Age | Commit message (Collapse) | Author | Files | Lines |
|
Reflect new project precedence option `AMReX_BUILD_SHARED_LIBS`
in superbuild install logic defaults for AMReX.
|
|
Add Ascent and SENSEI minimal versions to our dependency doc page.
|
|
Document how to work-around Jupyter HDF5 issues on OLCF until they
fix it.
|
|
Slurm at Perlmutter is not yet configured to do proper affinity
control itself. Thus, we do it ourselves.
Co-authored-by: Kevin Gott <kngott@lbl.gov>
|
|
* allow arbitrary laser antenna normal - picmi
* always check that laser propagation direction is along the normal direction of the antenna if the antenna normal is given
|
|
* AMReX: 22.11
* PICSAR: 22.11
* WarpX: 22.11
|
|
Migrate CUDA CI to use Ubuntu 18.04 -> 20.04
|
|
* Perlmutter (NERSC): Update Modules
Update after major system upgrade.
* Perlmutter (NERSC): Update Jobscript
* Perlmutter (NERSC): Remove Early Access Warning
* Perlmutter (NERSC): Process Affinity
Co-authored-by: Kevin Gott <kngott@lbl.gov>
Co-authored-by: Kevin Gott <kngott@lbl.gov>
|
|
* Implement D+D -> n+He3 fusion
* Fix logic for fusion reaction
* Check products in a different place
* Correct compilation error
* Implement D+D -> T+p cross-section
* Update example
* Use clearer naming convention for fusion types
* Revert changes to example input script
* Add analysis script
* Progress on tests
* Use 2 species in test
* Correct momentum of colliding species
* Update test
* Update test
* Generalize species names in fusion tests
* Update benchmarks
* Correct typo
* Updated scripts
* Update script so that it works for D+T and D+D
* Update CI
* Add benchmark file
* Correct typo
* Fix compilation on GPU
* Update RZ CI test
* Implement Deuterium-Helium reaction
* Apply suggestions from code review
Co-authored-by: Edoardo Zoni <59625522+EZoni@users.noreply.github.com>
Co-authored-by: Edoardo Zoni <59625522+EZoni@users.noreply.github.com>
|
|
|
|
* Begin writing BTD FAQ notes
* FAQ notes on BTD
* access figure from github online
* update figure and improve faq answers
|
|
* cell center BTD functors for RZ with openpmd
* add RZ modes to output varnames too
* update varnames once and set map for RZ fields in BTfunctor
* fix access to varname field instead of unallocated m_varnames, and update hi ki-index after flush
* set back trasnformed fields to false if fields_to_plot is none. Separate PArticle and Field io further. Also, clean up Particle Buffer initialize so its consistent with Field box, boxarray, dmap, and geom
* clean commented line
* assert for single rz mode
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* remove unused var
* Apply suggestions from code review
From Axels' review
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
* adding comments, doxygen, and clean-up
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|
|
* Generalize species names in fusion tests
* Update benchmarks
* Correct typo
|
|
* Added error checks of the user input
* Fixed literals and const's
|
|
* workaround to fix invalid memory access while using PML + Moving Window + Load Balancing + Timers
* fixed bug
* add argument to shiftMF to avoid updating cost for PML
|
|
* ABLASTR: Always Sync Option
Add an option to ABLASTR/WarpX to always sync nodal cells that are
owned by multiple blocks. This can be used for debugging.
* Fix typo
Co-authored-by: Edoardo Zoni <59625522+EZoni@users.noreply.github.com>
* Use New Option in One CI Test
* Fix Logic: Force only to be `true`, not `false`
Co-authored-by: Edoardo Zoni <59625522+EZoni@users.noreply.github.com>
Co-authored-by: Edoardo Zoni <ezoni@lbl.gov>
|
|
|
|
* Docs: LUMI (CSC)
Start documenting LUMI usage instructions for WarpX.
* Add more system info
|
|
|
|
* Add warning to set zmin and zmax for ions
and also change the ion acceleration example to set these
parameters so that users are aware that memory overflow could
happen for production-size examples
furthermore, correct units in comment after definition of
r0 and Lcut
Replace zmin and zmax with variables for electrons
* Reset checksum for regression test
Reset benchmark for LaserIonAcc2d test because the particle
initialization in the input changed.
|
|
* Replace "none" BC with Neumann for electrostatic
* Added checking for Neumann boundaries
* Updated documentation regarding Neumann BCs
* [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>
|
|
plotfile/openPMD (#3424)
* print when writing openPMD, Ascent, Sensei
* remove unnecesssary variables, include min_digits
* print snapshot id when writing btd
* fix ascent print
* compile without warnings
* still working on Ascent
* still working Ascent
* Print when BTD buffer is flushed
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* print when writing openPMD, Ascent, Sensei
* remove unnecesssary variables, include min_digits
* print snapshot id when writing btd
* fix ascent print
* compile without warnings
* still working on Ascent
* still working Ascent
* Print when BTD buffer is flushed
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* add warning message if BTD not full
* adjust for CI, no in-situ vis BTD
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* revert to correct plotfile naming
* change warning topic to BTD
* nicer format for print statements
* extend examples to fill BTD
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Reset Checksums of BTD_ReducedSliceDiag
* Upgrade CI Test LaserAccelerationBoost
* Apply suggestions from code review
Co-authored-by: Edoardo Zoni <59625522+EZoni@users.noreply.github.com>
* improve BTD warning notes
* edit final step notes
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Fix formatting
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edoardo Zoni <ezoni@lbl.gov>
Co-authored-by: Edoardo Zoni <59625522+EZoni@users.noreply.github.com>
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|
|
oneAPI 2022.2.0 hangs for `-O2` and higher: #3442
|
|
* add reduced diagnostics to picmi
* added test of picmi reduced diagnostics
* changes requested during PR review
|
|
|
|
* Fix stair-case solver in 2D Cartesian
* Generalization for the macroscropic solver
* Update automated test
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update benchmark
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
|
* Do not query redundant do_back_transformed_particles
* Only set m_do_back_transformed_particles to true if write_species is true
|
|
* Fix unused import warnings in python analysis scripts
* [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>
|
|
Allow users to overwrite the default for the "used inputs" file that we write.
|
|
Mamba is now pre-installed on OLCF Jupyter base images! :)
|
|
|
|
* NVCC generates code that tends to trigger ICEs in GCC 7.x
* CTAD erratum in GCC 8: https://gcc.gnu.org/projects/cxx-status.html#cxx17
|
|
Fix:
```
$ cmake -S . -B build_dp ...
-- The C compiler identification is AppleClang 13.0.0.13000029
-- The CXX compiler identification is AppleClang 13.0.0.13000029
...
CMake Error at /usr/local/Cellar/cmake/3.24.2/share/cmake/Modules/FindPackageHandleStandardArgs.cmake:230 (message):
-- Configuring incomplete, errors occurred!
Could NOT find OpenMP_CXX (missing: OpenMP_CXX_FLAGS OpenMP_CXX_LIB_NAMES)
```
```
libomp 14.0.6 is already installed but outdated (so it will be upgraded).
==> Downloading https://ghcr.io/v2/homebrew/core/libomp/manifests/15.0.2
==> Downloading https://ghcr.io/v2/homebrew/core/libomp/blobs/sha256:a4e0796616d09221e2a486c95f9aa7c12d3c617e594b1d463a8f479bd4fa45c2
==> Downloading from https://pkg-containers.githubusercontent.com/ghcr1/blobs/sha256:a4e0796616d09221e2a486c95f9aa7c12d3c617e594b1d463a8f479bd4fa45c2?se=2022-10-13T20%3A50%3A00Z&sig=a68J671xfsvWip4TqYo5%2B2E0U5AswwCQDocHLNc6XXE%3D&sp=r&spr=https&sr=b&sv=2019-12-12
==> Upgrading libomp
14.0.6 -> 15.0.2
==> Pouring libomp--15.0.2.big_sur.bottle.tar.gz
==> Caveats
libomp is keg-only, which means it was not symlinked into /usr/local,
because it can override GCC headers and result in broken builds.
For compilers to find libomp you may need to set:
export LDFLAGS="-L/usr/local/opt/libomp/lib"
export CPPFLAGS="-I/usr/local/opt/libomp/include"
==> Summary
🍺 /usr/local/Cellar/libomp/15.0.2: 7 files, 1.7MB
==> Running `brew cleanup libomp`...
Disable this behaviour by setting HOMEBREW_NO_INSTALL_CLEANUP.
Hide these hints with HOMEBREW_NO_ENV_HINTS (see `man brew`).
Removing: /usr/local/Cellar/libomp/14.0.6... (7 files, 1.6MB)
Removing: /Users/runner/Library/Caches/Homebrew/libomp--14.0.6... (548.8KB)
```
|
|
* Correct particle positions outside refined injection
* Improve test
* Update benchmark
* Update Examples/Physics_applications/laser_acceleration/analysis_refined_injection.py
|
|
The current implementation only accepted parameters for ADIOS2
engines if the engine type was also specified. That is not
necessary, the default (bp4) will usually be taken and thus
the documentation did not work.
This fixes it, keeping the engine type optional in the inputs.
|
|
Add more instructions how to debug errors seen in *Release* mode.
|
|
|
|
|
|
* initial work to clean WarpX Utils
* remove AMRCore from Ionization tables
* progress
* refactoring of a part of the utils directory
* fix bug
* fixed bug
* fixed bug
* remove debug line accidentally slipped into the code
* remove debug line accidentally slipped into the code
* remove debug line accidentally slipped into the code
* cleaning
* fixed bug
|
|
* Fix synchronization of nodal points in subcycling
* Update checksum
|
|
* Updated 1D CI tests with particles
* Updated required version of yt to 4.1.0
* Fixed CI analysis of collisionXYZ
Fixed the fetching of the particle data
|
|
* Spack Desktop: Simpler Visualization
Reduce the dependencies for in situ vis, avoiding to
build DevilRay in Ascent, which takes very long right now.
We mostly use Ascent with VTK-m for now, so that works.
* CUDA Ascent & VTK-M: static
* ADIOS2: w/o CUDA
See https://github.com/ornladios/ADIOS2/issues/3332
|
|
* Doc: BLAS++/LAPACK++ on GitHub
BLAS++ and LAPACK++ development has moved to GitHub :tada:
Welcome! :)
* Update bitbucket mention to GH
* SLATE: Rename Missing Links
Co-authored-by: Mark Gates <mgates3@icl.utk.edu>
|
|
We forgot to set the shape (particle number) of constant
particle records. This violates the openPMD standard and
complicates post-processing.
|
|
Add values that are checked to the msg, e.g., I see:
```
z-slice in lab-frame (0.049998) is outside the buffer domain
physical extent (0.049998 to 0.050000).
```
|
|
Unbreak the RTD build. We need to do this all relative to the
root directory.
|
|
* Doc: Dev FAQ Pinned Memory
Since I spoke to two devs in 24hrs about the topic, let us document
it.
Co-authored-by: Ryan Sandberg <RSandberg@lbl.gov>
Co-authored-by: David Grote <dpgrote@lbl.gov>
|
|
* AMReX: Weekly Update
One commit after 22.10 release, to fix oneAPI DPC++ 2022.02:
22.10-1-g13aa4df0f5
* PICSAR: 22.10
* WarpX: 22.10
|
|
* Use parser for input parameters of type long
* Revert "Use parser for input parameters of type long"
This reverts commit 9573bb3f693f1247e77faa433fd96dc294e68361.
* Use parser for inputs of type long
* add safeCasttoLong function
* Fix typo in comment
|
|
* Include WarpX specific documentation
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Undo the precommit change of import in __init__.py
* Docs: Install Pure Python WarpX PICMI
Allows to use these classes in our Sphinx autodocs in the manual.
* Sphinx: Add Napoleon Extension
https://sphinxcontrib-napoleon.readthedocs.io
* PICMI Doc Strings: Fix Some Warnings
* Removed the DocumentedMetaClass since it was moved to picmistandard
* Defined the languate in Docs/source/conf.py
* Cleaned up minor issues in the rst files
* Updated existing doc strings in picmi.py
* Reformatted doc strings in Regression/Checksum/checksumAPI.py
* Reformatted doc strings in Regression/Checksum
* First set of WarpX specific documentation
* Updated to picmistandard version 0.0.20
* Cleaned up the Python document page
Fixed up the text to be more clear.
Added section headers for each picmi class (for easy reference).
* Further updates to picmi.py doc strings
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|