function
* Replacing T_Algo by CylindricalYeeAlgorithm
* Fix typo
* ParallelFor for Br, Bt and Bz separated
* Compiled after removing unecessary Br
* Changes suggested by reviewer
* No need to compute r before if()
* Corrected real and imaginary parts of Bz
* Remove vscode file
Co-authored-by: Remi Lehe <remi.lehe@normalesup.org>
|
|
* Split Initialization Functions of Spectral Coefficients
* Simplify Initialization of Coefficients Without Averaging
* Do Not Store Coefficients C1,S1,C3,S3 With Averaging
* Simplify Initialization of Coefficients With Averaging
* Add amrex:: Prefix Following WarpX Style Guidelines
* Match Names Of Coefficients With/Without Averaging
* 'pow' Cannot Be Used in a Constant Expression
* Update Doxygen Documentation
|
|
Save 1-2 GB by cleaning this out.
|
|
Add more badges to shout-out to communities & dependencies.
|
|
* More Badges: Install, ECP
* update: funding ECP
* haaands
|
|
|
|
HIP 4.1 introduces issues with C targets which break the
MPI feature test in CMake's `FindMPI.cmake` for C.
|
|
|
|
- fix minor typos in example commands
- simplify/clarify `git clone` to be uniform
- modernize Juwels section to CMake
|
|
Add more badges to the README file (more badges, more creds,
obviously).
We will also declare the development phase from now on *beta*
(21.04+) :tada:.
|
|
Fix a missing note.
|
|
|
|
The big restructuring & streamlining of our documentation.
|
|
Print the WarpX and PICSAR version to the terminal.
Fetched from `git` as usual. Will use a proper tag-prefixed
description as soon as we add tags in the git history.
|
|
Add a verbose message if IPO is used and also build AMReX with
IPO/LTO.
WarpX binary size (default build, GCC 9.3.0 on x86_64 with
Ubuntu 20.04):
* no IPO: 217MiB
* WarpX IPO, AMReX no IPO: 155MiB
* both WarpX and AMReX with IPO: 127M
|
|
Update AMReX to 384ca2341889a8f1db78e87cb4b1735bd2d58554.
No new changes in PICSAR.
|
|
|
|
|
|
* PSATD+RZ: Fix Warnings
Fix a couple of warnings in PSATD + RZ.
* BesselRoots: Missing Include & using
* Kernel & Util: Unused Code & Var
|
|
dims (#1785)
* Fix for diff spatial dims
* eol
* Update LoadBalanceCosts.cpp
* Update LoadBalanceCosts.cpp
* Restart CI
|
|
|
|
- remove some binaries we do not use in CI
- build shared to avoid any copies of AMReX
|
|
* Introduce relative_time for deposition
* Deposit current at arbitrary time
* Update documentation
* Combine parameters in DepositCurrent
* Explicitly show the time at which the current is deposited
|
|
|
|
|
|
The sorting defaults changed to per-cell.
|
|
|
|
ParaView 5.9+ adds initial support for openPMD files.
We need the filename passed to the `openPMD::Series` constructor
as a little text helper file to open the files properly, this adds this.
|
|
* add option to pass a0 instead of e_max for the laser
e_max xor a0 must be specified
* update doc for laser a0
* a0: Update Examples
Using all options so they are tested.
* reset benchmarks (precision E0->a0)
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|
|
* KnlTimer cleanup
* Update KernelTimer.H
* minor
* Update Source/Particles/Deposition/CurrentDeposition.H
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|
|
* add reference to WarpX slides in the acknowledge us section
* eol
* replace WarpX citation with the 3-stage one in acknowledgement section
|
|
* Read boundary and set periodicity, enumerate BC types, added support for periodic
* eol
* separate particle and field boudnary structs
* Update comment for particle struct
* the valid values are for field/particle
* adding docs
* eol
* remove warning
* add doc only for periodic
* typo
|
|
* Move laser particle container
* Update CMake file
* Fix compilation errors
|
|
* Add function definitions for Silver-Mueller boundary condition
* Add test file
* Implement Silver-Mueller equations
* Add equations for Bz
* Add equation in the z direction
* Add equation for Bx
* Add 3D implementation
* Add documentation, automated tests, and warnings
* Include second automated test and evaluate benchmark
* Fix unused variable warning
* Fix minor bugs
* Update benchmark
* Apply suggestions from code review
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
* Incorporate review comments
* Incorporate more suggestions
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|
|
Modern FFTW versions support CMake as an install method, which
provides the legacy goodies (`.pc`) files and proper
`FFTW3(f)Config.cmake` files for dependency propagation.
Conda-forge uses the CMake-based method to deploy. Checking this,
I realized the `.pc` files add unconditionally `libm` to the linker,
which is wrong on Windows and fails the builds.
Thus, forward looking we would search first for CMake based installs
of FFTW and in case this search is not successful, we continue to
search for the old Autotools logic and `.pc` files, which are still
routinely generated by some package managers.
But, due to an issue in FFTW3 3.3.9 on Linux/macOS with Autotools
installs, we make `Config.cmake` file based search still opt-in for
now.
|