Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
* Fix NVCC Regression in Parser
Fix a regression with the recompilation of the parser that shows
up with NVCC and GNUmake on Summit.
* Modify Parser's makefile to remove #line and rename bison's output file to .cpp
Co-authored-by: Weiqun Zhang <weiqunzhang@lbl.gov>
|
|
* Remove a few segfaults when there is zero particle species
* Docs: Update Example
* We don't have `part_per_proc` so exemplify with something else.
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|
|
|
|
|
|
* Document for ascent replay
* ascent replated repository
* Apply suggestions from code review
* Review of replay - rendering plot without run simulation
* modify Review of replay - accept Axel suggestion
* modify replay - remove png file
* Remove PNG
Too large for git
* Replay: Restructure Workflow
* Remove duplicated files
Well documented in the LWFA example and text now.
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|
|
|
|
parser (#1448)
When compiling WarpX I got this warning message:
`wp_parser.lex.c:1356:17: warning: ‘void yyunput(int, char*)’ defined but not used [-Wunused-function]`
This can be fixed by adding `%option nounput` option to `wp_parser.l` and re-generating the parser with `make`.
|
|
Removes an outdated environment activation line in DPC++.
|
|
For HIP, AddNoRet for float is much faster because it has hardware support.
|
|
Add AMD HIP support to CMake and test it in CI (single-precision
build).
|
|
|
|
* replace wherever possible .reset(new with = make_unique
* fixed bug
* fixed bug
* revert WarpXOpenPMD.cpp to the original version
* removed another .reset(new
|
|
* New reduced diag: number of macroparticles
* Add parentheses in header line
* Apply suggestions from code review
Co-authored-by: Luca Fedeli <luca.fedeli.88@gmail.com>
* Add some consts
* Fuse some lines with std::ofstream
Co-authored-by: Luca Fedeli <luca.fedeli.88@gmail.com>
|
|
* turn off diagnostics
* document input to enable/disable diagnostics
* Update Docs/source/running_cpp/parameters.rst
|
|
|
|
|
|
* fix warning concerning comparison of int and unsigned int
* fix warning concerning unused variables
|
|
* write species flag to toggle particle output
* document write_species
* Update Source/Diagnostics/Diagnostics.cpp
* write species flag to toggle particle output
* document write_species
* Update Source/Diagnostics/Diagnostics.cpp
|
|
Try to update icc/icpc tests to oneAPI beta09.
|
|
(#1361)
Co-authored-by: Remi Lehe <remi.lehe@normalesup.org>
|
|
* Separate rho per species diagnostics from particle output
* Update input file of CI test dumping rho per species
* Update documentation
* Use m_output_species_names and m_all_species_names
* Use std::string::rfind and improve error message
* Rename m_all_species to m_output_species for consistency
|
|
|
|
|
|
Change old Travis-CI badge to new Azure Pipelines badge.
|
|
* added the phase argument for the gaussian laser
* removed signature
* added laser phase to picmi api
* added doc string for phi0
* Update parameters.rst
typo fix
Co-authored-by: Remi Lehe <remi.lehe@normalesup.org>
|
|
* RZ_SP fix
* RZ SP fix
* Fix cufft for single precision RZ.
Co-authored-by: Weiqun Zhang <weiqunzhang@lbl.gov>
|
|
* rocFFT support
* rocfft in 2d rz PSATD solver
|
|
|
|
* PICMI: read flag for laser continuous injection
* Simplify assignment of do_continuous_injection
* Update PICMI version
Co-authored-by: Remi Lehe <remi.lehe@normalesup.org>
|
|
Try to work-around the newly broken Nvidia apt repos for CUDA 11.0.
- Refs.: https://gitlab.com/nvidia/container-images/cuda/
- Nvidia bug report: 3144329
|
|
* Update inputs_3d
* Update inputs_2d
* Add ndt
* Doc
* Doc
* Update inputs_3d
* Update inputs_2d
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|
|
* replace some reset(new ) with make_unique
* fixed bug
|
|
This reflects an upstream change in AMReX' CMake
options.
|
|
|
|
* Fix warnings for 2D/3D PSATD builds
* Fix warnings for RZ PSATD builds
* Update FIXME comment: no known bug
* Fix remaining warning for RZ PSATD builds
|
|
* Made particles.use_fdtd_nci_corr false for RZ
* Added AMREX ALWAYS ASSERT
* Fixed AMREX ASSERT
* Revert "Fixed AMREX ASSERT"
This reverts commit 902fcaabbf0c01b75cd7b38e7fd198e75cf1703d.
* Revert "Added AMREX ALWAYS ASSERT"
This reverts commit 7c56270730d0e1e4c9a2ab705cc0027e7ecc1a79.
* Fixed confusion between commits
|
|
* Plot raw fields: fix bug for rho data
* Add query for <diag_name>.plot_raw_rho
* Check if MultiFabs are allocated and fix bug for coarse patch
* Add same check for raw F data
|
|
(#1409)
* Add clean error message when user makes a typo in fields to output
* Add a space
|
|
* fixed unused variable warning
* fixed unwanted copy
|
|
|
|
|
|
* [Draft] Doc: Lassen (LLNL)
Document installation and usage on Lassen (LLNL).
* [Draft] Doc: Quartz (LLNL)
Document installation and usage on Quartz (LLNL).
|
|
* Use fewer guard cells for deposition
* Update some CI benchmarks
* Revert "Update some CI benchmarks"
This reverts commit 2f40062557d75e3f6b2d3ba284332b8a92a0d404.
* Check if shapes and guard cells are compatible only for rho
* Do not allocate one extra guard cell for J
* Add more relaxed check on shapes and guard cells for J
|
|
* Increase size of sigma_star by 1 in nodal mode
* Add do_nodal to FabFactory<SigmaBox>
* Add ifdef WARPX_USE_PSATD directives
* Increase size of sigma_star arrays by one regardless of nodal state
|
|
Since our CI is cloud-based, we can get various kinds of
CPUs in different runs. For machine-precision tests, I got
the feeling that some tests that rely heavily on exact numbers
are overly strict with respect to guarantees between various
CPU generations, esp. since we rely on fastmath.
This will output the CPU type at the beginning of runs, so we
can find out more when tests fail with "checksum" (output data
fingerprint) errors.
|
|
|
|
* Update inputs_3d
* Apply suggestions from code review
Co-authored-by: Axel Huebl <axel.huebl@plasma.ninja>
|
|
|
|
|