Age | Commit message (Collapse) | Author | Files | Lines |
|
With the old E4S packages gone, we need to update out `Boost`
dependency.
|
|
also complains about .value(), which has an exception check
|
|
* ChargeDeposition: Remove Global WarpX Variable
Remove Global WarpX variable use in `ChargeDeposition`.
* CurrentDeposition: Add Missing Include
|
|
Give a clean error message if a `WarpX_thirdparty_src` directory
does not exist.
|
|
* add more readability checks to clang-tidy CI test
* reformat check list
* starting to address issues found with clang-tidy£
* addressing issues
* remove check
* address issues
* address all issues
* address issue found with github code scanning
|
|
SignalHandling::CheckSignals() (#4240)
|
|
* Update Crusher Docs and Scripts
The updated Crusher scripts and documentation were kept very close to the ones
for Frontier since they were found to work very well, currently.
The original Frontier docs were authored by Axel Huebl <axelhuebl@lbl.gov>.
* Directly ref to Frontier docs for system issues
Due to the two systems having a lot of similarities now we just reference
the section in the Frontier docs for the known system issues.
* [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>
|
|
Typos.
|
|
* Add Python bindings to extract the timestep
* More secure array access
|
|
Update slurm changes and pinning.
|
|
|
|
Congratulations!! :)
|
|
|
|
* Work-Around: Segfault in MPI_Init with HIP
See:
https://docs.olcf.ornl.gov/systems/crusher_quick_start_guide.html#olcfdev-1655-occasional-seg-fault-during-mpi-init
* Move to ABLASTR
All that counts is that HIP is initialized before GPU-aware MPI.
* Add Exception
|
|
In the spirit of a recent nvcc compiler bug, simplify the capture
logic from implicit to explicit for the particle filter lambda.
|
|
* Remove AddToMultiFabMap functions from WarpX class
* fixed bug
|
|
|
|
* Frontier (OLCF): MPI_Waitall issues
Document to use `export MPICH_SMP_SINGLE_COPY_MODE=NONE` in job scripts.
According to earlier attempts by Marco, this might not yet solve the issue,
but it is the documented workaround from the OLCF side so far, so we
should add it for now.
* Comments in LUMI Submission Template
|
|
* Doc: HPC Python Separate
Currently, due to global variables used in AMReX to store implicit
state instead of using handlers, we cannot build shared libraries
for Python (pyAMReX + pyWarpX + pyImpactX) with a *static*
`libamrex_Nd.a`. Thus, we automatically build a shared AMReX
library in our superbuild the moment that we need Python bindings.
That can confuse people that just use the C++ application with
AMReX inputs files, which then also depends on this shared AMReX
library (which also needs to be copied around, for instance).
This update the HPC docs to just build App and bindings twice
(twice the time spent), to work around this current limitation
of AMReX.
* Apply suggestions from code review
Co-authored-by: Remi Lehe <remi.lehe@normalesup.org>
* Apply suggestions from code review
Co-authored-by: Remi Lehe <remi.lehe@normalesup.org>
* Docs: Update other cluster descriptions as well.
* Install Dependencies: `git fetch --prune`
Fix an issue first seen in blaspp/lapackpp where fetch
failed because an unused remote branch was gone.
---------
Co-authored-by: Remi Lehe <remi.lehe@normalesup.org>
|
|
|
|
Add constant record components for position and positionOffset
in 1D3V and 2D3V simulations.
Follows our conventions in
https://warpx.readthedocs.io/en/latest/developers/dimensionality.html#conventions
and simplifies post-processing, e.g., emittance calculations.
|
|
* Fix: WarpX Version
Fix the WarpX version info.
* Fix Version: pyWarpX
* Version: More Python __version__
|
|
* CI: Shrink NVHPC Size
Try to shrink the size occupied in the NVHPC CI runs. Currently
running out-of-disk...
* NVHPC: Thin out Install
- remove profilers (big)
- remove static libs (big)
- remove examples (small)
|
|
|
|
|
|
* Fix Repeated Plasma Lens: Start
The current implementation has problems finding the right
index of the current plasma lens if:
- the lens started before z=0
- the lens did not start between z=0...first-period
This should fix it.
* Document Start, Define an End
|
|
|
|
|
|
* BTD RZ test: add missing checksum analysis
* [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>
|
|
* BoxArray in BTD has single box
* cleaning to remove all instances of max_box_size
* Assert BA size to be 1 for BTD and fix an inline comment
* fix typo
|
|
Accidentially now always built AMReX was shared library,
which is only needed if we build with Python.
|
|
|
|
* Remove unused/untested Python file
* Remove additional file
|
|
|
|
|
|
|
|
Because breathe is not yet ready.
|
|
Add an include guard to this header file.
|
|
|
|
Frontier uses a Low-noise Mode Layout for its nodes which pins system
processes via the default core specification `-S 8` to the first core
every L3 region.
|
|
|
|
(#4116)
* include current density in diagnostic output even if an electromagnetic solver is not used
* do not redeposit current for the magnetostatic solver
* update CI benchmarks for tests that previously did not register current density
* fix remaining failing CI test
* Apply suggestions from code review
Co-authored-by: Edoardo Zoni <59625522+EZoni@users.noreply.github.com>
* do not output current density in collision tests
* update `JFunctor` constructor doc-string
* code cleanup - reduce code duplication
* specify `Ex Ey Ez Bx By Bz` fields to plot for collision CI tests
* specify `Er Et Ez Br Bt Bz` as output for rz collision test
* rename `InterpolateToDst` to `InterpolateMFForDiag`
* only deposit current density once per diagnostic output (if not already deposited)
* write deposited current for all directions into `current_fp` during diagnostic step deposition
* use `amrex::make_alias` to directly deposit current density into `warpx.current_fp` during diagnostic step deposition
* add class variable `solver_deposits_current` to `FullDiagnostics` so that there is only one place where the determination is made whether current should be deposited during diagnostic output
* correct logic to determine `m_solver_deposits_current`
---------
Co-authored-by: Edoardo Zoni <59625522+EZoni@users.noreply.github.com>
|
|
* Fix: Remove Overly Exported Classes
Luckily, many of them should not contain global state.
* Add Missing WarpX.H Include
|
|
* added nodal electrostatic solver
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* sphere test for electrostatic nodal solver
* add 3d single bunch test
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* adds single bunch ci test
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* updated tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* updated checksums nodal electrostatic
* Improve readability
* Change `assert` condition
* Clean up
---------
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>
|
|
|
|
* Implement boosted frame automated test (1D)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Implement boosted frame automated test (1D)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* time_chunk_size=50
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Remi Lehe <remi.lehe@normalesup.org>
|
|
* Extensive clean up of PlasmaInjection
* Small cleanup of parseMomentum
* Fix string argument reference
* Remove unneeded surface_flux
|
|
|
|
* Update Python installation instructions on HPC
* Update Python installation instructions on HPC
---------
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|
|
|