commit 61734d3da08b88a5c35a7c57c20644e122fa5e36
Author: Pedro Costa
Date: Wed Sep 30 16:43:17 2020 +0000
typo fixes (#1427)
Docs/sphinx_documentation/source/AmrCore.rst
Docs/sphinx_documentation/source/AmrLevel.rst
Docs/sphinx_documentation/source/Basics.rst
Docs/sphinx_documentation/source/Fortran.rst
Docs/sphinx_documentation/source/GridCreation.rst
Docs/sphinx_documentation/source/IO.rst
Docs/sphinx_documentation/source/LinearSolvers.rst
Docs/sphinx_documentation/source/LoadBalancing.rst
commit c75c4a65ec50580504f87e6ea7cbd4d314644a17
Author: Weiqun Zhang
Date: Tue Sep 29 21:55:31 2020 -0700
Signed distance function (#1425)
## Summary
* Add FillImpFunc to fill MultiFab with the value of implicit function
* average_down_nodal: optional argument to specify the safety of mfiter
* Add FillSignedDistance to fill MultiFab with signed distance.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_MultiFabUtil.H
Src/EB/AMReX_EBFabFactory.H
Src/EB/AMReX_EB_utils.H
Src/EB/AMReX_EB_utils.cpp
commit ea9f8208cd820326cbd38d65307ae7f1181b48fb
Author: Andrew Myers
Date: Tue Sep 29 14:53:33 2020 -0700
Call sync at the end of the particle transformation functions. (#1424)
The `ParticleTransformations` test was broken on Tulip. The problem was missing `synchronize` statements in the test code itself. However, it is safer if these functions perform the synchronization internally.
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_ParticleTransformation.H
commit 23f2eb2dc97b56038d38b4079ee161800c8e40c2
Author: mic84
Date: Tue Sep 29 12:41:26 2020 -0700
Doc: update documentation on CMake+CUDA (#1421)
* Doc: update documentation on CMake+CUDA
* Update Docs/sphinx_documentation/source/GPU.rst
Co-authored-by: Axel Huebl
* Update Docs/sphinx_documentation/source/GPU.rst
Co-authored-by: Axel Huebl
* CMake: bump up minimum required CUDA version
* Doc: update doc
Co-authored-by: Axel Huebl
Docs/sphinx_documentation/source/BuildingAMReX.rst
Docs/sphinx_documentation/source/GPU.rst
Tools/CMake/AMReX_SetupCUDA.cmake
commit f9a441ae372198bf6efe3af6eaca3b967ee3291c
Author: Weiqun Zhang
Date: Tue Sep 29 09:16:03 2020 -0700
Fix 1D CUDA compilation failure due to #1332 (#1423)
The most vexing parse in C++ strikes again.
This closes #1422.
Src/Base/AMReX_GpuLaunchMacrosG.H
commit b7ef5a650d7e3accb9f698b7ec929c87cbf24711
Author: Weiqun Zhang
Date: Tue Sep 29 08:24:51 2020 -0700
Remove the call to set device flags. (#1417)
## Summary
This is no longer needed and may result in failure due to that the device
has already been set and initialized in another code.
## Additional background
Exa-Wind has a code that uses both Kokkos and AMReX. It seems that
the call to cudaSetDeviceFlags fails because of that.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_GpuDevice.cpp
commit ef6789d593814c78ce1de0456fb5fa524a1f9c13
Author: Andrew Myers
Date: Mon Sep 28 17:30:23 2020 -0700
Fix new bug in clearNeighhborsGPU (#1420)
* Fix new bug in clearNeighhborsGPU
* fix typo
Src/Particle/AMReX_NeighborParticlesGPUImpl.H
commit c4116b67f7f176229a749bc818b7610a61a9a2f8
Author: Andrew Myers
Date: Mon Sep 28 14:17:55 2020 -0700
these functions need to be marked __host__ __device__ to compile with HIP (#1419)
Src/Particle/AMReX_ParticleUtil.H
commit e4f0bcd12845e81daa8c13d9839e9662c26bfd08
Author: Andrew Myers
Date: Mon Sep 28 13:11:30 2020 -0700
Define ParticleTile in AddParticles (#1409)
* Remove GetPosition and SetPosition from ParIter
* define the particle tile before adding particles to it
Src/Particle/AMReX_ParticleContainerI.H
commit aadf8eb8711b816d3234745a98c6170918bca7c3
Author: Andrew Myers
Date: Mon Sep 28 13:09:33 2020 -0700
Generalize clearNeighbors for multiple levels and remove faulty assertions. (#1418)
* Generalize clearNeighbors for multiple levels and remove faulty assertions.
* also fix cpu version
Src/Particle/AMReX_NeighborParticlesGPUImpl.H
Src/Particle/AMReX_NeighborParticlesI.H
commit 3a9381141a1fe78180be578e7b0e1b24a7532171
Author: Andrew Myers
Date: Fri Sep 25 14:15:18 2020 -0700
fix decltype for the case where have a random engine (#1414)
Src/Particle/AMReX_ParticleTransformation.H
commit d7ed0d7a41d92ecfb0a4cee2ff023597e5e9f875
Author: Andrew Myers
Date: Fri Sep 25 14:15:09 2020 -0700
always clear neighbor particles before calling Redistribute() (#1415)
Src/Particle/AMReX_NeighborParticles.H
commit 4a4d585621b0fa34980464baab0466b5e09bc781
Author: asalmgren
Date: Fri Sep 25 09:50:37 2020 -0700
1) remove flags setCGVerbose and setCGMaxIter since those are more accurately called (#1413)
setBottomVerbose and setBottomMaxIter. The "setCG..." flags are misleading since
they actually apply to non-CG bottom solvers as well.
2) add documentation in the Linear Solvers section about "maxorder" -- this was not previously documented.
Docs/sphinx_documentation/source/LinearSolvers.rst
Src/F_Interfaces/LinearSolvers/AMReX_multigrid_fi.cpp
Src/F_Interfaces/LinearSolvers/AMReX_multigrid_mod.F90
Src/LinearSolvers/MLMG/AMReX_MLMG.H
Tests/LinearSolvers/MLMG/inputs
Tests/LinearSolvers/MLMG/inputs.boxes
Tests/LinearSolvers/MLMG/solve_with_mlmg.cpp
Tutorials/Basic/HeatEquation_EX3_C/Source/advance.cpp
Tutorials/LinearSolvers/ABecLaplacian_F/inputs
Tutorials/LinearSolvers/ABecLaplacian_F/inputs-rt-abeclap-lev
Tutorials/LinearSolvers/ABecLaplacian_F/inputs-rt-poisson-com
Tutorials/LinearSolvers/ABecLaplacian_F/mytest.F90
Tutorials/LinearSolvers/MAC_Projection_EB/inputs_3d
Tutorials/LinearSolvers/MAC_Projection_EB/main.cpp
Tutorials/LinearSolvers/MultiComponent/inputs
Tutorials/LinearSolvers/MultiComponent/main.cpp
Tutorials/LinearSolvers/Nodal_Projection_EB/inputs_3d
Tutorials/LinearSolvers/Nodal_Projection_EB/main.cpp
Tutorials/SDC/MISDC_ADR_2d/Source/main.cpp
commit bd7c2b8db69b4a3cdb40bf3f23bf8eea952d5071
Author: Pedro Costa
Date: Fri Sep 25 16:13:23 2020 +0000
typo fixes (#1411)
## Summary
some small typos I found while reading the documentation
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Co-authored-by: Weiqun Zhang
Docs/sphinx_documentation/source/Basics.rst
commit 9499db8a1f406d1691b845f203c3a15c5f28c744
Author: Andrew Myers
Date: Fri Sep 25 08:43:04 2020 -0700
Remove GetPosition and SetPosition from ParIter (#1408)
These were added for WarpX, but they are not used any more.
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_ParIter.H
commit 9f43ecbb0df8e37e4d0aa9511f76cc5e91c52d9d
Author: Andrew Myers
Date: Thu Sep 24 13:58:32 2020 -0700
filterParticles and filterAndTransformParticles need to take callables with and without RNG (#1407)
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_ParticleTransformation.H
Src/Particle/AMReX_ParticleUtil.H
Src/Particle/AMReX_WriteBinaryParticleData.H
commit 605e276d42d660eaf28e8a7f68636d30f624738f
Author: Axel Huebl
Date: Thu Sep 24 08:10:29 2020 -0700
ParallelDescriptor: Check MPI If Self (#1406)
## Summary
Only check the requested MPI Threading level if AMReX also initialized MPI.
Follow-up to #1351
## Additional background
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_ParallelDescriptor.cpp
commit 3cad68f7a504044186d0caa20649197d2794aa8b
Author: Donald E. Willcox
Date: Wed Sep 23 18:02:34 2020 -0700
Enable compilation on Cori interactive nodes (#1405)
## Summary
This enables compilation in Cori interactive jobs.
Previously, in interactive jobs on Cori Haswell and KNL, `hostname -f` returned a string like `nid00018` so we were loading `Make.unknown`, causing compilation failure.
@WeiqunZhang suggested checking the `NERSC_HOST` environment variable instead, which is defined on Cori login, interactive (Haswell & KNL), and GPU nodes.
This PR moves the check in `Make.machines` for Cori to follow the logic for Cori GPU, and we check if `NERSC_HOST` is set to `cori`.
I've tested this works on Haswell & KNL interactive jobs and that we are still setting `which_computer` correctly on Cori GPU.
## Additional background
## Checklist
The proposed changes:
- [X] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/Make.machines
commit 2b31386adec5efda010ff08261d5f0658027e96c
Author: Weiqun Zhang
Date: Wed Sep 23 17:23:48 2020 -0700
DPCPP: use C++17 and disable mkl warnings (#1404)
* DPCPP: use C++17 and disable mkl warnings
* Explicitly use C++17 because MKL contains C++17 codes.
Co-authored-by: Axel Huebl
Src/Base/AMReX_Random.H
Src/Base/AMReX_RandomEngine.H
Tools/GNUMake/comps/dpcpp.mak
commit 766a2385cf37e97291f7e0d4fa5ebe1ed55f7fe5
Author: Axel Huebl
Date: Wed Sep 23 16:32:19 2020 -0700
Docs: C++17 & Particle SP (#1403)
Update the build docs for GNUmake:
- single precision for particles
- typo in `USE_DPCPP` option
- add C++17 option
Equivalent CMake options are up-to-date, so no change is needed.
Docs/sphinx_documentation/source/BuildingAMReX.rst
commit fbee3e3a3b1ebd912dbe6c1b6bb29d36d10ab153
Author: mic84
Date: Wed Sep 23 16:16:11 2020 -0700
CMake: enforce C++17 as minimum C++ standard for ENABLE_DPCPP=ON (#1402)
Tools/CMake/AMReX_Config.cmake
commit 577c8221804b5a2185c0356d6ab46e538722b9d1
Author: Weiqun Zhang
Date: Wed Sep 23 14:47:56 2020 -0700
Use RNG in Particle I/O filter (#1401)
In order to support the use of RNG in particle I/O filters, we need to be
able to call two types of filter functions. The new type has a
RandomEngine argument.
Src/Particle/AMReX_WriteBinaryParticleData.H
commit e9007d6832efe63e94ff180d97f2212cad9e13ac
Author: Pedro Costa
Date: Wed Sep 23 17:58:05 2020 +0000
nitpicking :) (#1400)
## Summary
typo fix
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Docs/sphinx_documentation/source/Basics.rst
commit 5f85756e8523423913bd46a7d318def6f1889103
Author: Weiqun Zhang
Date: Wed Sep 23 10:31:10 2020 -0700
Update dpcpp flags (#1399)
* -fsycl-unnamed-lambda is enabled by default now. So we do not need to
pass the flag anymore.
* Early optimizations are enabled by default in beta09. This makes a lot of
tests crash. So we disable it with -fno-sycl-early-optimizations.
Tools/CMake/AMReXSYCL.cmake
Tools/GNUMake/comps/dpcpp.mak
commit f75ae5897d699a0fa4835818b5d5ab83535e5be8
Author: mic84
Date: Wed Sep 23 08:09:22 2020 -0700
NodalProjector: fix comments (#1396)
Src/LinearSolvers/Projections/AMReX_NodalProjector.H
Src/LinearSolvers/Projections/AMReX_NodalProjector.cpp
commit 804f844b0634e5fde9fd52a36c12fcb3069fcc66
Author: kngott <30483578+kngott@users.noreply.github.com>
Date: Tue Sep 22 15:11:36 2020 -0700
Fix inaccurate assertion. (#1395)
Src/Base/AMReX_BoxList.cpp
commit a18b88f27ca0b5c69cd935d7af3011d954322d9a
Author: mic84
Date: Tue Sep 22 13:47:12 2020 -0700
CMake: remove LinearSolvers/MultiComponent from tutorials (#1394)
Tutorials/LinearSolvers/MultiComponent/CMakeLists.txt
commit 784f23144d3edda49ff812e6c601ca3ec1468372
Author: Weiqun Zhang
Date: Tue Sep 22 13:33:25 2020 -0700
Fix function type for HIP (#1393)
Currently in HIP we do not have a way to do SNFINAE based on whether the
function is callable on host or not. So we have to replace the device
lambda with host and device lambda.
Src/Base/AMReX_MultiFab.cpp
commit 9198b1b9a7f7f77ff04261aada6284301db77391
Author: Weiqun Zhang
Date: Mon Sep 21 21:08:18 2020 -0700
DPCPP support for mulitple GPUs (#1392)
Src/Base/AMReX_GpuDevice.cpp
commit d6eadf8c32ae211febeaa2aa1e6e1df5a70807cb
Author: Weiqun Zhang
Date: Mon Sep 21 20:21:53 2020 -0700
Intall intel-oneapi-mkl-devel in CI (#1391)
.github/workflows/dependencies/dependencies_dpcpp.sh
commit 9a272a1701dbedd1f0b993720f157cf340d12f33
Author: Weiqun Zhang
Date: Mon Sep 21 13:52:33 2020 -0700
Random Number Generation on Device (#1363)
In order to support RNG on device with DPC++, we have to change the API
because DPC++ does not support global device variables. We introduce a new
ParallelForRNG function that takes a device lambda with an extra
RandomEngine argument. Random numbers can be generated by calling
amrex::Random(RandomEngine const&). To maintain backward compatibility in
the short term, the old way of calling `amrex::Random()` on device is kept.
But it will be removed soon because no locking is needed in the new way
unlike in `Random()`. Note that only the device version of
`amrex::Random()` is deprecated.
Src/Base/AMReX_Extension.H
Src/Base/AMReX_GpuDevice.H
Src/Base/AMReX_GpuDevice.cpp
Src/Base/AMReX_GpuLaunch.H
Src/Base/AMReX_GpuLaunchFunctsC.H
Src/Base/AMReX_GpuLaunchFunctsG.H
Src/Base/AMReX_Random.H
Src/Base/AMReX_Random.cpp
Src/Base/AMReX_RandomEngine.H
Src/Base/AMReX_Reduce.H
Src/Base/CMakeLists.txt
Src/Base/Make.package
Tests/AsyncOut/multifab/main.cpp
Tests/GPU/RandomNumberGeneration/inputs
Tests/GPU/RandomNumberGeneration/main.cpp
Tests/Particles/ParallelContext/main.cpp
Tests/Particles/Redistribute/main.cpp
Tutorials/GPU/ParallelReduce/main.cpp
commit 3c8dad86006c5ef256884733bff99b9806be46d4
Author: Andrew Myers
Date: Mon Sep 21 13:51:30 2020 -0700
Remove redundant particle test (#1389)
The same functionality is already tested in `Tests/Particles/Redistribute`.
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tests/Particles/GNUmakefile
Tests/Particles/Make.package
Tests/Particles/main.cpp
Tests/Particles/test.py
commit 1e35d0d50287bc32a02a94b03466a750c9e491d1
Author: Axel Huebl
Date: Mon Sep 21 10:24:23 2020 -0700
CMake & DPC++: Avoid Flag Deduplication (#1388)
## Summary
Avoid deduplication of `-mlong-double-64 -Xclang -mlong-double-64` into `-mlong-double-64 -Xclang`.
Update the apt package name, which was changed by upstream for beta09.
Refs.:
- https://gitlab.kitware.com/cmake/cmake/-/issues/15826
- https://gitlab.kitware.com/cmake/cmake/-/merge_requests/1841
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
.github/workflows/dependencies/dependencies_dpcpp.sh
Tools/CMake/AMReXSYCL.cmake
commit 1cec808bc2f814033350b09f01f4b64a9112f2cd
Author: Weiqun Zhang
Date: Mon Sep 21 08:33:10 2020 -0700
More on Gpu kernel fusing (#1332)
## Summary
* Add Gpu::KernelInfo argument to ParallelFor to allow the user to indicate
whether the kernel is an candidate for fusing.
* For MFIter, if the local size is less or equal to 3, the fuse region is
turned on and small kernels marked fusable will be fused.
* Add launch macros for fusing.
* Add fusing to a number of functions used by linear solvers. Note that
there are a lot more amrex functions need to be updated for fusing.
* Optimize reduction for bottom solve.
* Consolidate memcpy in communication functions.
* Option to use device memory in communication kernels for packing and
unpacking buffers. But it's currently turned off because the performance
was not improved in testing. In fact, it was worse than using pinned
memory. But this might change in the future. So the option is kept.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX.cpp
Src/Base/AMReX_FBI.H
Src/Base/AMReX_FabArray.H
Src/Base/AMReX_FabArrayBase.cpp
Src/Base/AMReX_FabArrayCommI.H
Src/Base/AMReX_FabArrayUtility.H
Src/Base/AMReX_Gpu.H
Src/Base/AMReX_GpuFuse.H
Src/Base/AMReX_GpuFuse.cpp
Src/Base/AMReX_GpuKernelInfo.H
Src/Base/AMReX_GpuLaunch.H
Src/Base/AMReX_GpuLaunchFunctsC.H
Src/Base/AMReX_GpuLaunchFunctsG.H
Src/Base/AMReX_GpuLaunchMacrosG.H
Src/Base/AMReX_MFIter.H
Src/Base/AMReX_MFIter.cpp
Src/Base/AMReX_MultiFab.cpp
Src/Base/AMReX_MultiFabUtil.cpp
Src/Base/AMReX_Reduce.H
Src/Base/CMakeLists.txt
Src/Base/Make.package
Src/LinearSolvers/MLMG/AMReX_MLABecLaplacian.cpp
Src/LinearSolvers/MLMG/AMReX_MLCellLinOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLPoisson.cpp
commit 626b5f55c1c3d143e918f5e5e95cb231e497e896
Author: Weiqun Zhang
Date: Mon Sep 21 08:29:39 2020 -0700
Make max GPU streams runtime parameter (#1386)
## Summary
Add a new runtime parameter, amrex.max_gpu_streams.
## Additional background
This can be used to work around a HIP compiler bug by setting `amrex.max_gpu_streams=1`.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_GpuDevice.H
Src/Base/AMReX_GpuDevice.cpp
commit 50e3994bb3cb3d4fdd1a680ae10176bc99dcd4d5
Author: Weiqun Zhang
Date: Sun Sep 20 13:40:05 2020 -0700
Switch to `use mpi` from `include 'mpif.h'` (#1385)
Src/F_Interfaces/Base/AMReX_parallel_mod.F90
commit 816ef0b6d37580fc86c5c1119ecdfa7288cf83e5
Author: Weiqun Zhang
Date: Sun Sep 20 13:31:59 2020 -0700
Fix a typo in Advection_F make file (#1384)
## Summary
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tutorials/FortranInterface/Advection_F/Source/Src_3d/Make.package
commit 35a0533f8726e3741a5bc7cadc68d9e6926a6893
Author: Weiqun Zhang
Date: Sun Sep 20 13:31:24 2020 -0700
Fix tensor solver with periodic boundaries (#1383)
## Summary
Remove erroneous calls to EnforcePeriodicity. When coarse/fine boundary
meets periodic boundary with fine grids form an L-shape, the uncovered fine
ghost cell at the corner is actually "multi-valued". That is the cells with
the same indices has different values in different boxes. That's why
calling EnforcePeriodicity is erroneous. For CPU runs, this bug introduces
deterministic errors, but the solver still converges. For GPU runs, this
causes convergence issues too because EnforcePeriodicity is not
deterministic when the values inside domain are not consistent.
## Additional background
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [x] changes answers in the test suite to more than roundoff level
- [x] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/LinearSolvers/MLMG/AMReX_MLEBTensorOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLTensorOp.cpp
commit 7c5cce3687a5b663355b1a367f0734760dcb4a0b
Author: Andrew Myers
Date: Fri Sep 18 17:15:53 2020 -0700
Remove old / obsolete / non-test tests. (#1381)
Tests/Algoim/CMakeLists.txt
Tests/Algoim/GNUmakefile
Tests/Algoim/Make.package
Tests/Algoim/main.cpp
Tests/BBIOBenchmark/BBIOTest.cpp
Tests/BBIOBenchmark/BBIOTestDriver.cpp
Tests/BBIOBenchmark/CMakeLists.txt
Tests/BBIOBenchmark/GNUmakefile
Tests/BBIOBenchmark/README
Tests/BaseFabTesting/CMakeLists.txt
Tests/BaseFabTesting/GNUmakefile
Tests/BaseFabTesting/Make.package
Tests/BaseFabTesting/inputs
Tests/BaseFabTesting/main.cpp
Tests/C_BaseLib/AMRPROFTEST_F.H
Tests/C_BaseLib/AMRProfTestBL.cpp
Tests/C_BaseLib/CMakeLists.txt
Tests/C_BaseLib/GNUmakefile
Tests/C_BaseLib/TPROFILER.F
Tests/C_BaseLib/TPROFILER_F.H
Tests/C_BaseLib/fillfab.f
Tests/C_BaseLib/mt19937int.out
Tests/C_BaseLib/t8BIT.cpp
Tests/C_BaseLib/tBA.cpp
Tests/C_BaseLib/tCArena.cpp
Tests/C_BaseLib/tDM.cpp
Tests/C_BaseLib/tDir.cpp
Tests/C_BaseLib/tFAC.cpp
Tests/C_BaseLib/tFB.cpp
Tests/C_BaseLib/tFillFab.cpp
Tests/C_BaseLib/tMF.cpp
Tests/C_BaseLib/tMFcopy.cpp
Tests/C_BaseLib/tParmParse.cpp
Tests/C_BaseLib/tProfiler.cpp
Tests/C_BaseLib/tRABcast.cpp
Tests/C_BaseLib/tRan.cpp
Tests/C_BaseLib/tUMap.cpp
Tests/C_BaseLib/tVisMF.cpp
Tests/C_BaseLib/tVisMF2.cpp
Tests/C_BaseLib/tread.cpp
Tests/DataServicesTest0/CMakeLists.txt
Tests/DataServicesTest0/DataServicesTest0.cpp
Tests/DataServicesTest0/GNUmakefile
Tests/GPU/CudaGraphs/BuildingGraphs/GNUmakefile
Tests/GPU/CudaGraphs/BuildingGraphs/Make.package
Tests/GPU/CudaGraphs/BuildingGraphs/inputs_3d
Tests/GPU/CudaGraphs/BuildingGraphs/main.cpp
Tests/GPU/CudaGraphs/BuildingGraphs/run.corigpu
Tests/GPU/CudaGraphs/CrazyGraphs/GNUmakefile
Tests/GPU/CudaGraphs/CrazyGraphs/main.cpp
Tests/GPU/CudaGraphs/CrazyGraphs/run.corigpu
Tests/GPU/CudaGraphs/GraphBoundary/GNUmakefile
Tests/GPU/CudaGraphs/GraphBoundary/Make.package
Tests/GPU/CudaGraphs/GraphBoundary/Prob.H
Tests/GPU/CudaGraphs/GraphBoundary/inputs_3d
Tests/GPU/CudaGraphs/GraphBoundary/main.cpp
Tests/GPU/CudaGraphs/GraphBoundary/profile.sh
Tests/GPU/CudaGraphs/GraphBoundary/run.corigpu
Tests/GPU/CudaGraphs/GraphBoundary/run.summit
Tests/GPU/CudaGraphs/GraphInitTest/GNUmakefile
Tests/GPU/CudaGraphs/GraphInitTest/Make.package
Tests/GPU/CudaGraphs/GraphInitTest/inputs_3d
Tests/GPU/CudaGraphs/GraphInitTest/main.cpp
Tests/GPU/CudaGraphs/GraphReuseCopy/GNUmakefile
Tests/GPU/CudaGraphs/GraphReuseCopy/Make.package
Tests/GPU/CudaGraphs/GraphReuseCopy/inputs_3d
Tests/GPU/CudaGraphs/GraphReuseCopy/main.cpp
Tests/GPU/CudaGraphs/GraphReuseCopy/run.corigpu
Tests/GPU/CudaGraphs/GraphWithMemcpy/GNUmakefile
Tests/GPU/CudaGraphs/GraphWithMemcpy/Make.package
Tests/GPU/CudaGraphs/GraphWithMemcpy/inputs_3d
Tests/GPU/CudaGraphs/GraphWithMemcpy/main.cpp
Tests/GPU/CudaGraphs/GraphWithMemcpy/run.corigpu
Tests/GPU/CudaGraphs/Readme.txt
Tests/GPU/CuptiTest/Exec/CUDA/GNUmakefile
Tests/GPU/CuptiTest/Exec/CUDA/Make.package
Tests/GPU/CuptiTest/Exec/CUDA/myfunc.H
Tests/GPU/CuptiTest/Exec/CUDA/myfunc.cpp
Tests/GPU/CuptiTest/Exec/CUDA/mykernel.H
Tests/GPU/CuptiTest/Make.CUPTI
Tests/GPU/CuptiTest/Source/Make.package
Tests/GPU/CuptiTest/Source/main.cpp
Tests/GPU/Locking/GNUmakefile
Tests/GPU/Locking/Make.package
Tests/GPU/Locking/inputs
Tests/GPU/Locking/main.cpp
Tests/GPU/libamrex_CUDA/GNUmakefile
Tests/GPU/libamrex_CUDA/README.md
Tests/GPU/libamrex_CUDA/main.cpp
Tests/IOBenchmark/GNUmakefile
Tests/IOBenchmark/IOTest.cpp
Tests/IOBenchmark/IOTestDriver.cpp
Tests/IOBenchmark/README
Tests/IOBenchmark/inputs
Tests/IOBenchmark/inputs.dirname
Tests/IOBenchmark/inputs.dss
Tests/IOBenchmark/inputs.dssmf
Tests/IOBenchmark/inputs.nft
Tests/IOBenchmark/inputs.small
Tests/MKDir/GNUmakefile
Tests/MKDir/MKDir.cpp
Tests/NoFort/GNUmakefile
Tests/NoFort/Make.package
Tests/NoFort/main.cpp
Tests/PnetCDFBenchmark/GNUmakefile
Tests/PnetCDFBenchmark/Make.package
Tests/PnetCDFBenchmark/ReadMe
Tests/PnetCDFBenchmark/WritePlotfilePnetCDF.H
Tests/PnetCDFBenchmark/WritePlotfilePnetCDF.cpp
Tests/PnetCDFBenchmark/inputs
Tests/PnetCDFBenchmark/main.cpp
Tests/ProfTests/HeatEquation_EX1_C/Exec/GNUmakefile
Tests/ProfTests/HeatEquation_EX1_C/Exec/inputs_2d
Tests/ProfTests/HeatEquation_EX1_C/Exec/inputs_3d
Tests/ProfTests/HeatEquation_EX1_C/Source/Make.package
Tests/ProfTests/HeatEquation_EX1_C/Source/advance.cpp
Tests/ProfTests/HeatEquation_EX1_C/Source/advance_2d.f90
Tests/ProfTests/HeatEquation_EX1_C/Source/advance_3d.f90
Tests/ProfTests/HeatEquation_EX1_C/Source/init_phi_2d.f90
Tests/ProfTests/HeatEquation_EX1_C/Source/init_phi_3d.f90
Tests/ProfTests/HeatEquation_EX1_C/Source/main.cpp
Tests/ProfTests/HeatEquation_EX1_C/Source/myfunc.H
Tests/ProfTests/HeatEquation_EX1_C/Source/myfunc_F.H
Tests/ProfTests/ThirdParty/Exec/GNUmakefile
Tests/ProfTests/ThirdParty/Exec/inputs_2d
Tests/ProfTests/ThirdParty/Exec/inputs_3d
Tests/ProfTests/ThirdParty/Source/Make.package
Tests/ProfTests/ThirdParty/Source/advance.cpp
Tests/ProfTests/ThirdParty/Source/advance_2d.f90
Tests/ProfTests/ThirdParty/Source/advance_3d.f90
Tests/ProfTests/ThirdParty/Source/init_phi_2d.f90
Tests/ProfTests/ThirdParty/Source/init_phi_3d.f90
Tests/ProfTests/ThirdParty/Source/main.cpp
Tests/ProfTests/ThirdParty/Source/myfunc.H
Tests/ProfTests/ThirdParty/Source/myfunc_F.H
Tests/SinglePrecision/GNUmakefile
Tests/SinglePrecision/Make.package
Tests/SinglePrecision/main.cpp
Tests/Slice/GNUmakefile
Tests/Slice/Make.package
Tests/Slice/inputs
Tests/Slice/main.H
Tests/Slice/main.cpp
Tests/SliceWithInterp/inputs
Tests/SliceWithInterp/main.cpp
Tests/Stream/GNUmakefile
Tests/Stream/Make.package
Tests/Stream/inputs
Tests/Stream/main.cpp
Tests/ThirdPartyLib/GNUmakefile
Tests/ThirdPartyLib/Readme
Tests/ThirdPartyLib/bar.F90
Tests/ThirdPartyLib/foo.cpp
Tests/ThirdPartyLib/main.c
Tests/TypeCheck/GNUmakefile
Tests/TypeCheck/Make.package
Tests/TypeCheck/f.f90
Tests/TypeCheck/f_F.H
Tests/Vectorization/GNUmakefile
Tests/Vectorization/Make.package
Tests/Vectorization/kc.H
Tests/Vectorization/kc.cpp
Tests/Vectorization/kdecl.H
Tests/Vectorization/kf.F90
Tests/Vectorization/main.cpp
commit 28a454d7ea1f696dd721834b9b4e668294bf50b5
Author: mic84
Date: Fri Sep 18 16:23:04 2020 -0700
CMake: enable testing via CTest. (#1362)
.github/workflows/linux.yml
CMakeLists.txt
Tests/Algoim/CMakeLists.txt
Tests/AsyncOut/multifab/CMakeLists.txt
Tests/BBIOBenchmark/CMakeLists.txt
Tests/BaseFabTesting/CMakeLists.txt
Tests/CMakeLists.txt
Tests/C_BaseLib/CMakeLists.txt
Tests/DataServicesTest0/CMakeLists.txt
Tests/FillBoundaryComparison/CMakeLists.txt
Tests/Particles/NeighborParticles/CMakeLists.txt
Tests/Particles/ParticleMesh/CMakeLists.txt
Tests/Particles/ParticleReduce/CMakeLists.txt
Tests/Particles/ParticleTransformations/CMakeLists.txt
Tests/Particles/Redistribute/CMakeLists.txt
Tests/complementIn/CMakeLists.txt
commit 9831ee93eb844ce43a37c8b3af138cc57e893964
Author: Andrew Myers
Date: Fri Sep 18 15:58:59 2020 -0700
Remove reliance on managed memory from AMReX_ParticleCommunication.H/.cpp (#1380)
Src/Particle/AMReX_ParticleCommunication.H
Src/Particle/AMReX_ParticleCommunication.cpp
commit eb2b7fd5419a4b2e89d32c6ec0ac8e7bfd59df69
Author: Andrew Myers
Date: Fri Sep 18 10:25:48 2020 -0700
fix signed vs. unsigned issue in AMReX_ParticleCommunication.cpp (#1379)
Src/Particle/AMReX_ParticleCommunication.cpp
commit 6fbd2868a38396391260c0ac61f23c5eac5d6d6b
Author: Weiqun Zhang
Date: Fri Sep 18 07:24:18 2020 -0700
Fix #1375 (#1378)
A Bug was introduced in #1375. The host vector should have an initial size.
Src/Particle/AMReX_ParticleTile.H
commit 6ce92531fd1fb5fce70506ddd214aacf0db06501
Author: Weiqun Zhang
Date: Thu Sep 17 19:09:33 2020 -0700
DPC++ & HIP: NumberOfParticles on device (#1377)
## Summary
For DPC++ and HIP, we have to run NumberOfParticles on device.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_ParticleContainerI.H
commit e4023270d871f6850f9fe8ff744745e7d74fdaac
Author: Weiqun Zhang
Date: Thu Sep 17 18:30:12 2020 -0700
Put static constexpr members back to ParticleTile (#1376)
## Summary
They should not be removed in #1375.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_ParticleTile.H
commit 4b2f5fa1fc7d13c150a41f74ff33b5fbe4395bdf
Author: Weiqun Zhang
Date: Thu Sep 17 17:21:44 2020 -0700
Avoid accessing device memory directly on host (#1375)
## Summary
ParticleLocator and ParticleTile contain vectors using device memory. To
remove the dependency on unified memory we avoid accessing them directly on
the host.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_ParticleLocator.H
Src/Particle/AMReX_ParticleReduce.H
Src/Particle/AMReX_ParticleTile.H
commit 39f32bd4f9f6d9f0a9455f7f1790337a780a1e09
Author: Andrew Myers
Date: Thu Sep 17 11:34:21 2020 -0700
Add explicit syncs to InitParticles in the Test routines. (#1374)
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tests/Particles/AsyncIO/main.cpp
Tests/Particles/NeighborParticles/MDParticleContainer.cpp
Tests/Particles/ParallelContext/main.cpp
Tests/Particles/ParticleReduce/main.cpp
Tests/Particles/ParticleTransformations/main.cpp
Tests/Particles/Redistribute/main.cpp
commit 3cf2461f97ba66b8da5d62cacf35b43ab6584b67
Author: Weiqun Zhang
Date: Thu Sep 17 09:48:48 2020 -0700
User defined literals in DPC++ (#1366)
It is now supported in beta09 with `-Xclang -mlong-double-64`.
Src/Base/AMReX_REAL.H
Tools/CMake/AMReXSYCL.cmake
Tools/GNUMake/comps/dpcpp.mak
commit 5a2d460801039b79a9d90ac2bf116c186d1220f9
Author: mic84
Date: Wed Sep 16 19:02:22 2020 -0700
GNUMake: remove unused define for HIP compilation (#1372)
* GNUMake: remove unused AMREX_HIP_PLATFORM define
* GNUMake: rename HIP_PLATFORM to HIP_COMPILER for consistency
Tools/GNUMake/Make.defs
Tools/GNUMake/comps/hip.mak
commit 59036167c676ae1acc31b18e061f99573af36111
Author: Axel Huebl
Date: Wed Sep 16 17:09:11 2020 -0700
.gitignore: build/ (#1373)
## Summary
besides our currently assumed temporary build directories, a temporar build/ directory is quite common in CMake (although discouraged to build inside the tree at all, but this is safe enough).
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
.gitignore
commit ea27c4008901d88fa41af08592bbd2355f23f569
Author: Mark Meredith
Date: Wed Sep 16 17:38:46 2020 -0400
Fix uninitialized compiler warnings for: xslope yslope zslope (#1371)
## Summary
Getting compiler warnings when building MFiX-Exa file that includes `AMReX_EB_slopes_K.H`
## Additional background
Building with GCC 7.5.0:
```
../src/convection/mfix_slopes.cpp: In member function 'void mfix::mfix_compute_slopes(int, amrex::Real, amrex::MultiFab&, const amrex::Vector&, const amrex::Vector&, const amrex::Vector&, int, std::map, amrex::PODVector > >&)':
../src/convection/mfix_slopes.cpp:235:52: error: 'zslope' may be used uninitialized in this function [-Werror=maybe-uninitialized]
zs_fab(i,j,k,slopes_comp+n) = eb_slopes[2];
../src/convection/mfix_slopes.cpp:216:52: error: 'yslope' may be used uninitialized in this function [-Werror=maybe-uninitialized]
ys_fab(i,j,k,slopes_comp+n) = eb_slopes[1];
../src/convection/mfix_slopes.cpp:197:52: error: 'xslope' may be used uninitialized in this function [-Werror=maybe-uninitialized]
xs_fab(i,j,k,slopes_comp+n) = eb_slopes[0];
```
## Checklist
The proposed changes:
- [X] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Co-authored-by: Mark Meredith
Src/EB/AMReX_EB_slopes_K.H
commit 7aacbd10c8b6b0cf463d04d3418783587090467f
Author: mic84
Date: Tue Sep 15 17:41:31 2020 -0700
CMake: remove duplicate sources (#1370)
Src/Base/CMakeLists.txt
commit 940a9499bb7c5c4f1aa209d4e501d34e20eeafff
Author: mic84
Date: Tue Sep 15 14:50:59 2020 -0700
CMake: re-organize tutorials (#1333)
* CMake: begin refactoring tutorials
* CMake: refactor Tutorials/Basic
* CMake: setup_tutorials is now able to figure out include paths by itself
* CMake: add optional arguments ot setup_tutorial
* CMake: refactor Tutorials/Amr
* CMake: refactor Tutorials/LinearSolvers
* CMake: add tutorials in Tutorials/EB
* Tutorials: change name of file for consistency with 2D case
* CMake: refactor Tutorials/FortranInterface
* CMake: refactor Tutorials/Particles
* CMake: forgot to refactor Tutorials/LinearSolvers/NodeTensorLap/
* CMake: update CI
* CMake: some clean-up
* CMake: forgot to update the CI for win and mac
* CMake: add missing F_interfaces sources to build set
* CMake: 2D setup for tutorials
* CMake: 1D setup for tutorials
* CMake: EB is made avaialble only if DIM>1
* CMake: exclude MPI-only tutorials from build if ENABLE_MPI=OFF
* CMake: some turotorials need Fortran enabled
* CMake: exclude CUDA-unaware tutorials from CUDA builds
* CMake: this should fix the Windows CI
* CMake: add tutorials in Tutorials/ForkJoin
* CMake: fix mispelled variable
* CMAke: Particles/CellSortedParticles tutorial is not CUDA-ready
* CMake: change naming scheme for tutorials target to avoid conflicts
* CMake: add tutorials in Tutorials/Blueprint
* CMake: replace option ENABLE_TUTORIALS with AMReX_BUILD_TUTORIALS
* Update Tools/CMake/AMReX_Options.cmake
* Update Tutorials/Amr/Advection_AmrLevel/CMakeLists.txt
* CMake: this is not needed anymore
* CMake: must use setup_target_for_cuda_compilation
* CMake: fix build setup for Particles/ElectromagneticPIC
* CMake: add tutorials in Tutorials/GPU
* Fix EBCNS particles tutorials.
The syntax for reductions has changed since itwas written.
* CMake: improvements to setup_tutorials()
* CMake: document how to build tutorials
Co-authored-by: Axel Huebl
.github/workflows/linux.yml
.github/workflows/macos.yml
.github/workflows/windows.yml
CMakeLists.txt
Docs/sphinx_documentation/source/BuildingAMReX.rst
Src/F_Interfaces/CMakeLists.txt
Tools/CMake/AMReX_Options.cmake
Tutorials/Amr/Advection_AmrCore/CMakeLists.txt
Tutorials/Amr/Advection_AmrLevel/CMakeLists.txt
Tutorials/Basic/HeatEquation_EX1_C/CMakeLists.txt
Tutorials/Basic/HeatEquation_EX1_CF/CMakeLists.txt
Tutorials/Basic/HeatEquation_EX1_F/CMakeLists.txt
Tutorials/Basic/HeatEquation_EX2_C/CMakeLists.txt
Tutorials/Basic/HeatEquation_EX2_CF/CMakeLists.txt
Tutorials/Basic/HeatEquation_EX3_C/CMakeLists.txt
Tutorials/Basic/HelloWorld_C/CMakeLists.txt
Tutorials/Basic/HelloWorld_F/CMakeLists.txt
Tutorials/Basic/PrefixSum_MultiFab/CMakeLists.txt
Tutorials/Basic/main_C/CMakeLists.txt
Tutorials/Basic/main_F/CMakeLists.txt
Tutorials/Blueprint/AssignMultiLevelDensity/CMakeLists.txt
Tutorials/Blueprint/CellSortedParticles/CMakeLists.txt
Tutorials/Blueprint/HeatEquation_EX1_C/CMakeLists.txt
Tutorials/CMakeLists.txt
Tutorials/EB/CNS/CMakeLists.txt
Tutorials/EB/Donut/CMakeLists.txt
Tutorials/EB/GeometryGeneration/CMakeLists.txt
Tutorials/EB/LevelSet/CMakeLists.txt
Tutorials/EB/MacProj/CMakeLists.txt
Tutorials/EB/Poisson/CMakeLists.txt
Tutorials/ForkJoin/MLMG/CMakeLists.txt
Tutorials/ForkJoin/Simple/CMakeLists.txt
Tutorials/FortranInterface/Advection_F/CMakeLists.txt
Tutorials/FortranInterface/Advection_F/Source/Src_3d/Make.package
Tutorials/FortranInterface/Advection_F/Source/Src_3d/advect_3d_mod.F90
Tutorials/FortranInterface/Advection_octree_F/CMakeLists.txt
Tutorials/FortranInterface/Advection_octree_F2/CMakeLists.txt
Tutorials/GPU/CNS/CMakeLists.txt
Tutorials/GPU/EBCNS/CMakeLists.txt
Tutorials/GPU/EBCNS/Source/CNS.cpp
Tutorials/GPU/EBCNS/Source/CNS_K.H
Tutorials/GPU/Launch/CMakeLists.txt
Tutorials/GPU/ParallelReduce/CMakeLists.txt
Tutorials/GPU/ParallelScan/CMakeLists.txt
Tutorials/LinearSolvers/ABecLaplacian_C/CMakeLists.txt
Tutorials/LinearSolvers/ABecLaplacian_F/CMakeLists.txt
Tutorials/LinearSolvers/MAC_Projection_EB/CMakeLists.txt
Tutorials/LinearSolvers/MultiComponent/CMakeLists.txt
Tutorials/LinearSolvers/NodalPoisson/CMakeLists.txt
Tutorials/LinearSolvers/Nodal_Projection_EB/CMakeLists.txt
Tutorials/LinearSolvers/NodeTensorLap/CMakeLists.txt
Tutorials/Particles/CellSortedParticles/CMakeLists.txt
Tutorials/Particles/ElectromagneticPIC/CMakeLists.txt
Tutorials/Particles/ElectrostaticPIC/CMakeLists.txt
Tutorials/Particles/NeighborList/CMakeLists.txt
commit a2284e9368e079dd4030ec9e0c41512f191848cf
Author: Andrew Myers
Date: Tue Sep 15 08:31:16 2020 -0700
Reorder struct (#1368)
This reverts a breaking change introduced in 1337.
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_ArrayOfStructs.H
Src/Particle/AMReX_Particle.H
commit 8929086e08058ac61d096414363341485afec34b
Author: Andrew Myers
Date: Mon Sep 14 08:30:11 2020 -0700
Implement Particle in a way that does not involve UB. (#1337)
Src/Particle/AMReX_Particle.H
Tests/Particles/Redistribute/main.cpp
commit 6fa3cc67ed0997e12877cdf3b5baf6d594d1c538
Author: WeiqunZhang
Date: Mon Sep 14 07:56:41 2020 -0700
Updates for oneAPI beta09 (#1365)
Src/Base/AMReX_GpuQualifiers.H
Tools/GNUMake/Make.defs
Tools/GNUMake/Make.rules
Tools/GNUMake/comps/dpcpp.mak
commit 82339abdc742453d076d0bfeca7d9ed4e479eb0c
Author: WeiqunZhang
Date: Sat Sep 12 18:24:06 2020 -0700
Revert "DPCPP Beta09: Package Broken (#1361)" (#1364)
This reverts commit 198a4958cc2cbfd5a7e1d639d5ff750489104df6.
The workaround does not work with #1363 because of mkl. We have now
temporarily removed dpc++ from the list of required checks. So we can
revert this change now. Once the packages are fixed, we will make the check
require again.
.github/workflows/linux.yml
commit 5175acf9efc2e8aeb37fc581480aba034a1fbb3f
Author: Candace Gilet
Date: Fri Sep 11 12:55:43 2020 -0400
amrex::EB_average_down(): Fix indexing error. (#1360)
Src/EB/AMReX_EBMultiFabUtil.cpp
commit 198a4958cc2cbfd5a7e1d639d5ff750489104df6
Author: Axel Huebl
Date: Fri Sep 11 07:55:44 2020 -0700
DPCPP Beta09: Package Broken (#1361)
## Summary
The Ubuntu packages seem to forget the `env/var.sh` activation
scripts for the compiler packages for beta09.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
.github/workflows/linux.yml
commit de4aab4f20ebffa862a0f9c29daa2717d8abcbd3
Author: Andrew Myers
Date: Thu Sep 10 17:00:08 2020 -0700
only add one copy of the ghost particle regardless of how many isects we have (#1359)
Src/Particle/AMReX_ParticleContainerI.H
commit fe4c98e749e40c65f03046a3efc9df5f948ced69
Author: WeiqunZhang
Date: Thu Sep 10 13:47:36 2020 -0700
SetBoxArray (#1310)
If SetBoxArray is called in Level 0's MakeNewLevelFromScrach, the set
BoxArray will be used. This makes it consistent with the behavior of
SetDistributionMap.
Src/AmrCore/AMReX_AmrMesh.H
Src/AmrCore/AMReX_AmrMesh.cpp
commit b1fcd62bd7f98fce1be98313307697d61320687d
Author: WeiqunZhang
Date: Thu Sep 10 13:46:42 2020 -0700
Remove duplicated alias. (#1311)
Src/Base/AMReX_GpuAllocators.H
commit a1eafe7ab57324e7b574ab5b636ba0fc98ea7635
Author: asalmgren
Date: Wed Sep 9 19:36:15 2020 -0700
Logic wasn't quite right for the calculation in amrex_calc_slopes_ext… (#1358)
…dir_eb routine
of slope not seeing EB cells that is adjacent to and tangential to ext_dir boundary
## Summary
Previous commit of AMReX_EB_slopes.cpp had messed up the logic for calcluation of the slope in amrex_calc_slopes_extdir_eb routine in cells not seeing any EB but next to ext_dir bondary and tangential to boundary
## Additional background
## Checklist
The proposed changes:
- [X] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/EB/AMReX_EB_slopes_K.H
commit 27c21a0c56623022a9b012c0ec2ec837e4971366
Author: WeiqunZhang
Date: Wed Sep 9 19:03:25 2020 -0700
Gpu::single_task (#1357)
Add a new version that uses the current default stream.
Src/Base/AMReX_GpuLaunchFunctsG.H
commit 6b032b4188e77e2b089e21c74586f5a2e6c1e5d2
Author: asalmgren
Date: Wed Sep 9 08:48:34 2020 -0700
Add non-EB slopes and extend the generality of EB slope routines in 2D and 3D (#1350)
## Summary
This commit adds a file with the non-EB slope routines, and updates the EB-aware routines to correctly compute slopes where an EB boundary intersects an hoextrap/extdir face, e.g. inflow. This does not assume the geometry intersects without a slant. This has been tested in 2D and 3D (with a function f(x,y,z) = 10 + 5 x + 2 y + 3 z specified at cell centroids) and with slanted cylinders intersecting inflow in each coordinate direction.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_Slopes_K.H
Src/Base/CMakeLists.txt
Src/Base/Make.package
Src/EB/AMReX_EB_slopes_K.H
commit 7cc452c65185efbb07e6503cb79206d10c0dc35c
Author: Axel Huebl
Date: Wed Sep 9 08:40:32 2020 -0700
Async IO: Check Runtime Threading (#1351)
## Summary
Even if `AMREX_MPI_THREAD_MULTIPLE` is defined at compile time, applications might provide their own MPI init routines and/or MPI implementations, runtime arguments or environment arguments can influence the actually provided MPI thread safety level at runtime.
Therefore, we need to check the MPI thread safety level again at runtime where we require a certain minimum.
## Additional background
WarpX PR: https://github.com/ECP-WarpX/WarpX/pull/1298
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_AsyncOut.cpp
Src/Base/AMReX_ParallelDescriptor.H
Src/Base/AMReX_ParallelDescriptor.cpp
commit f542f98d72c2cd651a0eae4d092c12239976db3f
Author: Axel Huebl
Date: Wed Sep 9 08:31:53 2020 -0700
Docs: MPI_THREAD_MULTIPLE (#1355)
## Summary
Add GNUmake and CMake docs about control of `MPI_THREAD_MULTIPLE`.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [x] are described in the proposed changes to the AMReX documentation, if appropriate
Docs/sphinx_documentation/source/BuildingAMReX.rst
commit a9aa65cc3841375b762324fe79516f5cfdee7364
Author: Andrew Myers
Date: Tue Sep 8 17:17:57 2020 -0700
also use atomics in the particle_dx != dx case (#1356)
Src/Particle/AMReX_Particle_mod_K.H
commit c59619f5ff32a78494b0b2328914fe108254844c
Author: Andrew Myers
Date: Tue Sep 8 15:53:07 2020 -0700
Fix EM pic tutorial - the syntax for reductions has changed since it was written. (#1354)
Tutorials/Particles/ElectromagneticPIC/Exec/CUDA/Evolve.cpp
Tutorials/Particles/ElectromagneticPIC/Exec/CUDA/em_pic_K.H
commit 96d1381e1f20adf48df84ff4f16601c887dac239
Author: WeiqunZhang
Date: Tue Sep 8 12:34:35 2020 -0700
USE_EB (#1349)
Changes to make AMR codes compiled with EB but not building any Geometry
work as if it is all regular.
Src/Amr/AMReX_AmrLevel.cpp
Src/AmrCore/AMReX_FillPatchUtil_I.H
commit a2aebd24854b5d375899a44156603f4bc9c85a64
Author: Andrew Myers
Date: Tue Sep 8 10:49:03 2020 -0700
Always recompute offset, dx, and inv_dx, and the roundoff_domain when either the ProbDomain or Domain is changed. (#1353)
Src/Base/AMReX_Geometry.H
Src/Base/AMReX_Geometry.cpp
commit cd075c81967aca29a17a52af3f41310a8426b121
Author: Axel Huebl
Date: Tue Sep 8 08:13:49 2020 -0700
Fallthrough: ICC Warning (#1352)
## Summary
`gnu::fallthrough` is not defined with ICC 20.2.1.20200602 and throws a warning at compile-time in ParamsParse.
## Additional background
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_Extension.H
commit 9ec96e6f9b286ccd26a61a910f0fb97a08a377fe
Author: Brian Friesen
Date: Sun Sep 6 09:19:33 2020 -0700
bcf/hpcsdk-updates (#1348)
## Summary
This PR fixes two wrong compile/link flags in the NVIDIA HPC SDK which are documented in #1347.
## Additional background
The wrong/missing flags are documented in #1347.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/comps/hpcsdk.mak
commit 31421f70c3cf73ae2cc9aaa4e7e024f554a69e20
Author: asalmgren
Date: Fri Sep 4 17:35:28 2020 -0700
Add slopes routines (#1346)
* Add the calculation of slopes using a least squares fit when the values in each cell
are defined to be at cell centroids.
* Add to the description of each routine
* EB_SLOPES_K_H --> AMREX_EB_SLOPES_K_H
Src/EB/AMReX_EB_slopes_K.H
Src/EB/CMakeLists.txt
Src/EB/Make.package
commit 7e0d7196dd1e8ac91625030622b360cbcafa21ce
Author: WeiqunZhang
Date: Fri Sep 4 17:27:40 2020 -0700
No inline for assertion (#1345)
It is recommended that assertion calls on device are not inlined because
of performance issues.
Src/Base/AMReX.H
Src/Base/AMReX.cpp
Src/Base/AMReX_Extension.H
Src/Base/AMReX_GpuDevice.cpp
commit a7166b92dc0b58b1ed96110ed2a43341513dcdd3
Author: WeiqunZhang
Date: Fri Sep 4 16:21:32 2020 -0700
Fix mkconfig.py (#1344)
## Summary
In the script, Intel, Cray and PGI should be quoted as string.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/libamrex/mkconfig.py
commit de8a23ac50146b732aeaadf874ead2cae93753c8
Author: Andrew Myers
Date: Fri Sep 4 14:43:29 2020 -0700
Add -Xcudafe --diag_suppress=esa_on_defaulted_function_ignored to suppress spurious warnings from nvcc. (#1342)
Tools/CMake/AMReX_SetupCUDA.cmake
Tools/GNUMake/comps/nvcc.mak
commit d8747f753f9f5524aba04e1a70fee3d1665654bd
Author: Axel Huebl
Date: Fri Sep 4 13:39:44 2020 -0700
CI: Move DPC++ dependencies (#1336)
Move install script into unified location.
## Summary
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
.github/workflows/dependencies/dependencies_dpcpp.sh
.github/workflows/linux.yml
commit b9e746e447e896aa0df049845d573e11f7db10ab
Author: Rajeev Jain
Date: Fri Sep 4 15:36:46 2020 -0500
Update AMReX_multifab_mod.F90 (#1339)
#1338
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/F_Interfaces/Base/AMReX_multifab_mod.F90
commit 2ae64e7f472e81000c25ca97eb7587f4d43d0960
Author: Andrew Myers
Date: Fri Sep 4 13:35:18 2020 -0700
PR 1324 did not account for runtime-added components (#1341)
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_ParticleContainerI.H
commit e34912f3bc26d532fa9c3ee608ec5e50c4ce26f9
Author: WeiqunZhang
Date: Thu Sep 3 10:29:23 2020 -0700
update CHANGES for 20.09 (#1334)
CHANGES
commit c7cef052978eb509972a8f2c6b43771f22cf4fff
Author: Andrew Myers
Date: Wed Sep 2 19:11:10 2020 -0700
Add support for extended particle range to the Fortran interfaces (#1331)
* add support for extended particle range to the Fortran interfaces
* replace c_long -> amrex_long for Windows support
Src/F_Interfaces/Particle/AMReX_particlecontainer_fi.cpp
Src/F_Interfaces/Particle/AMReX_particlecontainer_mod.F90
Tutorials/FortranInterface/Advection_F/Exec/SingleVortex/Prob_2d.f90
Tutorials/FortranInterface/Advection_F/Exec/SingleVortex/Prob_3d.f90
commit 64e772795b9f9b09db11e4fbdf30552331850965
Author: WeiqunZhang
Date: Wed Sep 2 16:43:45 2020 -0700
Avoid divide-by-zero in 2D EB (#1319)
Src/EB/AMReX_EB2_2D_C.cpp
Src/EB/AMReX_EB2_3D_C.cpp
commit 3d65510d12ef10fe2a7d4b0d0bf9164615689597
Author: WeiqunZhang
Date: Wed Sep 2 16:34:45 2020 -0700
EB smoother (#1330)
* EB smoother
In #1327, the interpolation coefficients for EB stencil in cell-centered
linear solvers were fixed. Here, the EB smoother is also updated to use the
correct coefficients.
* fix order in interpolation coefficient
* Reduce max order to 2 for cut cells.
* If using hypre or petsc for EB solver, reduce order to 2.
Src/LinearSolvers/MLMG/AMReX_MLCellLinOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLEBABecLap_K.H
Src/LinearSolvers/MLMG/AMReX_MLLinOp_K.H
Src/LinearSolvers/MLMG/AMReX_MLMG.cpp
commit 4128f25ffa247802ab93e8fae7133b80a5e3d59f
Author: Roberto Porcu <53792251+rporcu@users.noreply.github.com>
Date: Wed Sep 2 14:21:45 2020 -0400
Gpu synchronization is needed to avoid deadlock situations (#1328)
Src/Particle/AMReX_ParticleCommunication.H
commit 65f25c7db1ba2f88654cff8e91d24cf3ff45b075
Author: OscarAntepara <52221614+OscarAntepara@users.noreply.github.com>
Date: Tue Sep 1 17:26:10 2020 -0700
Fix bug to correctly compute the interp_coeff array (#1327)
Src/LinearSolvers/MLMG/AMReX_MLEBABecLap_K.H
commit d5e29d965070a36c6e232ef114d1f5bd88b701e3
Author: WeiqunZhang
Date: Tue Sep 1 17:02:51 2020 -0700
EB extend_domain_face (#1321)
## Summary
When there are a lot of ghost cells, extend_domain_face option could fail in
the current development branch. This PR fixes it.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Docs/sphinx_documentation/source/Basics.rst
Src/EB/AMReX_EB2_Level.H
commit e4444fa80264ae716f9cadd4d7e8847fcf3d833b
Author: Andrew Myers
Date: Tue Sep 1 17:02:26 2020 -0700
Add explicit default copy ctors for ParticleIDWrapper and ParticleCPUWrapper (#1323)
This should fix a warning with DPC++.
Src/Particle/AMReX_Particle.H
commit 98b73e64bfcc839ba168b3e423d4d2f8e4914c28
Author: Klaus
Date: Tue Sep 1 19:01:42 2020 -0500
Add parameter that stores OpenMP support version (#1325)
## Summary
Add a parameter to `AMReX_omp_mod` that can be used by Fortran code to check whether AMReX was built with OpenMP support, and, if yes, what the version number (_OPENMP) was at the time AMReX was compiled.
## Additional background
The name of the parameter is `amrex_omp_support`, the type is (default kind) integer.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new **(mini)**-capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Co-authored-by: Klaus Weide <>
Src/Base/AMReX_omp_mod.F90
commit 1ffbfc1caab2dae19db0e7fa27202938300f9ced
Author: Andrew Myers
Date: Tue Sep 1 17:00:43 2020 -0700
Use Long for maxnextid in plot and checkfiles to accomodate the expanded range. (#1326)
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_ParticleHDF5.H
Src/Particle/AMReX_ParticleIO.H
Src/Particle/AMReX_WriteBinaryParticleData.H
commit 57d5bfeb97b03ce037714a08260d78660b6774be
Author: WeiqunZhang
Date: Tue Sep 1 16:46:23 2020 -0700
ParticleContainer: communicate_real_comp and commuicate_int_comp (#1324)
Having both host and device copies of these two so that we do not rely on
unified memory.
Src/Particle/AMReX_ParticleCommunication.H
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_Particles.H
commit b7479f0a662a9b256f1e42d1cd3c628457a53c1c
Author: mic84
Date: Tue Sep 1 11:15:35 2020 -0700
CMake: re-write genex evaluation functions (#1287)
* CMake: start adding genex evaluation functions
* CMake: more genex evaluation functions
* CMake: add support for more genex types
* CMake: finalize new function to evaluate genex
* CMake: slighly change the interface of function eval_genex
* CMake: deploy eval_genex.
* CMake: set MSVC required flags using regex
* CMake: final improvements to new genex evaluation helpers
Tools/CMake/AMReXBuildInfo.cmake
Tools/CMake/AMReXGenexHelpers.cmake
Tools/CMake/AMReXInstallHelpers.cmake
Tools/CMake/AMReXTargetHelpers.cmake
Tools/CMake/AMReXTypecheck.cmake
Tools/CMake/AMReX_Config.cmake
commit 6dd02ab17f06478d7fbea8e91ba274f932004801
Author: Andrew Myers
Date: Thu Aug 27 19:08:59 2020 -0700
Need another overload of the operator= for ParticleIDWrapper and ParticleCPUWrapper. (#1317)
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_Particle.H
commit 5e9dbc1dd2fb8675de53bc42bd8b3d13f9eacb88
Author: Andrew Myers
Date: Thu Aug 27 11:12:18 2020 -0700
Extend the number of unique particles per cpu we can have at once. (#1315)
Currently, we use two signed integers to store id numbers for each particle as well as the rank it was generated on. This allows unique combinations of 'id', and 'cpu' numbers to be generated without any communication between ranks. However, this does waste some space, since it's unlikely that 2**31-1 MPI ranks will be used any time soon, while the same limit for the id has actually been overflowed in real-world WarpX simulations. To address this, in this PR, we still use 64 bits to represent the combination of (id, cpu), but we devote 40 bits to the id and only 24 to the cpu. This allows ~0.5 trillion unique particles on each of 16.7 million MPI ranks, which should be good enough for the foreseeable future.
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_Particle.H
Src/Particle/AMReX_ParticleContainerI.H
commit d48f68f1a641084009e597d955ce3036d36b1755
Author: WeiqunZhang
Date: Wed Aug 26 09:00:42 2020 -0700
Add ParmParse parameters to Tutorials/GPU/Launch (#1314)
Tutorials/GPU/Launch/main.cpp
commit 655f10afbc68b5c4279d2ca78c7add8b449f71e4
Author: WeiqunZhang
Date: Tue Aug 25 15:57:40 2020 -0700
HIP: printf works now on device (#1312)
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_GpuPrint.H
commit 02530bc8a600e2a3589b875208f372dd2e462a31
Author: Klaus
Date: Tue Aug 25 15:44:05 2020 -0500
Add get_pmap method to amrex_distromap (Fortran) (#1307)
## Summary
Add a new method to the `amrex_distromap` type in F_Interfaces. `get_pmap` fills a caller-owned array of PEs.
Changes to make this more elegant are welcome.
## Additional background
We need something like this in order to actually look into a distribution mapping computed by AMReX from Fortran code.
(Currently this functionality is desired for Implementing a parallel restart capability from FLASH checkpoint files in FLASH.)
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Co-authored-by: Klaus Weide <>
Src/F_Interfaces/Base/AMReX_distromap_fi.cpp
Src/F_Interfaces/Base/AMReX_distromap_mod.F90
commit 7b72e38bbbe63d2039a5e0ef7a4833792c99beff
Author: kngott <30483578+kngott@users.noreply.github.com>
Date: Tue Aug 25 12:04:39 2020 -0700
Nsight documentation fix. (#1309)
* Fix Nsight docs.
* Missing syntax.
Docs/sphinx_documentation/source/External_Profiling_Tools.rst
commit 51c4a990b42e37d84cc0c2b8055403d351febbb2
Author: Klaus
Date: Tue Aug 25 12:25:42 2020 -0500
Fix instruction for local development branch (#1305)
## Summary
Just a simple one-word doc change
## Additional background
I tried to follow instructions in `CONTRIBUTING.md` and found that the `git pull` command as documented before this change resulted in an error message.
## Checklist
The proposed changes:
- [x] fix a bug in AMReX documentation
Co-authored-by: Klaus Weide <>
CONTRIBUTING.md
commit a34e8e42d52b7725552ea97e5e089a90b96833cc
Author: WeiqunZhang
Date: Mon Aug 24 19:22:19 2020 -0700
Modify the use of HOST and HOSTNAME in make system (#1303)
## Summary
PR #1275 fixed a bug in the use of `HOST` and `HOSTNAME` in the make system.
However, the fix broke on Summit because `HOSTNAME` on Summit login nodes
is set to something like `login1`. We now set `host_name` to `hostname -f` and
`host_name_short` to `hostname -s` command, which is the behavior before the fix
in #1275. We add a new make variable `host_name_env` that is set to
`HOSTNAME` and `HOST` from the environment and use it for tulip.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/Make.machines
commit e82080e322f238435f925da35fb8bae36365269d
Author: Marc Day
Date: Mon Aug 24 18:00:46 2020 -0700
Move declaration of Array4 object (#1302)
## Summary
The autogen function for refinement tagging on physical box doesn't make use of a field, but the controller was trying to make an Array4 using the field that doesn't exist in this case. This fix moves the declaration down into a place where the data is guaranteed to exist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/AmrCore/AMReX_ErrorList.cpp
commit 21269eff092d0a03aff9269b1200c0e408fde90e
Author: Andrew Myers
Date: Mon Aug 24 12:31:26 2020 -0700
Remove early exit for the 1 box case in RedistributeGPU. (#1301)
This early exit had a bug in that it did not properly remove invalid particles.
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_Particles.H
commit bb831b49bbb3b853a18ee0e16d955aafe251da7d
Author: WeiqunZhang
Date: Sat Aug 22 21:04:40 2020 -0700
add Scan::InclusiveSum and ExclusiveSum for CPU (#1299)
## Summary
Add Scan::InclusiveSum and ExclusiveSum for CPU to avoid ifdef.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_Scan.H
commit 91f387dad9f57dfa884b4e5d03b88ff18f7b8c12
Author: WeiqunZhang
Date: Sat Aug 22 16:00:51 2020 -0700
Turn on AsyncOut by default for HIP (#1298)
## Summary
Currently for HIP, The_Arena allocates device memory. We turn on AsyncOut
by default so that we can write plotfiles and VisMF files. In the long term, we
should also fix VisMF::Write so that it can work with device memory.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_AsyncOut.cpp
commit ba556dfe28afaa6704ea049f8200a8b53cffd888
Author: mxxw
Date: Sat Aug 22 09:29:21 2020 -0700
removed AMREX_NOEXCEPT from main.cpp:L139 to avoid 'error: expected b… (#1297)
removed AMREX_NOEXCEPT from main.cpp:L139 to avoid 'error: expected body of lambda expression'
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tutorials/GPU/ParallelReduce/main.cpp
commit ab300111b705486f10782f1966020907d9f66b66
Author: jmsexton03
Date: Fri Aug 21 14:58:25 2020 -0400
Add check for empty probin_file in Amr::restart (#1295)
## Summary
This adds the same empty check in Amr::restart that is used in Amr::InitializeInit
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Amr/AMReX_Amr.cpp
commit a70adecd122ef1a4441cabd0f6356f53389e606b
Author: mic84
Date: Fri Aug 21 11:33:50 2020 -0700
CMake: fix typo in source file name (#1296)
Tutorials/Basic/HeatEquation_EX1_F/CMakeLists.txt
commit 373cb8ac3143e1252ccd9b181280d355e9399f8a
Author: WeiqunZhang
Date: Fri Aug 21 08:06:10 2020 -0700
Fix a few warnings (#1293)
Src/Base/AMReX_FBI.H
Src/Base/AMReX_GpuPrint.H
Src/Base/AMReX_GpuReduce.H
Src/Base/AMReX_PhysBCFunct.H
Src/Base/AMReX_Reduce.H
Src/Particle/AMReX_ParticleContainerI.H
commit 1aef797d0dd5a701c696d02165237c2c8e39771c
Author: WeiqunZhang
Date: Thu Aug 20 20:19:30 2020 -0700
Extern HIP device function (#1294)
## Summary
We now need to use __attribute__((weak)) for extern HIP device function,
otherwise we get undefined hidden symbol error.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_GpuQualifiers.H
commit ce5e0d4bf4b27a88fc3eb6a2dab137b6b25e2bbb
Author: jmsexton03
Date: Thu Aug 20 23:19:02 2020 -0400
Update Sundials interface, documentation, and build to be version agnostic (#1289)
## Summary
Updates the make system to use SUNDIALS_ROOT and USE_SUNDIALS
Updates the documentation regarding sundials and tutorials
Removes some CVODE/SUNDIALS tutorials
This will require applications to update their make variables
## Additional background
Co-authored-by: Balos, Cody Joe
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [x] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Co-authored-by: Balos, Cody Joe
Co-authored-by: Michele Rosso
Docs/sphinx_documentation/source/CVODE_top.rst
Docs/sphinx_documentation/source/External_Frameworks_Chapter.rst
Docs/sphinx_documentation/source/SUNDIALS.rst
Docs/sphinx_documentation/source/SUNDIALS3.rst
Docs/sphinx_documentation/source/SUNDIALS_CVODE.rst
Docs/sphinx_documentation/source/SUNDIALS_top.rst
Docs/sphinx_tutorials/source/CVODE_Tutorial.rst
Docs/sphinx_tutorials/source/SUNDIALS_Tutorial.rst
Src/CMakeLists.txt
Src/Extern/SUNDIALS/CMakeLists.txt
Src/Extern/SUNDIALS/Make.package
Src/Extern/SUNDIALS/arkode_interface.f90
Src/Extern/SUNDIALS/cvode_interface.f90
Src/Extern/SUNDIALS3/CMakeLists.txt
Src/Extern/SUNDIALS3/Make.package
Src/Extern/SUNDIALS3/arkode_interface.f90
Src/Extern/SUNDIALS3/cvode_interface.f90
Src/Extern/SUNDIALS3/farkode.f90
Src/Extern/SUNDIALS3/fcvode.f90
Src/Extern/SUNDIALS3/fnvector_serial.f90
Src/Extern/SUNDIALS3/fnvector_serial_fprefix.f90
Src/Extern/SUNDIALS3/fsunlinsol_dense.f90
Src/Extern/SUNDIALS3/fsunmat_dense.f90
Src/Extern/SUNDIALS4/CMakeLists.txt
Src/Extern/SUNDIALS4/Make.package
Src/Extern/SUNDIALS4/arkode_interface.f90
Src/Extern/SUNDIALS4/cvode_interface.f90
Src/Extern/SUNDIALS4/farkode_arkstep_mod.f90
Src/Extern/SUNDIALS4/farkode_mod.f90
Src/Extern/SUNDIALS4/fcvode_mod.f90
Src/Extern/SUNDIALS4/fnvector_mod.f90
Src/Extern/SUNDIALS4/fnvector_serial_mod.f90
Src/Extern/SUNDIALS4/fsundials_types_mod.f90
Src/Extern/SUNDIALS4/fsunlinsol_dense_mod.f90
Src/Extern/SUNDIALS4/fsunlinsol_mod.f90
Src/Extern/SUNDIALS4/fsunmatrix_dense_mod.f90
Src/Extern/SUNDIALS4/fsunmatrix_mod.f90
Tools/CMake/AMReXThirdPartyLibraries.cmake
Tools/CMake/FindSUNDIALS.cmake
Tools/GNUMake/Make.defs
Tools/GNUMake/packages/Make.sundials
Tools/GNUMake/packages/Make.sundials3
Tools/GNUMake/packages/Make.sundials4
Tutorials/CVODE/EX1_F/GNUmakefile
Tutorials/CVODE/EX1_F/Make.package
Tutorials/CVODE/EX1_F/inputs
Tutorials/CVODE/EX1_F/integrate_ode.f90
Tutorials/CVODE/EX1_F/main.cpp
Tutorials/CVODE/EX1_F/myfunc_F.H
Tutorials/CVODE/EX1_F/ode_mod.f90
Tutorials/CVODE/EX2_F/GNUmakefile
Tutorials/CVODE/EX2_F/Make.package
Tutorials/CVODE/EX2_F/inputs
Tutorials/CVODE/EX2_F/integrate_ode_no_jac.f90
Tutorials/CVODE/EX2_F/integrate_ode_with_jac.f90
Tutorials/CVODE/EX2_F/main.cpp
Tutorials/CVODE/EX2_F/myfunc_F.H
Tutorials/CVODE/EX2_F/ode_mod.f90
Tutorials/CVODE/SUNDIALS3_finterface/EX1/CMakeLists.txt
Tutorials/CVODE/SUNDIALS3_finterface/EX1/main.cpp
Tutorials/CVODE/SUNDIALS3_finterface/EX_ark_analytic_fp/GNUmakefile
Tutorials/CVODE/SUNDIALS3_finterface/EX_ark_analytic_fp/Make.package
Tutorials/CVODE/SUNDIALS3_finterface/EX_ark_analytic_fp/ark_analytic_fp.f90
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_fp/Make.package
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_fp/cv_analytic_fp.f90
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_fp/cv_analytic_fp.out
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_sys_dns/GNUmakefile
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_sys_dns/Make.package
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_sys_dns/cv_analytic_sys_dns.f90
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_sys_dns/cv_analytic_sys_dns.out
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_sys_dns_jac/GNUmakefile
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_sys_dns_jac/Make.package
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_sys_dns_jac/cv_analytic_sys_dns_jac.f90
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_sys_dns_jac/cv_analytic_sys_dns_jac.out
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_brusselator_dns/GNUmakefile
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_brusselator_dns/Make.package
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_brusselator_dns/cv_brusselator_dns.f90
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_brusselator_dns/cv_brusselator_dns.out
Tutorials/CVODE/SUNDIALS4/EX_ark_analytic/GNUmakefile
Tutorials/CVODE/SUNDIALS4/EX_ark_analytic/Make.package
Tutorials/CVODE/SUNDIALS4/EX_ark_analytic/ark_analytic_f2003.f90
Tutorials/SUNDIALS/EX-CUSOLVER/GNUmakefile
Tutorials/SUNDIALS/EX-CUSOLVER/Make.CVODE
Tutorials/SUNDIALS/EX-CUSOLVER/Make.package
Tutorials/SUNDIALS/EX-CUSOLVER/README.md
Tutorials/SUNDIALS/EX-CUSOLVER/extern_probin.template
Tutorials/SUNDIALS/EX-CUSOLVER/inputs
Tutorials/SUNDIALS/EX-CUSOLVER/inputs_128
Tutorials/SUNDIALS/EX-CUSOLVER/inputs_256
Tutorials/SUNDIALS/EX-CUSOLVER/inputs_32
Tutorials/SUNDIALS/EX-CUSOLVER/inputs_64
Tutorials/SUNDIALS/EX-CUSOLVER/main.cpp
Tutorials/SUNDIALS/EX-CUSOLVER/make_cuda.sh
Tutorials/SUNDIALS/EX-CUSOLVER/make_cuda_cusolver.sh
Tutorials/SUNDIALS/EX-CUSOLVER/make_serial.sh
Tutorials/SUNDIALS/EX-CUSOLVER/react_cuda.cpp
Tutorials/SUNDIALS/EX-CUSOLVER/react_cuda_cusolver.cpp
Tutorials/SUNDIALS/EX-CUSOLVER/react_serial.cpp
Tutorials/SUNDIALS/EX-CUSOLVER/react_system.F90
Tutorials/SUNDIALS/EX-CUSOLVER/react_utils.F90
Tutorials/SUNDIALS/EX-CUSOLVER/test_react.H
Tutorials/SUNDIALS/EX-CUSOLVER/test_react_F.H
Tutorials/SUNDIALS/EX1_C/CMakeLists.txt
Tutorials/SUNDIALS/EX1_C/GNUmakefile
Tutorials/SUNDIALS/EX1_C/Make.package
Tutorials/SUNDIALS/EX1_C/SetIC.f90
Tutorials/SUNDIALS/EX1_C/inputs
Tutorials/SUNDIALS/EX1_C/inputs_2box
Tutorials/SUNDIALS/EX1_C/inputs_non_vectorized
Tutorials/SUNDIALS/EX1_C/main.cpp
Tutorials/SUNDIALS/EX1_C/myfunc_F.H
Tutorials/SUNDIALS/EX1_CUDA/GNUmakefile
Tutorials/SUNDIALS/EX1_CUDA/Make.package
Tutorials/SUNDIALS/EX1_CUDA/SetIC.f90
Tutorials/SUNDIALS/EX1_CUDA/inputs
Tutorials/SUNDIALS/EX1_CUDA/main.cpp
Tutorials/SUNDIALS/EX1_CUDA/myfunc_F.H
Tutorials/SUNDIALS/EX1_CUDA/ode_mod.f90
Tutorials/SUNDIALS/EX1_F/GNUmakefile
Tutorials/SUNDIALS/EX1_F/Make.package
Tutorials/SUNDIALS/EX1_F/inputs
Tutorials/SUNDIALS/EX1_F/integrate_ode.f90
Tutorials/SUNDIALS/EX1_F/main.cpp
Tutorials/SUNDIALS/EX1_F/myfunc_F.H
Tutorials/SUNDIALS/EX1_F/ode_mod.f90
Tutorials/SUNDIALS/EX2_F/GNUmakefile
Tutorials/SUNDIALS/EX2_F/Make.package
Tutorials/SUNDIALS/EX2_F/inputs
Tutorials/SUNDIALS/EX2_F/integrate_ode_no_jac.f90
Tutorials/SUNDIALS/EX2_F/integrate_ode_with_jac.f90
Tutorials/SUNDIALS/EX2_F/main.cpp
Tutorials/SUNDIALS/EX2_F/myfunc_F.H
Tutorials/SUNDIALS/EX2_F/ode_mod.f90
commit 418f90d6f6620008edc0385ce6f487cb555ae1fb
Author: Candace Gilet
Date: Thu Aug 20 21:15:56 2020 -0400
Fix typo (#1292)
## Summary
Fix a typo in AMReX_extrapolater_3D_K.H
## Additional background
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Amr/AMReX_extrapolater_3D_K.H
commit f38919982f59b25f49b9b2a4c948f20677f144e7
Author: asalmgren
Date: Thu Aug 20 16:26:04 2020 -0700
Un-comment lines which should not have been commented (#1291)
## Summary
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tests/LinearSolvers/EBTensor/MyTest.cpp
commit df022f23e57792da09feed9b9db74dc60e6d4847
Author: WeiqunZhang
Date: Thu Aug 20 14:14:12 2020 -0700
Skip MPI checking when doing make clean etc. (#1290)
## Summary
When we do `make clean`, `make realclean` etc., we do not need to check MPI. This can avoid unnecessary errors.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/sites/Make.unknown
commit 924ec34c858335848b9514590f41d221990f209d
Author: kngott <30483578+kngott@users.noreply.github.com>
Date: Thu Aug 20 13:55:11 2020 -0700
Get rid of incorrect comment. (#1288)
## Summary
Just deleting this old, spurious comment before I forget about it.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/Make.defs
commit 7c35abbcdf3882d0b83c5746a23c5f2eca24a2dc
Author: WeiqunZhang
Date: Thu Aug 20 10:20:27 2020 -0700
Physical locations in pltofiles (#1286)
Normally, the integer index of a Geometry's Domain starts with 0. But if
not, the physical location of boxes stored in plotfiles is incorrect,
because the RealBox constructor we use takes the physical location of index
0. So we need to shift the box before using RealBox.
Src/Base/AMReX_PlotFileUtil.cpp
commit 0748356f06fadf2f6caef6542f48d15c3e9391bc
Author: WeiqunZhang
Date: Thu Aug 20 08:20:26 2020 -0700
Misc. updates for ROCm 3.6.0 (#1285)
## Summary
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_BaseFab.H
Src/Base/AMReX_Box.H
Src/Base/AMReX_Dim3.H
Src/Base/AMReX_Geometry.H
Src/Base/AMReX_GpuAssert.H
Src/Base/AMReX_GpuComplex.H
Src/Base/AMReX_GpuPrint.H
Src/Base/AMReX_GpuReduce.H
Src/Base/AMReX_GpuUtility.H
Src/Base/AMReX_Math.H
Src/Base/AMReX_MemPool.cpp
Src/Base/AMReX_PlotFileDataImpl.H
Src/Base/AMReX_Random.cpp
Src/Base/AMReX_RealVect.H
Tools/GNUMake/comps/hip.mak
Tutorials/Basic/HeatEquation_EX2_C/Source/mykernel.H
Tutorials/GPU/CNS/Source/CNS_K.H
commit e33b6cd8b866c2f8f34a16491316f912960922c0
Author: Marc Day
Date: Wed Aug 19 17:56:11 2020 -0700
Add section to Make.unknown for intel mpi (#1284)
## Summary
Avoid another site-specific make setting by adding generic id for Intel MPI - seems to work for UT Austin's TACC center computer, Stampede2, for example.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/sites/Make.unknown
commit c35f7aad31ec34f55f3dcfab9b42f19a31224a18
Author: mic84
Date: Wed Aug 19 15:55:31 2020 -0700
CMake: re-factor third party libraries setup (#1252)
## Summary
Centralize setup of all parallel backends dependencies.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/CMakeLists.txt
Src/Extern/Conduit/CMakeLists.txt
Src/Extern/HYPRE/CMakeLists.txt
Src/Extern/PETSc/CMakeLists.txt
Src/Extern/SENSEI/CMakeLists.txt
Src/Extern/SUNDIALS4/CMakeLists.txt
Tools/CMake/AMReXConfig.cmake.in
Tools/CMake/AMReXThirdPartyLibraries.cmake
Tools/CMake/AMReX_Config.cmake
commit 547e0f040a8ec6c03b73a4a001eaa34fd369121e
Author: asalmgren
Date: Wed Aug 19 12:43:32 2020 -0700
1) replace gpu_regtest flag by plot_error flag (which can be used on … (#1283)
…CPU or GPU)
2) add analogous 2D test problem -- also with analytical solution
## Summary
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [X] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tests/LinearSolvers/EBTensor/Make.package
Tests/LinearSolvers/EBTensor/MyTest.cpp
Tests/LinearSolvers/EBTensor/MyTestPlotfile.cpp
Tests/LinearSolvers/EBTensor/MyTest_2D_K.H
Tests/LinearSolvers/EBTensor/MyTest_3D_K.H
Tests/LinearSolvers/EBTensor/inputs
Tests/LinearSolvers/EBTensor/inputs.rt.2d
commit 5a81f883ec4d83fbe4c31c8e18db023deade5623
Author: Jon Rood
Date: Wed Aug 19 10:36:34 2020 -0600
Add use of CUDA_HOME to Make.nrel. (#1282)
## Summary
We set `CUDA_HOME` on machines at NREL so this adds its use to `Make.nrel`. This is mostly so that `-lnvToolsExt` will happen automatically for us when using the tiny profiler.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/sites/Make.nrel
commit 192f638662e02c660e98bca5970c5e75b4c535b5
Author: WeiqunZhang
Date: Wed Aug 19 06:58:12 2020 -0700
fcompare (#1281)
## Summary
Remove unimplemented option from help message. Print more information on
which files have NaNs if there are.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/Plotfile/fcompare.cpp
commit 33fdf6bbc80eacca651766697e991c21bae8d32d
Author: WeiqunZhang
Date: Tue Aug 18 19:55:11 2020 -0700
HIP: exp (#1280)
## Summary
ROCm 3.6 has fixed std::exp.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tutorials/Basic/HeatEquation_EX1_C/Source/mykernel.H
commit 2d258baa734c8929035a1a2523d87695881c2008
Author: WeiqunZhang
Date: Tue Aug 18 19:54:47 2020 -0700
Fix a bug in VisMF::AsyncWrite (#1279)
## Summary
When not writing ghost cells, we need to make sure the header has the
correct number of ghost cells that is zero.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_VisMF.cpp
commit cb2cdfafde7c00f62e951d7060552c71167856b1
Author: WeiqunZhang
Date: Tue Aug 18 17:21:46 2020 -0700
Updates for Tulip (#1278)
## Summary
* Force DEBUG=FALSE for HIP to avoid a compiler bug.
* Modify makefile so that the dependency files are not dumped in the current
directory.
* Add support for MPI.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/Make.defs
Tools/GNUMake/Make.rules
Tools/GNUMake/sites/Make.frontier-coe
commit ce38e4b385fdd54f8bb67f254904802ceacc4079
Author: WeiqunZhang
Date: Tue Aug 18 15:15:54 2020 -0700
OMP CI (#1277)
Add an OMP CI test.
.github/workflows/linux.yml
commit 8688f76e033b5868762c178fe4dc20ef2930032a
Author: WeiqunZhang
Date: Tue Aug 18 14:57:39 2020 -0700
Fix typo for OMP (#1276)
The typo was introduced in #1274.
Src/AmrCore/AMReX_TagBox.cpp
commit b0735b5e604305f226b42816d9f0d3c882ff3c4c
Author: WeiqunZhang
Date: Tue Aug 18 14:34:07 2020 -0700
Fix Make.machines (#1275)
The tests on environment variables HOSTNAME and HOST were incorrect.
Tools/GNUMake/Make.machines
commit f71d122d0ed6831ae946e108116a88bee9b563d7
Author: WeiqunZhang
Date: Tue Aug 18 14:02:54 2020 -0700
TagBoxArray::hasTags (#1274)
In #1258, TagBoxArray::numtags was removed. However, IAMR still needs it.
So a new function, hasTags, is added for IAMR.
Src/AmrCore/AMReX_TagBox.H
Src/AmrCore/AMReX_TagBox.cpp
commit 6478336784067131d857a3f219220b6b00127751
Author: WeiqunZhang
Date: Mon Aug 17 21:19:11 2020 -0700
TagBoxArray on GPU (#1258)
# Summary
TagBoxArray functions can now run on GPU now. We no longer rely on unified memory for TagBoxArray functions.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/AmrCore/AMReX_TagBox.H
Src/AmrCore/AMReX_TagBox.cpp
Src/Base/AMReX_AsyncOut.cpp
Src/Base/AMReX_GpuAtomic.H
Src/Base/AMReX_GpuDevice.H
Src/Base/AMReX_Scan.H
commit 3235db6b295923ed454cc1bbdfe6ce3b6964e93c
Author: mxxw
Date: Mon Aug 17 16:36:17 2020 -0700
Make.machines (#1273)
## Summary
ensure host_name_short is always defined in Tools/GNUMake/Make.machines
## Additional background
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/Make.machines
commit 8b66f47e906c53cd0127531d97fd65508d8eb9f9
Author: WeiqunZhang
Date: Mon Aug 17 10:32:21 2020 -0700
DPC++: Missing synchronize() (#1268)
Missing Gpu::synnhronize() call after host to device memcpy.
Src/Base/AMReX_Reduce.H
commit addfc7d6f35604f43893d836c891138b373757d2
Author: WeiqunZhang
Date: Mon Aug 17 10:02:25 2020 -0700
Should use std::forward. (#1267)
Src/Base/AMReX_FabArrayUtility.H
commit 2c45e058537a3bffd212e13968e6df6562f01bc3
Author: WeiqunZhang
Date: Mon Aug 17 09:51:32 2020 -0700
Advection_AmrCore Tutorial (#1266)
Remove redundant MPI_Allreduce and cleanup.
Tutorials/Amr/Advection_AmrCore/Source/AdvancePhiAtLevel.cpp
Tutorials/Amr/Advection_AmrCore/Source/AmrCoreAdv.H
Tutorials/Amr/Advection_AmrCore/Source/AmrCoreAdv.cpp
commit 2d16fa694c4e08cc6ce922d31da730d4ab116985
Author: WeiqunZhang
Date: Mon Aug 17 09:03:24 2020 -0700
amrex::Math::abs for beta8 on DG1 (#1271)
## Summary
Due to a compiler bug (I believe), we have to modify amrex::Math::abs for
beta8 on DG1.
## Additional background
After spending many hours debugging Tutorial/Amr/Adevection_AmrCore/
on DG1, it is found that `MultiFab::norm0` does not return the correct
result, because of the use of `amrex::Math::abs`. After making the change
in this PR, the test now produces correct results with AMR.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_Math.H
commit b4128deaff6e9ec9a91b605271f1aa972a1d3809
Author: mxxw
Date: Sun Aug 16 18:21:43 2020 -0700
add support for additional AMD GPU SKUs and incl. poplar and redwood … (#1270)
#1168 Summary
Changed Tools/GNUMake/sites/Make.frontier-coe to
1. add support for additional AMD GPU SKUs;
2. include poplar and redwood in the list of frontier-coe machines
## Additional background
Changed
ifeq ($(which_computer),$(filter $(which_computer),tulip))
ifeq ($(USE_HIP),TRUE)
CXXFLAGS += --amdgpu-target=gfx906
HIPCC_FLAGS += --amdgpu-target=gfx906
endif
endif
to
ifeq (,$(filter $(which_computer),poplar redwood tulip))
$(error Unknown Frontier CoE computer, $(which_computer))
else
ifeq ($(USE_HIP),TRUE)
CXXFLAGS += --amdgpu-target=gfx906,gfx908
HIPCC_FLAGS += --amdgpu-target=gfx906,gfx908
endif
endif
For AMD GPUs, the compile target should be exact to avoid runtime errors,
i.e., one should generally not assume that a binary targeting an older GPU
will always run correctly on a newer GPU. However, building "fat binaries"
is possible, so that one does, typically, not need one separate binary
specifically built for each AMD GPU SKU available on the platform.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/sites/Make.frontier-coe
commit cf4e666a7268d83dffcbb8559c1608fbee067b7c
Author: mxxw
Date: Sun Aug 16 18:21:02 2020 -0700
updated Make.machines to include new frontier-coe machines and change… (#1265)
…d 'hostname -f' command to 'hostname'
## Summary
updated Make.machines:
1. added poplar and redwood to frontier-coe machines;
2. changed name string matching for frontier-coe to only
yield exact matches (filter); hence, !="" means exact match;
3. changed `hostname -f` command (which includes the domain) to `hostname` to
avoid false negatives, e.g., not recognizing `hostname -f`==tulip.cm.cluster as tulip;
## Additional background
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/Make.machines
commit bfaa57db45438d795293032609286d5c3146d8a3
Author: mxxw
Date: Sun Aug 16 16:58:39 2020 -0700
fixed bug in matching of COMP_VERSION to correctly treat CCE > 9 (#1269)
## Summary
changed Tools/GNUMake/comps/cray.mak to fix bugs in the matching of $(COMP_VERSION) in order to correctly treat CCE version > 9
## Additional background
Starting with version 9, the Cray CCE C and C++ compilers changed to being clang/LLVM based
which makes it necessary to change some flags compared to CCE versions <=8.
(This change does not apply to the Cray Fortran compiler versions.)
However, the original version check of
COMP_VERSION = $(shell echo $(CRAY_CC_VERSION) | cut -f 1 -d .)
in combination with the tests of "ifeq ($(COMP_VERSION),9)"
differentiated only between V9 and non-V9, which it assumed to be V8.
Howerver, now, we have V10 while V11 is being tested, all of which would
be treated like V8, which is incorrect.
To address this issue the compiler version classification is now done with
ifeq ($(shell expr $(COMP_VERSION) \>= 9), 1)
CCE_GE_V9 := TRUE
else
CCE_GE_V9 := FALSE
endif
and it is tested with
ifeq ($(CCE_GE_V9),TRUE)
## Checklist
The proposed changes:
- [X] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/comps/cray.mak
commit 3896fae0158d1456e238845250bda4264d5980d7
Author: asalmgren
Date: Sun Aug 16 16:17:37 2020 -0700
Change default max_level in AmrCore tutorial from 0 to 2. (#1264)
Tutorials/Amr/Advection_AmrCore/Exec/inputs
commit ac2128cc218ea5ba8124f2c78c808cccf1fab1f9
Author: asalmgren
Date: Fri Aug 14 17:33:32 2020 -0700
Replace amrcore tutorial (#1253)
## Summary
This replaces the existing tutorial with the more general, all C++ one developed for ATPESC 2020. This version works for CPU and GPU (compile-time option), in 2D and 3D (compile-time option), with and without subcycling (run-time option).
A .md file is also included in the directory which walks a new user through how to run this code and what to experiment with to understand AMR.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [x] are described in the proposed changes to the AMReX documentation, if appropriate
Co-authored-by: Weiqun Zhang
Docs/sphinx_tutorials/source/AMR_Tutorial.rst
Src/Base/AMReX_Box.H
Src/Base/AMReX_CoordSys.cpp
Src/Base/AMReX_DistributionMapping.cpp
Tutorials/Amr/Advection_AmrCore/CMakeLists.txt
Tutorials/Amr/Advection_AmrCore/Exec/GNUmakefile
Tutorials/Amr/Advection_AmrCore/Exec/GNUmakefile_movie
Tutorials/Amr/Advection_AmrCore/Exec/Make.Adv
Tutorials/Amr/Advection_AmrCore/Exec/Make.package
Tutorials/Amr/Advection_AmrCore/Exec/Prob.H
Tutorials/Amr/Advection_AmrCore/Exec/SingleVortex/Make.package
Tutorials/Amr/Advection_AmrCore/Exec/SingleVortex/Prob.f90
Tutorials/Amr/Advection_AmrCore/Exec/SingleVortex/face_velocity_2d.f90
Tutorials/Amr/Advection_AmrCore/Exec/SingleVortex/face_velocity_3d.f90
Tutorials/Amr/Advection_AmrCore/Exec/SingleVortex/inputs
Tutorials/Amr/Advection_AmrCore/Exec/inputs
Tutorials/Amr/Advection_AmrCore/Exec/inputs_for_scaling
Tutorials/Amr/Advection_AmrCore/Exec/paraview_amr101.py
Tutorials/Amr/Advection_AmrCore/README
Tutorials/Amr/Advection_AmrCore/README.md
Tutorials/Amr/Advection_AmrCore/Source/AdvancePhiAllLevels.cpp
Tutorials/Amr/Advection_AmrCore/Source/AdvancePhiAtLevel.cpp
Tutorials/Amr/Advection_AmrCore/Source/AmrCoreAdv.H
Tutorials/Amr/Advection_AmrCore/Source/AmrCoreAdv.cpp
Tutorials/Amr/Advection_AmrCore/Source/AmrCoreAdv_F.H
Tutorials/Amr/Advection_AmrCore/Source/DefineVelocity.cpp
Tutorials/Amr/Advection_AmrCore/Source/Kernels.H
Tutorials/Amr/Advection_AmrCore/Source/Make.package
Tutorials/Amr/Advection_AmrCore/Source/Src_2d/Adv_2d.f90
Tutorials/Amr/Advection_AmrCore/Source/Src_2d/Make.package
Tutorials/Amr/Advection_AmrCore/Source/Src_2d/compute_flux_2d.f90
Tutorials/Amr/Advection_AmrCore/Source/Src_2d/slope_2d.f90
Tutorials/Amr/Advection_AmrCore/Source/Src_3d/Adv_3d.f90
Tutorials/Amr/Advection_AmrCore/Source/Src_3d/Make.package
Tutorials/Amr/Advection_AmrCore/Source/Src_3d/compute_flux_3d.f90
Tutorials/Amr/Advection_AmrCore/Source/Src_3d/slope_3d.f90
Tutorials/Amr/Advection_AmrCore/Source/Src_K/Adv_K.H
Tutorials/Amr/Advection_AmrCore/Source/Src_K/Make.package
Tutorials/Amr/Advection_AmrCore/Source/Src_K/compute_flux_2D_K.H
Tutorials/Amr/Advection_AmrCore/Source/Src_K/compute_flux_3D_K.H
Tutorials/Amr/Advection_AmrCore/Source/Src_K/slope_K.H
Tutorials/Amr/Advection_AmrCore/Source/Src_nd/Make.package
Tutorials/Amr/Advection_AmrCore/Source/Src_nd/Tagging_nd.f90
Tutorials/Amr/Advection_AmrCore/Source/Tagging.H
Tutorials/Amr/Advection_AmrCore/Source/bc_fill.H
Tutorials/Amr/Advection_AmrCore/Source/bc_fill_nd.F90
Tutorials/Amr/Advection_AmrCore/Source/face_velocity.H
Tutorials/Amr/Advection_AmrCore/Source/main.cpp
Tutorials/GPU/Advection_AmrCore/CMakeLists.txt
Tutorials/GPU/Advection_AmrCore/Exec/Make.Adv
Tutorials/GPU/Advection_AmrCore/Exec/SingleVortex/GNUmakefile
Tutorials/GPU/Advection_AmrCore/Exec/SingleVortex/Make.package
Tutorials/GPU/Advection_AmrCore/Exec/SingleVortex/Prob.H
Tutorials/GPU/Advection_AmrCore/Exec/SingleVortex/face_velocity.H
Tutorials/GPU/Advection_AmrCore/Exec/SingleVortex/inputs
Tutorials/GPU/Advection_AmrCore/README
Tutorials/GPU/Advection_AmrCore/Source/AmrCoreAdv.H
Tutorials/GPU/Advection_AmrCore/Source/AmrCoreAdv.cpp
Tutorials/GPU/Advection_AmrCore/Source/AmrCoreAdv_F.H
Tutorials/GPU/Advection_AmrCore/Source/Kernels_3d.H
Tutorials/GPU/Advection_AmrCore/Source/Make.package
Tutorials/GPU/Advection_AmrCore/Source/Src_2d/Adv_2d.f90
Tutorials/GPU/Advection_AmrCore/Source/Src_2d/Make.package
Tutorials/GPU/Advection_AmrCore/Source/Src_2d/compute_flux_2d.f90
Tutorials/GPU/Advection_AmrCore/Source/Src_2d/slope_2d.f90
Tutorials/GPU/Advection_AmrCore/Source/Src_3d/Adv_3d.H
Tutorials/GPU/Advection_AmrCore/Source/Src_3d/Make.package
Tutorials/GPU/Advection_AmrCore/Source/Src_3d/compute_flux_3d.H
Tutorials/GPU/Advection_AmrCore/Source/bc_fill.H
Tutorials/GPU/Advection_AmrCore/Source/main.cpp
commit 192fa35ab4195456ba122bce5585e901d51e4b6d
Author: WeiqunZhang
Date: Fri Aug 14 16:12:46 2020 -0700
Rocm36 (#1263)
## Summary
Update make system for hip-clang. Remove some hip workarounds.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_BaseFab.H
Src/Base/AMReX_BaseFabUtility.H
Src/Base/AMReX_Extension.H
Src/Base/AMReX_GpuAsyncArray.H
Src/Base/AMReX_GpuAsyncArray.cpp
Src/Base/AMReX_GpuElixir.cpp
Src/Base/AMReX_GpuError.H
Src/Base/AMReX_GpuLaunch.H
Tools/GNUMake/Make.defs
Tools/GNUMake/comps/hip.mak
commit 6c36b5ffcf2c7fb97a9cf2ee7a2d6951c5dc1064
Author: WeiqunZhang
Date: Fri Aug 14 10:32:50 2020 -0700
Tensor solver and extension of EB domain face (#1262)
## Summary
ParmParse parameter eb2.extend_domain_face is now true by default. The
reason for the change is that for most applications it's hard to imagine one
would want it to be false. It will also solve the EB tensor solver issue
when there is a cut cell just outside the domain abutting a covered valid
cell.
Recent changes to the tensor solvers are incorrect and are reverted here.
They are incorrect because the ghost cells in the modified functions
actually have values at the cell centered. Although the users put domain
face values in the ghost cells, that is not what the solver does. The
solver stores the boundary values in boundary registers. Before the stencil
is being applied, bc function is called to fill the ghost cells with
properly interpolated or extrapolated values so that the stencil operations
are just like working on normal interior cells.
Revert "Correcting tensor cross terms computation for periodic bcs (#1254)"
This reverts commit 1197adc5be6144bae9e362c114ca16b879006180.
Revert "Changing computation at inflow/outflow for tensor solve (#1235)"
This reverts commit b391885e918d63fa1741f954d2792e31426a6204.
Revert "Changing computation at outflow for the EBTensor (#1187)"
This reverts commit 9cd538802ea5b3ab7e6db10d6046933e937aa449.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [x] changes answers in the test suite to more than roundoff level
- [x] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/EB/AMReX_EB2.cpp
Src/EB/AMReX_EB2_Level.H
Src/LinearSolvers/MLMG/AMReX_MLEBTensorOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLEBTensor_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLEBTensor_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLTensorOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLTensor_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLTensor_3D_K.H
commit 5f7a29477108d74834fc0d19ca966e8ea7b6a16f
Author: Andrew Myers
Date: Thu Aug 13 13:25:13 2020 -0700
reuse the tiling machinary in SortParticlesByBin. (#1257)
Src/Particle/AMReX_DenseBins.H
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_ParticleUtil.H
commit 1652a9da291f96dfa8c1f522a6ddf281b2ad252c
Author: asalmgren
Date: Thu Aug 13 10:26:44 2020 -0700
Fix formatting (1261)
## Summary
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Docs/sphinx_documentation/source/Basics.rst
commit 4c6040c0a3f5c96a2f26592d54f0284dfb524b8b
Author: asalmgren
Date: Thu Aug 13 08:34:56 2020 -0700
Add text about how to set default value in the code by passing name (#1260)
of routine to amrex::Initialize(...)
Docs/sphinx_documentation/source/Basics.rst
commit 04d0eecf4a09bcdb584368d89e931fb7a1978580
Author: WeiqunZhang
Date: Wed Aug 12 17:36:04 2020 -0700
nvcc -MM (#1259)
## Summary
`-MM` is not supported for nvcc <= 10.0.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tools/GNUMake/comps/nvcc.mak
commit e915f9e20be15cc070d4aa023474c8cc2f92840c
Author: Andrew Myers
Date: Wed Aug 12 12:16:23 2020 -0700
Domain precision (#1209)
Due to issues with pointing point precision, there are points inside the `ProbDomain()` that do not map to cells inside the `Domain()` box when binned using something like the `getParticleCell` function. This PR adds an additional domain, the "RoundoffDomain()", to the Geometry object. Points inside the roundoff domain are always sure to map to cells inside `Domain()`. The bounds are calculated when the Geometry object is constructed using bisection.
This helps solve a number of issues that can happen when you have particles very close to the domain boundaries.
While I have tested this with Nyx and WarpX, there is a chance that there could be a downstream effect on other codes. If the bisection fails when setting the roundoff domain on some problem, then we may need to adjust some tolerances.
Src/Base/AMReX_Algorithm.H
Src/Base/AMReX_Geometry.H
Src/Base/AMReX_Geometry.cpp
Src/Particle/AMReX_ParticleContainerI.H
commit 19db258fdd76dd5631d8205d8078623c6b97037b
Author: Axel Huebl
Date: Wed Aug 12 10:40:28 2020 -0700
CI: Fix oneAPI Activation (#1208)
now located at `/opt/intel/oneapi/setvars.sh`
Tools/CMake/AMReXSYCL.cmake
Tools/GNUMake/comps/dpcpp.mak
commit 4d4ef3e9b950639f2da9cafe2ee7fa3d572e227c
Author: Andrew Myers
Date: Wed Aug 12 10:27:34 2020 -0700
Give HDF5 benchmark option to read in a set of realistic grids. (#1236)
This should lead to more useful measurements for Castro.
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [x] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Tests/HDF5Benchmark/inputs
Tests/HDF5Benchmark/main.cpp
commit d89d48026e077f251408c2652c6ee2e5c5f80d51
Author: Mu-Hua Chien
Date: Wed Aug 12 13:26:02 2020 -0400
fix semicoarsening bugs for nodal solver (#1255)
## Summary
Fix a semicoarsening bug for nodal solver.
## Additional background
In avgdown coefficient, there is a place use coarsen ratio instead of the mg_coarsen_ratio_vec. Also, build the semi_avgdown for nodal solver.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_1D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLaplacian.cpp
commit 1628d314b33143c6747bd2b44a99426bd1ac3d09
Author: kngott <30483578+kngott@users.noreply.github.com>
Date: Wed Aug 12 10:22:46 2020 -0700
Fixes for Amrvis warnings from gcc 7.5 with DEBUG=TRUE, DIM=2 and DIM=3. (#1256)
## Summary
Warnings from compiling Amrvis using GCC 7.5 with DEBUG=TRUE, DIM=2 and DIM=3. (Build on dogora). Primarily ignore_unused, commenting out function parameter declarations and a few manual type casts. Some unique warning adjustments as well (e.g. a shadowed BL_PROFILER).
To be paired with a matching PR in the Amrvis repo.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_BLProfiler.cpp
Src/Base/AMReX_BoxList.cpp
Src/Extern/ProfParser/AMReX_BLProfStats.H
Src/Extern/ProfParser/AMReX_BLProfStats.cpp
Src/Extern/ProfParser/AMReX_BLProfUtilities.cpp
Src/Extern/ProfParser/AMReX_CommProfStats.H
Src/Extern/ProfParser/AMReX_CommProfStats.cpp
Src/Extern/ProfParser/AMReX_ProfParserBatch.cpp
Src/Extern/ProfParser/AMReX_RegionsProfStats.H
Src/Extern/ProfParser/AMReX_RegionsProfStats.cpp
Src/Extern/ProfParser/BLProfParser.y
Src/Extern/amrdata/AMReX_AmrData.cpp
Src/Extern/amrdata/AMReX_DataServices.cpp
commit 1197adc5be6144bae9e362c114ca16b879006180
Author: OscarAntepara <52221614+OscarAntepara@users.noreply.github.com>
Date: Tue Aug 11 14:02:22 2020 -0700
Correcting tensor cross terms computation for periodic bcs (#1254)
## Summary
Correcting tensor cross terms computation for periodic bcs
## Additional background
Fixing a bug introduced in PR #1235. Now it computes in the correct way the cross terms in the tensor solve for problems with periodic bcs.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [x] changes answers in the test suite to more than roundoff level
- [x] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/LinearSolvers/MLMG/AMReX_MLEBTensorOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLEBTensor_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLEBTensor_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLTensorOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLTensor_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLTensor_3D_K.H
commit a4ab677faf1279d20d7feb6f42263bcc7b1468f4
Author: WeiqunZhang
Date: Mon Aug 10 17:35:07 2020 -0700
Fix bug in FPinfo for EB (#1241)
PR #1224 introduced a bug that affects FillPatch on nodal grids for EB.
When there is only one ghost cell, the fine patch grids might be degenerate
in the sense that some boxes are only 1 node wide. In that case, we cannot
build cell-centered EBCellFlags unless there are ghost cells.
Src/Base/AMReX_FabArrayBase.cpp
commit 46b75867f5106fa363563a230e64a078a3cacc00
Author: Andrew Myers
Date: Mon Aug 10 14:00:15 2020 -0700
Fix curly brace warning with clang (#1251)
Src/Base/AMReX_Machine.cpp
commit a1c997e6ca9d3ad58b18881ef9f319334a2db161
Author: Andrew Myers
Date: Mon Aug 10 14:00:05 2020 -0700
fix some warnings that arise in the redistribute test (#1250)
Tests/Particles/Redistribute/main.cpp
commit c99e8ac4af561a960b98e7995d8fbb291aa70865
Author: Andrew Myers
Date: Mon Aug 10 13:49:52 2020 -0700
Remove _rt and _prt from the particle code, for reasons explained in PR 1243. (#1249)
Src/Particle/AMReX_Particle.H
Src/Particle/AMReX_Particle_mod_K.H
Src/Particle/AMReX_TracerParticle_mod_K.H
Src/Particle/AMReX_TracerParticles.cpp
commit 480d1fffe7c7ff7bddcca356ce656f1cecd7f308
Author: WeiqunZhang
Date: Mon Aug 10 13:41:07 2020 -0700
Remove Perilla (#1248)
Remove Perilla because it's incompatible with our GPU strategy and it's no
longer being maintained.
Docs/Doxygen/doxygen.conf
Docs/sphinx_documentation/make_api.py
Docs/sphinx_documentation/source/AsyncIter.rst
Docs/sphinx_documentation/source/AsyncIter_Chapter.rst
Docs/sphinx_documentation/source/index.rst
Src/Amr/AMReX_Amr.H
Src/Amr/AMReX_Amr.cpp
Src/Amr/AMReX_AmrLevel.H
Src/Amr/AMReX_AmrLevel.cpp
Src/Amr/AMReX_AsyncFillPatch.cpp
Src/Amr/Make.package
Src/AmrTask/AMFIter/AMReX_AmrTask.H
Src/AmrTask/AMFIter/AMReX_AsyncMFIter.H
Src/AmrTask/AMFIter/AMReX_AsyncMFIter.cpp
Src/AmrTask/AMFIter/AMReX_Connections.H
Src/AmrTask/AMFIter/Makefile
Src/AmrTask/Amr/AMReX_AmrLevelTask.H
Src/AmrTask/Amr/AMReX_AmrLevelTask.cpp
Src/AmrTask/Amr/AMReX_AmrTask.cpp
Src/AmrTask/Amr/Make.package
Src/AmrTask/Amr/Makefile
Src/AmrTask/AmrCore/AMReX_FillPatchUtil.H
Src/AmrTask/AmrCore/AMReX_FillPatchUtil.cpp
Src/AmrTask/Makefile
Src/AmrTask/arch.common
Src/AmrTask/arch/arch.mpi.generic
Src/AmrTask/arch/arch.serial
Src/AmrTask/graph/AMReX_AbstractTask.H
Src/AmrTask/graph/AMReX_AbstractTask.cpp
Src/AmrTask/graph/AMReX_Affinity.H
Src/AmrTask/graph/AMReX_Affinity.cpp
Src/AmrTask/graph/AMReX_DataTypes.H
Src/AmrTask/graph/AMReX_TaskGraph.H
Src/AmrTask/graph/AMReX_TaskGraph.cpp
Src/AmrTask/graph/Makefile
Src/AmrTask/graph/RTS.H
Src/AmrTask/make_defaults/Cori
Src/AmrTask/make_defaults/Edison
Src/AmrTask/make_defaults/Summit-dev
Src/AmrTask/rts_impls/MPI_Generic/Makefile
Src/AmrTask/rts_impls/MPI_Generic/README
Src/AmrTask/rts_impls/MPI_Generic/mylock.h
Src/AmrTask/rts_impls/MPI_Generic/rts.C
Src/AmrTask/rts_impls/MPI_Generic/rts_graphimpl.H
Src/AmrTask/rts_impls/MPI_Generic/rts_taskimpl.H
Src/AmrTask/rts_impls/README
Src/AmrTask/rts_impls/Serial/Makefile
Src/AmrTask/rts_impls/Serial/rts.C
Src/AmrTask/rts_impls/Serial/rts_graphimpl.H
Src/AmrTask/rts_impls/Serial/rts_taskimpl.H
Src/AmrTask/rts_impls/Utils/dl_malloc.c
Src/AmrTask/rts_impls/Utils/dl_malloc.h
Src/AmrTask/rts_impls/Utils/sysInfo.C
Src/AmrTask/rts_impls/Utils/sysInfo.H
Src/AmrTask/rts_impls/mpi/Make.package
Src/AmrTask/rts_impls/mpi/PackageQueue.H
Src/AmrTask/rts_impls/mpi/PackageQueue.cpp
Src/AmrTask/rts_impls/mpi/Perilla.H
Src/AmrTask/rts_impls/mpi/Perilla.cpp
Src/AmrTask/rts_impls/mpi/PerillaConfig.H
Src/AmrTask/rts_impls/mpi/PerillaRts.H
Src/AmrTask/rts_impls/mpi/PerillaRts.cpp
Src/AmrTask/rts_impls/mpi/perilla.mak
Src/AmrTask/rts_impls/mpi_omp/AsyncMultiFabUtil.H
Src/AmrTask/rts_impls/mpi_omp/AsyncMultiFabUtil.cpp
Src/AmrTask/rts_impls/mpi_omp/Barrier.H
Src/AmrTask/rts_impls/mpi_omp/Barrier.cpp
Src/AmrTask/rts_impls/mpi_omp/LocalConnection.H
Src/AmrTask/rts_impls/mpi_omp/Make.package
Src/AmrTask/rts_impls/mpi_omp/PackageQueue.H
Src/AmrTask/rts_impls/mpi_omp/PackageQueue.cpp
Src/AmrTask/rts_impls/mpi_omp/Perilla.H
Src/AmrTask/rts_impls/mpi_omp/Perilla.cpp
Src/AmrTask/rts_impls/mpi_omp/PerillaConfig.H
Src/AmrTask/rts_impls/mpi_omp/PerillaRts.H
Src/AmrTask/rts_impls/mpi_omp/PerillaRts.cpp
Src/AmrTask/rts_impls/mpi_omp/RGIter.H
Src/AmrTask/rts_impls/mpi_omp/RGIter.cpp
Src/AmrTask/rts_impls/mpi_omp/RegionGraph.H
Src/AmrTask/rts_impls/mpi_omp/RegionGraph.cpp
Src/AmrTask/rts_impls/mpi_omp/RegionQueue.H
Src/AmrTask/rts_impls/mpi_omp/RegionQueue.cpp
Src/AmrTask/rts_impls/mpi_omp/RemoteConnection.H
Src/AmrTask/rts_impls/mpi_omp/WorkerThread.H
Src/AmrTask/rts_impls/mpi_omp/WorkerThread.cpp
Src/AmrTask/rts_impls/mpi_omp/perilla.mak
Src/AmrTask/rts_impls/runtime_common/AsyncMultiFabUtil.H
Src/AmrTask/rts_impls/runtime_common/AsyncMultiFabUtil.cpp
Src/AmrTask/rts_impls/runtime_common/Barrier.H
Src/AmrTask/rts_impls/runtime_common/Barrier.cpp
Src/AmrTask/rts_impls/runtime_common/LocalConnection.H
Src/AmrTask/rts_impls/runtime_common/Make.package
Src/AmrTask/rts_impls/runtime_common/PerillaMemCheck.H
Src/AmrTask/rts_impls/runtime_common/PerillaMemCheck.cpp
Src/AmrTask/rts_impls/runtime_common/Perilla_common.cpp
Src/AmrTask/rts_impls/runtime_common/RGIter.H
Src/AmrTask/rts_impls/runtime_common/RGIter.cpp
Src/AmrTask/rts_impls/runtime_common/RegionGraph.H
Src/AmrTask/rts_impls/runtime_common/RegionGraph.cpp
Src/AmrTask/rts_impls/runtime_common/RegionQueue.H
Src/AmrTask/rts_impls/runtime_common/RegionQueue.cpp
Src/AmrTask/rts_impls/runtime_common/RemoteConnection.H
Src/AmrTask/rts_impls/runtime_common/WorkerThread.H
Src/AmrTask/rts_impls/runtime_common/WorkerThread.cpp
Src/AmrTask/rts_impls/runtime_common/mylock.h
Src/AmrTask/rts_impls/runtime_common/perilla.mak
Src/AmrTask/rts_impls/upcxx/Make.package
Src/AmrTask/rts_impls/upcxx/PackageQueue.H
Src/AmrTask/rts_impls/upcxx/PackageQueue.cpp
Src/AmrTask/rts_impls/upcxx/Perilla.H
Src/AmrTask/rts_impls/upcxx/Perilla.cpp
Src/AmrTask/rts_impls/upcxx/PerillaConfig.H
Src/AmrTask/rts_impls/upcxx/PerillaRts.H
Src/AmrTask/rts_impls/upcxx/PerillaRts.cpp
Src/AmrTask/rts_impls/upcxx/perilla.mak
Src/AmrTask/todolist
Src/Base/AMReX.cpp
Src/Base/AMReX_BaseFab.H
Src/Base/AMReX_FabArray.H
Src/Base/AMReX_FabArrayBase.H
Src/Base/AMReX_MemPool.cpp
Src/Base/AMReX_MultiFab.H
Src/Base/AMReX_MultiFab.cpp
Src/Base/AMReX_MultiFabUtil_Perilla.H
Src/Base/AMReX_MultiFabUtil_Perilla.cpp
Src/Base/Make.package
Tools/CMake/AMReX_Defines.cmake
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/Make.Adv.mpi
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/Make.Adv.upcxx
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/SingleVortex/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/SingleVortex/Make.package
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/SingleVortex/Prob.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/SingleVortex/face_velocity_2d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/SingleVortex/face_velocity_3d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/SingleVortex/inputs
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/SingleVortex/probin
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/UniformVelocity/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/UniformVelocity/Make.package
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/UniformVelocity/Prob.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/UniformVelocity/face_velocity_2d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/UniformVelocity/inputs
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/UniformVelocity/probdata.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/UniformVelocity/probin
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/README
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Adv.H
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Adv.cpp
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/AdvBld.cpp
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Adv_F.H
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Adv_advance.cpp
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Adv_dt.cpp
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Adv_io.cpp
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Adv_setup.cpp
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Make.package
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_2d/Adv_2d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_2d/Make.package
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_2d/compute_flux_2d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_2d/slope_2d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_3d/Adv_3d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_3d/Make.package
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_3d/compute_flux_3d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_3d/slope_3d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_nd/Adv_nd.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_nd/Make.package
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_nd/Tagging_nd.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_nd/tagging_params.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/main.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/Make.Adv.mpi
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/Make.Adv.upcxx
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/SingleVortex/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/SingleVortex/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/SingleVortex/Prob.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/SingleVortex/face_velocity_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/SingleVortex/face_velocity_3d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/SingleVortex/inputs
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/SingleVortex/probin
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/UniformVelocity/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/UniformVelocity/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/UniformVelocity/Prob.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/UniformVelocity/face_velocity_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/UniformVelocity/inputs
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/UniformVelocity/probdata.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/UniformVelocity/probin
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/README
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Adv.H
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Adv.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/AdvBld.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Adv_F.H
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Adv_advance.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Adv_dt.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Adv_io.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Adv_setup.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_2d/Adv_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_2d/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_2d/compute_flux_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_2d/slope_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_3d/Adv_3d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_3d/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_3d/compute_flux_3d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_3d/slope_3d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_nd/Adv_nd.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_nd/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_nd/Tagging_nd.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_nd/tagging_params.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/main.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/Make.Adv.mpi
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/Make.Adv.upcxx
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/SingleVortex/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/SingleVortex/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/SingleVortex/Prob.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/SingleVortex/face_velocity_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/SingleVortex/face_velocity_3d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/SingleVortex/inputs
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/SingleVortex/probin
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/UniformVelocity/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/UniformVelocity/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/UniformVelocity/Prob.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/UniformVelocity/face_velocity_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/UniformVelocity/inputs
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/UniformVelocity/probdata.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/UniformVelocity/probin
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/README
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Adv.H
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Adv.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/AdvBld.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Adv_F.H
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Adv_advance.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Adv_dt.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Adv_io.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Adv_setup.cpp
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_2d/Adv_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_2d/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_2d/compute_flux_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_2d/slope_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_3d/Adv_3d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_3d/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_3d/compute_flux_3d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_3d/slope_3d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_nd/Adv_nd.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_nd/Make.package
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_nd/Tagging_nd.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_nd/tagging_params.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/main.cpp
commit b391885e918d63fa1741f954d2792e31426a6204
Author: OscarAntepara <52221614+OscarAntepara@users.noreply.github.com>
Date: Mon Aug 10 11:17:32 2020 -0700
Changing computation at inflow/outflow for tensor solve (#1235)
## Summary
Computation at inflow boundaries for the MLEBTensor and inflow/outflow for MLTensor has been changed.
## Additional background
At outflow the tensor solve uses extrapolated values from the interior cell to the outflow face. At inflow, the tensor solve uses boundary data, evaluated at the face, that has been stored in the ghost cell next to the inflow face.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [x] changes answers in the test suite to more than roundoff level
- [x] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/LinearSolvers/MLMG/AMReX_MLEBTensorOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLEBTensor_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLEBTensor_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLTensorOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLTensor_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLTensor_3D_K.H
commit ad1b75d9557fddf126f67be4b802327c1a4415e5
Author: asalmgren
Date: Mon Aug 10 10:46:17 2020 -0700
Remove sensei tutorials (#1247)
* Remove SENSEI tutorials.
* Update tutorial documentation since we no longer have SENSEI tutorials
Docs/sphinx_tutorials/source/SENSEI_Tutorial.rst
Docs/sphinx_tutorials/source/index.rst
Tutorials/SENSEI/Advection_AmrCore/CMakeLists.txt
Tutorials/SENSEI/Advection_AmrCore/Exec/Make.Adv
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/GNUmakefile
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/Make.package
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/Prob.f90
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/face_velocity_2d.f90
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/face_velocity_3d.f90
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/inputs
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/histogram.py
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/histogram_python.xml
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/render_catalyst.py
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/render_iso_catalyst_2d.py
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/render_iso_catalyst_2d.xml
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/render_iso_catalyst_3d.py
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/render_iso_catalyst_3d.xml
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/render_iso_libsim_2d.session
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/render_iso_libsim_2d.xml
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/render_iso_libsim_3d.session
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/render_iso_libsim_3d.xml
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/sensei/write_vtk.xml
Tutorials/SENSEI/Advection_AmrCore/README
Tutorials/SENSEI/Advection_AmrCore/README_SENSEI.md
Tutorials/SENSEI/Advection_AmrCore/Source/AmrCoreAdv.H
Tutorials/SENSEI/Advection_AmrCore/Source/AmrCoreAdv.cpp
Tutorials/SENSEI/Advection_AmrCore/Source/AmrCoreAdv_F.H
Tutorials/SENSEI/Advection_AmrCore/Source/Make.package
Tutorials/SENSEI/Advection_AmrCore/Source/Src_2d/Adv_2d.f90
Tutorials/SENSEI/Advection_AmrCore/Source/Src_2d/Make.package
Tutorials/SENSEI/Advection_AmrCore/Source/Src_2d/compute_flux_2d.f90
Tutorials/SENSEI/Advection_AmrCore/Source/Src_2d/slope_2d.f90
Tutorials/SENSEI/Advection_AmrCore/Source/Src_3d/Adv_3d.f90
Tutorials/SENSEI/Advection_AmrCore/Source/Src_3d/Make.package
Tutorials/SENSEI/Advection_AmrCore/Source/Src_3d/compute_flux_3d.f90
Tutorials/SENSEI/Advection_AmrCore/Source/Src_3d/slope_3d.f90
Tutorials/SENSEI/Advection_AmrCore/Source/Src_nd/Make.package
Tutorials/SENSEI/Advection_AmrCore/Source/Src_nd/Tagging_nd.f90
Tutorials/SENSEI/Advection_AmrCore/Source/bc_fill_nd.F90
Tutorials/SENSEI/Advection_AmrCore/Source/main.cpp
Tutorials/SENSEI/Advection_AmrLevel/CMakeLists.txt
Tutorials/SENSEI/Advection_AmrLevel/Exec/Make.Adv
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/GNUmakefile
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/Make.package
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/Prob.f90
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/face_velocity_2d.f90
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/face_velocity_3d.f90
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/inputs
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/inputs.tracers
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/probin
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/histogram.py
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/histogram_python.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/iso_extract.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/read_adios2_bp4_default.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/read_adios2_sst_default.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/render_catalyst.py
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/render_iso_catalyst_2d.py
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/render_iso_catalyst_2d.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/render_iso_catalyst_3d.py
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/render_iso_catalyst_3d.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/render_iso_libsim_2d.session
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/render_iso_libsim_2d.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/render_iso_libsim_3d.session
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/render_iso_libsim_3d.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/render_libsim.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/write_adios2_bp4.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/write_adios2_sst.xml
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/sensei/write_vtk.xml
Tutorials/SENSEI/Advection_AmrLevel/README
Tutorials/SENSEI/Advection_AmrLevel/README_SENSEI.md
Tutorials/SENSEI/Advection_AmrLevel/Source/Adv_F.H
Tutorials/SENSEI/Advection_AmrLevel/Source/AmrLevelAdv.H
Tutorials/SENSEI/Advection_AmrLevel/Source/AmrLevelAdv.cpp
Tutorials/SENSEI/Advection_AmrLevel/Source/LevelBldAdv.cpp
Tutorials/SENSEI/Advection_AmrLevel/Source/Make.package
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_2d/Adv_2d.f90
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_2d/Make.package
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_2d/compute_flux_2d.f90
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_2d/slope_2d.f90
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_3d/Adv_3d.f90
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_3d/Make.package
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_3d/compute_flux_3d.f90
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_3d/slope_3d.f90
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_nd/Adv_nd.f90
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_nd/Make.package
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_nd/Tagging_nd.f90
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_nd/tagging_params.f90
Tutorials/SENSEI/Advection_AmrLevel/Source/main.cpp
Tutorials/SENSEI/README.md
commit 63d386514d5167d5849d11e0cc6dbb146c46a37b
Author: WeiqunZhang
Date: Sun Aug 9 20:22:25 2020 -0700
oneAPI Beta 8 (#1243)
In beta8, C++11 user defined literals are broken because of the use of long
double. We have implemented a workaround by overloading operator "" _rt
with `const char*` argument and our own `atof`. But, because Clang does not
evaluate the new version of the operator at compile time, we now avoid using
`_rt` in AMReX. In the long term, we may have to remove it entirely if the
long double issue is not resolved in DPC++.
With the workaround, we can enable the DPC++ CI test again.
.github/workflows/linux.yml
Src/Amr/AMReX_Amr.cpp
Src/Amr/AMReX_extrapolater_2D_K.H
Src/Amr/AMReX_extrapolater_3D_K.H
Src/AmrCore/AMReX_AmrCore.cpp
Src/AmrCore/AMReX_ErrorList.cpp
Src/AmrCore/AMReX_Interp_1D_C.H
Src/AmrCore/AMReX_Interp_2D_C.H
Src/AmrCore/AMReX_Interp_3D_C.H
Src/AmrTask/Amr/AMReX_AmrTask.cpp
Src/Base/AMReX_Dim3.H
Src/Base/AMReX_FArrayBox.cpp
Src/Base/AMReX_GpuReduce.H
Src/Base/AMReX_MultiFab.cpp
Src/Base/AMReX_MultiFabUtil_nd_C.H
Src/Base/AMReX_REAL.H
Src/Base/AMReX_Random.cpp
Src/Base/AMReX_Scan.H
Src/Boundary/AMReX_InterpBndryData_2D_K.H
Src/Boundary/AMReX_InterpBndryData_3D_K.H
Src/EB/AMReX_algoim_K.H
Src/LinearSolvers/MLMG/AMReX_MLCGSolver.cpp
Src/LinearSolvers/MLMG/AMReX_MLEBABecLap_K.H
Src/LinearSolvers/MLMG/AMReX_MLLinOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLLinOp_K.H
Src/LinearSolvers/MLMG/AMReX_MLMG.cpp
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_1D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeTensorLap_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeTensorLap_3D_K.H
Tutorials/Basic/HeatEquation_EX1_C/Source/main.cpp
Tutorials/Basic/HeatEquation_EX1_C/Source/mykernel.H
Tutorials/Basic/HeatEquation_EX1_CF/Source/main.cpp
Tutorials/Basic/HeatEquation_EX2_C/Source/main.cpp
Tutorials/Basic/HeatEquation_EX2_CF/Source/main.cpp
Tutorials/Basic/HeatEquation_EX3_C/Source/main.cpp
Tutorials/GPU/CNS/Exec/RT/cns_prob.H
Tutorials/GPU/CNS/Exec/RT/cns_prob_parm.H
Tutorials/GPU/CNS/Exec/Sod/cns_prob.H
Tutorials/GPU/CNS/Exec/Sod/cns_prob_parm.H
Tutorials/GPU/CNS/Source/CNS.cpp
Tutorials/GPU/CNS/Source/CNS_K.H
Tutorials/GPU/CNS/Source/CNS_advance.cpp
Tutorials/GPU/CNS/Source/CNS_parm.H
Tutorials/GPU/CNS/Source/CNS_parm.cpp
Tutorials/GPU/CNS/Source/hydro/CNS_hydro_K.H
Tutorials/GPU/CNS/Source/main.cpp
commit e3f25aa6b9ed74229d203110fa59fb2d0e815bdf
Author: WeiqunZhang
Date: Sun Aug 9 20:21:44 2020 -0700
Fix a bug in Tutorials/GPU/CNS (#1245)
Fix a bug in deriving velocity and a number of warnings for
Tutorials/GPU/CNS.
Src/Base/AMReX_FBI.H
Tutorials/GPU/CNS/Exec/RT/cns_prob.cpp
Tutorials/GPU/CNS/Exec/Sod/cns_prob.cpp
Tutorials/GPU/CNS/Source/CNS.cpp
Tutorials/GPU/CNS/Source/CNS_advance.cpp
Tutorials/GPU/CNS/Source/CNS_derive.cpp
commit cd0a8034f389b40882900b3faa92d84beed13201
Author: WeiqunZhang
Date: Sun Aug 9 20:21:04 2020 -0700
FPE trap in DPC++ (#1244)
Due to a bug in DPC++ compiler, we disable fpe trapping.
Src/Base/AMReX.cpp
commit a2eecef9c2efab1548ad14d6b5083f14451441bb
Author: drangara <69211175+drangara@users.noreply.github.com>
Date: Sun Aug 9 16:34:44 2020 -0400
Small volfrac (#1228)
## Summary
When extend_domain_face is set to true we do the following: If we set a cell that lies on one of the 6 faces of the domain as covered because its volfrac is less than the small_volfrac , we are forcing the corresponding ghost cells on that face to also be covered.
## Additional background
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [x] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/EB/AMReX_EB2_2D_C.cpp
Src/EB/AMReX_EB2_3D_C.cpp
Src/EB/AMReX_EB2_C.H
Src/EB/AMReX_EB2_Level.H
commit d8d270056e953904c04d3ae3221006389128531e
Author: Jon Rood
Date: Sun Aug 9 10:30:38 2020 -0600
Remove error message for unknown MPI in Make.nrel since we want to be able to use MPT for MPI. (#1242)
Tools/GNUMake/sites/Make.nrel
commit 09db04bf308dbfe797baa19b9118466f545ee16c
Author: Andrew Myers
Date: Fri Aug 7 16:39:17 2020 -0700
fix some warnings that arise when USE_SINGLE_PRECISION_PARTICLES=TRUE but PRECISION=DOUBLE (#1240)
Src/F_Interfaces/Particle/AMReX_particlecontainer_fi.cpp
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_ParticleUtil.H
Src/Particle/AMReX_TracerParticle_mod_K.H
Src/Particle/AMReX_TracerParticles.cpp
commit 881397d57f8fee10dc24db747e07a7bb3685d4ec
Author: WeiqunZhang
Date: Fri Aug 7 16:23:35 2020 -0700
GNU Make dependency (#1239)
Use `-MM` instead of `-M` to generate dependency in GNU Make system, because
we do not need to include dependency on headers in the system header
directories.
Tools/GNUMake/Make.defs
commit b4ba4429672be4589d535b7e6c83272fee5fdfe3
Author: WeiqunZhang
Date: Fri Aug 7 13:51:41 2020 -0700
Oversetmask (#1232)
Switch the overset mask convention to that 1 means unknown and 0 means
known. The new convention is more convenient for AMR-Wind.
Src/LinearSolvers/MLMG/AMReX_MLABecLap_1D_K.H
Src/LinearSolvers/MLMG/AMReX_MLABecLap_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLABecLap_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLABecLaplacian.H
Src/LinearSolvers/MLMG/AMReX_MLABecLaplacian.cpp
Src/LinearSolvers/MLMG/AMReX_MLLinOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLNodeLinOp.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLinOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLTensorOp.H
Src/LinearSolvers/MLMG/AMReX_MLTensor_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLTensor_3D_K.H
Tests/LinearSolvers/CellOverset/MyTest.cpp
Tests/LinearSolvers/NodalOverset/MyTest.cpp
Tests/LinearSolvers/TensorOverset/MyTest.cpp
commit cd71ce08c17edb1aeeaa89d00bb9b072713c26a8
Author: WeiqunZhang
Date: Fri Aug 7 13:40:39 2020 -0700
Remove old solvers (#1238)
Src/LinearSolvers/C_CellMG and C_TensorMG are removed. They have been
superseded by Src/LinearSolvers/MLMG.
Src/LinearSolvers/C_CellMG/AMReX_ABecLaplacian.H
Src/LinearSolvers/C_CellMG/AMReX_ABecLaplacian.cpp
Src/LinearSolvers/C_CellMG/AMReX_ABec_1D.F90
Src/LinearSolvers/C_CellMG/AMReX_ABec_2D.F90
Src/LinearSolvers/C_CellMG/AMReX_ABec_3D.F90
Src/LinearSolvers/C_CellMG/AMReX_ABec_F.H
Src/LinearSolvers/C_CellMG/AMReX_ABec_UTIL.F90
Src/LinearSolvers/C_CellMG/AMReX_CGSolver.H
Src/LinearSolvers/C_CellMG/AMReX_CGSolver.cpp
Src/LinearSolvers/C_CellMG/AMReX_LO_1D.F90
Src/LinearSolvers/C_CellMG/AMReX_LO_2D.F90
Src/LinearSolvers/C_CellMG/AMReX_LO_3D.F90
Src/LinearSolvers/C_CellMG/AMReX_LO_F.H
Src/LinearSolvers/C_CellMG/AMReX_LP_1D.F90
Src/LinearSolvers/C_CellMG/AMReX_LP_2D.F90
Src/LinearSolvers/C_CellMG/AMReX_LP_3D.F90
Src/LinearSolvers/C_CellMG/AMReX_LP_F.H
Src/LinearSolvers/C_CellMG/AMReX_Laplacian.H
Src/LinearSolvers/C_CellMG/AMReX_Laplacian.cpp
Src/LinearSolvers/C_CellMG/AMReX_LinOp.H
Src/LinearSolvers/C_CellMG/AMReX_LinOp.cpp
Src/LinearSolvers/C_CellMG/AMReX_MG_1D.F90
Src/LinearSolvers/C_CellMG/AMReX_MG_2D.F90
Src/LinearSolvers/C_CellMG/AMReX_MG_3D.F90
Src/LinearSolvers/C_CellMG/AMReX_MG_F.H
Src/LinearSolvers/C_CellMG/AMReX_MultiGrid.H
Src/LinearSolvers/C_CellMG/AMReX_MultiGrid.cpp
Src/LinearSolvers/C_CellMG/AMReX_lo_bctypes.fi
Src/LinearSolvers/C_CellMG/Make.package
Src/LinearSolvers/C_CellMG/OpenSource.txt
Src/LinearSolvers/C_CellMG4/AMReX_ABec2.H
Src/LinearSolvers/C_CellMG4/AMReX_ABec2.cpp
Src/LinearSolvers/C_CellMG4/AMReX_ABec2_2D.F90
Src/LinearSolvers/C_CellMG4/AMReX_ABec2_3D.F90
Src/LinearSolvers/C_CellMG4/AMReX_ABec2_F.H
Src/LinearSolvers/C_CellMG4/AMReX_ABec4.H
Src/LinearSolvers/C_CellMG4/AMReX_ABec4.cpp
Src/LinearSolvers/C_CellMG4/AMReX_ABec4_2D.F90
Src/LinearSolvers/C_CellMG4/AMReX_ABec4_3D.F90
Src/LinearSolvers/C_CellMG4/AMReX_ABec4_F.H
Src/LinearSolvers/C_CellMG4/Make.package
Src/LinearSolvers/C_CellMG4/OpenSource.txt
Src/LinearSolvers/C_TensorMG/AMReX_DV_2D.F
Src/LinearSolvers/C_TensorMG/AMReX_DV_3D1.F
Src/LinearSolvers/C_TensorMG/AMReX_DV_3D2.F
Src/LinearSolvers/C_TensorMG/AMReX_DV_3D3.F
Src/LinearSolvers/C_TensorMG/AMReX_DivVis.H
Src/LinearSolvers/C_TensorMG/AMReX_DivVis.cpp
Src/LinearSolvers/C_TensorMG/AMReX_DivVis_F.H
Src/LinearSolvers/C_TensorMG/AMReX_MCCGSolver.H
Src/LinearSolvers/C_TensorMG/AMReX_MCCGSolver.cpp
Src/LinearSolvers/C_TensorMG/AMReX_MCINTERPBNDRYDATA_2D.F
Src/LinearSolvers/C_TensorMG/AMReX_MCINTERPBNDRYDATA_3D.F
Src/LinearSolvers/C_TensorMG/AMReX_MCINTERPBNDRYDATA_F.H
Src/LinearSolvers/C_TensorMG/AMReX_MCInterpBndryData.H
Src/LinearSolvers/C_TensorMG/AMReX_MCInterpBndryData.cpp
Src/LinearSolvers/C_TensorMG/AMReX_MCLO_2D.F
Src/LinearSolvers/C_TensorMG/AMReX_MCLO_3D.F
Src/LinearSolvers/C_TensorMG/AMReX_MCLO_F.H
Src/LinearSolvers/C_TensorMG/AMReX_MCLinOp.H
Src/LinearSolvers/C_TensorMG/AMReX_MCLinOp.cpp
Src/LinearSolvers/C_TensorMG/AMReX_MCMultiGrid.H
Src/LinearSolvers/C_TensorMG/AMReX_MCMultiGrid.cpp
Src/LinearSolvers/C_TensorMG/DV_2D.mF
Src/LinearSolvers/C_TensorMG/DV_3D1.mF
Src/LinearSolvers/C_TensorMG/DV_3D2.mF
Src/LinearSolvers/C_TensorMG/DV_3D3.mF
Src/LinearSolvers/C_TensorMG/DV_3D4.mF
Src/LinearSolvers/C_TensorMG/Format.m
Src/LinearSolvers/C_TensorMG/Make.package
Src/LinearSolvers/C_TensorMG/OpenSource.txt
Src/LinearSolvers/C_TensorMG/Optimize.m
Src/LinearSolvers/C_TensorMG/amrex_tmg_util.F90
Src/LinearSolvers/C_TensorMG/visc2d.m
Src/LinearSolvers/C_TensorMG/visc2d.ma
Src/LinearSolvers/C_TensorMG/visc2d.nb
Src/LinearSolvers/C_TensorMG/visc3d.m
Src/LinearSolvers/C_TensorMG/visc3d.ma
Src/LinearSolvers/C_TensorMG/visc3d.nb
Tests/LinearSolvers/C_CellMG/COEF_2D.F
Tests/LinearSolvers/C_CellMG/COEF_3D.F
Tests/LinearSolvers/C_CellMG/COEF_F.H
Tests/LinearSolvers/C_CellMG/GNUmakefile
Tests/LinearSolvers/C_CellMG/GNUmakefile.dumpi
Tests/LinearSolvers/C_CellMG/MACOPERATOR_2D.F
Tests/LinearSolvers/C_CellMG/MACOPERATOR_3D.F
Tests/LinearSolvers/C_CellMG/MACOPERATOR_F.H
Tests/LinearSolvers/C_CellMG/MACPROJ_2D.F
Tests/LinearSolvers/C_CellMG/MACPROJ_3D.F
Tests/LinearSolvers/C_CellMG/MacOpMacDrivers.H
Tests/LinearSolvers/C_CellMG/MacOperator.H
Tests/LinearSolvers/C_CellMG/MacOperator.cpp
Tests/LinearSolvers/C_CellMG/Make.package
Tests/LinearSolvers/C_CellMG/Palette
Tests/LinearSolvers/C_CellMG/amrvis.defaults
Tests/LinearSolvers/C_CellMG/dumpi/NOTE
Tests/LinearSolvers/C_CellMG/dumpi/inputs.3d.4096core
Tests/LinearSolvers/C_CellMG/dumpi/inputs.3d.512core
Tests/LinearSolvers/C_CellMG/dumpi/inputs.3d.64core
Tests/LinearSolvers/C_CellMG/dumpi/inputs.3d.8core
Tests/LinearSolvers/C_CellMG/inputs.2d
Tests/LinearSolvers/C_CellMG/inputs.3d
Tests/LinearSolvers/C_CellMG/macprojTest.cpp
Tests/LinearSolvers/C_CellMG/main.cpp
Tests/LinearSolvers/C_CellMG/vpramps.dat
Tests/LinearSolvers/C_TensorMG/GNUmakefile
Tests/LinearSolvers/C_TensorMG/Make.package
Tests/LinearSolvers/C_TensorMG/Palette
Tests/LinearSolvers/C_TensorMG/TestMCViscBndry.H
Tests/LinearSolvers/C_TensorMG/TestMCViscBndry.cpp
Tests/LinearSolvers/C_TensorMG/amrvis.defaults
Tests/LinearSolvers/C_TensorMG/inputs
Tests/LinearSolvers/C_TensorMG/inputs2D
Tests/LinearSolvers/C_TensorMG/inputs3D
Tests/LinearSolvers/C_TensorMG/inputs8
Tests/LinearSolvers/C_TensorMG/main_2D.F
Tests/LinearSolvers/C_TensorMG/main_3D.F
Tests/LinearSolvers/C_TensorMG/main_F.H
Tests/LinearSolvers/C_TensorMG/probin
Tests/LinearSolvers/C_TensorMG/testVI.cpp
Tests/LinearSolvers/C_TensorMG/vpramps.dat
Tools/CompileTesting/compiletesting.py
commit 6de33d5b79d614dcdd3c7f4923676c7e8ad503dd
Author: mic84
Date: Fri Aug 7 12:25:10 2020 -0700
MacProjector: allow for re-use of the object and enhance multi-level algorithm. (#1225)
* MacProjector: build RHS each time project() is called
* MacProjector: average down velocities before and after projection
Src/LinearSolvers/Projections/AMReX_MacProjector.H
Src/LinearSolvers/Projections/AMReX_MacProjector.cpp
commit b9ed579c052acac1fb73ced3ac2e9e63b17b3046
Author: Andrew Myers
Date: Fri Aug 7 11:59:09 2020 -0700
Fix unused variable warnings in AMReX_LinOp.* (#1237)
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/LinearSolvers/C_CellMG/AMReX_LinOp.H
Src/LinearSolvers/C_CellMG/AMReX_LinOp.cpp
commit 53d32def6b48c00270f97f4efd6251aa55cb6e7a
Author: Candace Gilet
Date: Thu Aug 6 17:06:06 2020 -0400
Fix typo. (#1234)
## Summary
Fix a typo in AMReX_extrapolater_3D_K.H
## Additional background
Some IAMR runs were crashing with out of bounds errors. This fixes those, and brings this C version in agreement with what the old F90 version was doing.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Amr/AMReX_extrapolater_3D_K.H
commit fe8cc5f3facab97cd39df5c06f6a3e2cc43ab378
Author: asalmgren
Date: Thu Aug 6 11:15:56 2020 -0700
Fix AMREX_HOME in Tests and Tutorials only (#1233)
* 1) Replace all instances of "AMREX_HOME ?=" by "AMREX_HOME =" in Tests and Tutorials only.
Since these exist within the amrex directory structure the default should be that they
always use the amrex they are part of, not the one set by the user's environment variable
2) Replace all instances of BoxLib/BOXLIB/boxlib by AMReX.
Tests/Algoim/GNUmakefile
Tests/AsyncOut/multifab/GNUmakefile
Tests/BBIOBenchmark/GNUmakefile
Tests/BaseFabTesting/GNUmakefile
Tests/GPU/CudaGraphs/BuildingGraphs/GNUmakefile
Tests/GPU/CudaGraphs/GraphBoundary/GNUmakefile
Tests/GPU/CudaGraphs/GraphInitTest/GNUmakefile
Tests/GPU/CudaGraphs/GraphReuseCopy/GNUmakefile
Tests/GPU/CudaGraphs/GraphWithMemcpy/GNUmakefile
Tests/GPU/CuptiTest/Exec/CUDA/GNUmakefile
Tests/GPU/Fuse/GNUmakefile
Tests/GPU/Locking/GNUmakefile
Tests/GPU/RandomNumberGeneration/GNUmakefile
Tests/GPU/Vector/GNUmakefile
Tests/GPU/libamrex_CUDA/GNUmakefile
Tests/HDF5Benchmark/GNUmakefile
Tests/LinearSolvers/CellEB/GNUmakefile
Tests/LinearSolvers/CellEB2/GNUmakefile
Tests/LinearSolvers/CellOverset/GNUmakefile
Tests/LinearSolvers/EBConvergenceTest/GNUmakefile
Tests/LinearSolvers/EBTensor/GNUmakefile
Tests/LinearSolvers/EBflux_grad/GNUmakefile
Tests/LinearSolvers/MLMG/GNUmakefile
Tests/LinearSolvers/NodalOverset/GNUmakefile
Tests/LinearSolvers/NodeEB/GNUmakefile
Tests/LinearSolvers/TensorOverset/GNUmakefile
Tests/MKDir/GNUmakefile
Tests/NoFort/GNUmakefile
Tests/Particles/AssignDensity/GNUmakefile
Tests/Particles/AssignMultiLevelDensity/GNUmakefile
Tests/Particles/AsyncIO/GNUmakefile
Tests/Particles/GNUmakefile
Tests/Particles/GhostsAndVirtuals/GNUmakefile
Tests/Particles/InitFromAscii/GNUmakefile
Tests/Particles/Intersection/GNUmakefile
Tests/Particles/NeighborParticles/GNUmakefile
Tests/Particles/ParallelContext/GNUmakefile
Tests/Particles/ParticleIterator/GNUmakefile
Tests/Particles/ParticleMesh/GNUmakefile
Tests/Particles/ParticleReduce/GNUmakefile
Tests/Particles/ParticleTransformations/GNUmakefile
Tests/Particles/Redistribute/GNUmakefile
Tests/Particles/SparseBins/GNUmakefile
Tests/Particles/TypeDescriptor/GNUmakefile
Tests/PnetCDFBenchmark/GNUmakefile
Tests/ProfTests/HeatEquation_EX1_C/Exec/GNUmakefile
Tests/ProfTests/ThirdParty/Exec/GNUmakefile
Tests/SinglePrecision/GNUmakefile
Tests/Slice/GNUmakefile
Tests/Stream/GNUmakefile
Tests/TypeCheck/GNUmakefile
Tests/Vectorization/GNUmakefile
Tests/complementIn/GNUmakefile
Tutorials/Amr/Advection_AmrCore/Exec/SingleVortex/GNUmakefile
Tutorials/Amr/Advection_AmrLevel/Exec/SingleVortex/GNUmakefile
Tutorials/Amr/Advection_AmrLevel/Exec/UniformVelocity/GNUmakefile
Tutorials/Amr/Advection_AmrLevel/Source/Src_2d/Adv_2d.f90
Tutorials/Amr/Advection_AmrLevel/Source/Src_3d/Adv_3d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/SingleVortex/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Exec/UniformVelocity/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/README
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_2d/Adv_2d.f90
Tutorials/AmrTask/MiniApps/Adv_Async_OnDemand/Source/Src_3d/Adv_3d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/SingleVortex/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Exec/UniformVelocity/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/README
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_2d/Adv_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync/Source/Src_3d/Adv_3d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/SingleVortex/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Exec/UniformVelocity/GNUmakefile
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/README
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_2d/Adv_2d.f90
Tutorials/AmrTask/MiniApps/Adv_phaseAsync_rgi/Source/Src_3d/Adv_3d.f90
Tutorials/Basic/HeatEquation_EX1_C/Exec/GNUmakefile
Tutorials/Basic/HeatEquation_EX1_CF/Exec/GNUmakefile
Tutorials/Basic/HeatEquation_EX1_F/GNUmakefile
Tutorials/Basic/HeatEquation_EX1_F/inputs
Tutorials/Basic/HeatEquation_EX2_C/Exec/GNUmakefile
Tutorials/Basic/HeatEquation_EX2_CF/Exec/GNUmakefile
Tutorials/Basic/HeatEquation_EX3_C/Exec/GNUmakefile
Tutorials/Basic/HelloWorld_C/GNUmakefile
Tutorials/Basic/HelloWorld_F/GNUmakefile
Tutorials/Basic/PrefixSum_MultiFab/GNUmakefile
Tutorials/Basic/main_C/GNUmakefile
Tutorials/Basic/main_F/GNUmakefile
Tutorials/Blueprint/AssignMultiLevelDensity/GNUmakefile
Tutorials/Blueprint/CellSortedParticles/GNUmakefile
Tutorials/Blueprint/HeatEquation_EX1_C/Exec/GNUmakefile
Tutorials/CVODE/SUNDIALS2_finterface/EX1/GNUmakefile
Tutorials/CVODE/SUNDIALS2_finterface/EX2/GNUmakefile
Tutorials/CVODE/SUNDIALS3_cppversion/EX1_CUDA_NVEC/GNUmakefile
Tutorials/CVODE/SUNDIALS3_cppversion/EX1_SERIAL_NVEC/GNUmakefile
Tutorials/CVODE/SUNDIALS3_finterface/EX1/GNUmakefile
Tutorials/CVODE/SUNDIALS3_finterface/EX_ark_analytic_fp/GNUmakefile
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_fp/GNUmakefile
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_sys_dns/GNUmakefile
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_analytic_sys_dns_jac/GNUmakefile
Tutorials/CVODE/SUNDIALS3_finterface/EX_cv_brusselator_dns/GNUmakefile
Tutorials/CVODE/SUNDIALS4/EX-CUSOLVER/GNUmakefile
Tutorials/CVODE/SUNDIALS4/EX_ark_analytic/GNUmakefile
Tutorials/EB/GeometryGeneration/GNUmakefile
Tutorials/EB/MacProj/GNUmakefile
Tutorials/EB/Poisson/GNUmakefile
Tutorials/ForkJoin/MLMG/GNUmakefile
Tutorials/ForkJoin/Simple/GNUmakefile
Tutorials/FortranInterface/Advection_F/Source/Src_2d/advect_2d_mod.F90
Tutorials/FortranInterface/Advection_F/Source/Src_3d/Adv_3d.f90
Tutorials/FortranInterface/Advection_octree_F/Source/Src_2d/advect_2d_mod.F90
Tutorials/GPU/Advection_AmrCore/Exec/SingleVortex/GNUmakefile
Tutorials/GPU/Launch/GNUmakefile
Tutorials/GPU/ParallelReduce/GNUmakefile
Tutorials/GPU/ParallelScan/GNUmakefile
Tutorials/LinearSolvers/ABecLaplacian_C/GNUmakefile
Tutorials/LinearSolvers/ABecLaplacian_F/GNUmakefile
Tutorials/LinearSolvers/MAC_Projection_EB/GNUmakefile
Tutorials/LinearSolvers/MultiComponent/GNUmakefile
Tutorials/LinearSolvers/NodalPoisson/GNUmakefile
Tutorials/LinearSolvers/Nodal_Projection_EB/GNUmakefile
Tutorials/LinearSolvers/NodeTensorLap/GNUmakefile
Tutorials/MUI/Exec_01/GNUmakefile
Tutorials/MUI/Exec_02/GNUmakefile
Tutorials/Particles/ElectromagneticPIC/Exec/CUDA/GNUmakefile
Tutorials/Particles/ElectromagneticPIC/Exec/OpenACC/GNUmakefile
Tutorials/Particles/ElectromagneticPIC/Exec/OpenMP/GNUmakefile
Tutorials/SDC/MISDC_ADR_2d/Exec/GNUmakefile
Tutorials/SENSEI/Advection_AmrCore/Exec/SingleVortex/GNUmakefile
Tutorials/SENSEI/Advection_AmrLevel/Exec/SingleVortex/GNUmakefile
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_2d/Adv_2d.f90
Tutorials/SENSEI/Advection_AmrLevel/Source/Src_3d/Adv_3d.f90
commit 888e804b77fad6754368396bb728e3ec491ed610
Author: Andrew Myers
Date: Wed Aug 5 13:23:09 2020 -0700
use rvalues for functions that take callables (#1231)
Src/Particle/AMReX_DenseBins.H
Src/Particle/AMReX_NeighborList.H
Src/Particle/AMReX_NeighborParticles.H
Src/Particle/AMReX_NeighborParticlesI.H
Src/Particle/AMReX_ParticleCommunication.H
Src/Particle/AMReX_SparseBins.H
commit 30fe5a59e9730683a28f7291ae96310d9560d917
Author: Michael Zingale
Date: Wed Aug 5 12:32:34 2020 -0400
add asserts for bounds checking to Array1D and Array2D (#1230)
## Summary
Array1D and Array2D did not check if the index was in bounds. This adds an `AMREX_ASSERT` that catches out of bounds issues when compiled with `DEBUG = TRUE`.
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_Array.H
commit e2dac9c4e6fa2a4163ec778b4958af843077069c
Author: WeiqunZhang
Date: Tue Aug 4 16:12:50 2020 -0700
Optimization of the construction of SFC (#1227)
## Summary
The Morton numbers are explicitly computed and stored in SFCTokens. This
speeds up the sorting in making space filling curve significantly.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_DistributionMapping.cpp
Src/Base/AMReX_Print.H
commit 9d76c3732bd03111c0279630761e5439e70c9b49
Author: Andrew Myers
Date: Tue Aug 4 15:00:39 2020 -0700
Refactoring of particle buffer map (#1229)
Src/Particle/AMReX_DenseBins.H
Src/Particle/AMReX_ParticleBufferMap.H
Src/Particle/AMReX_ParticleCommunication.H
Src/Particle/AMReX_ParticleLocator.H
Src/Particle/AMReX_ParticleUtil.H
commit d570a8c6647abc85a32cbcbe4af886fa2dde3782
Author: WeiqunZhang
Date: Tue Aug 4 11:13:27 2020 -0700
Ignore unused indices in AMREX_LOOP_3D and 4D (#1226)
Src/Base/AMReX_BoxArray.cpp
Src/Base/AMReX_Loop.H
commit f706d8aceb4272eb3e8f700cf77150cf4967aa20
Author: WeiqunZhang
Date: Tue Aug 4 09:20:57 2020 -0700
Optimization of FPinfo and complementIn (#1224)
## Summary
Use simplified BoxArray in FPinfo and parallelize FPinfo. Add parallel version of
BoxList::parallelComplementIn and use it in AmrMesh::MakeNewGrids.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Amr/AMReX_AmrLevel.cpp
Src/AmrCore/AMReX_AmrMesh.cpp
Src/AmrCore/AMReX_FillPatchUtil_I.H
Src/Base/AMReX_Box.H
Src/Base/AMReX_Box.cpp
Src/Base/AMReX_BoxArray.cpp
Src/Base/AMReX_BoxDomain.cpp
Src/Base/AMReX_BoxList.H
Src/Base/AMReX_BoxList.cpp
Src/Base/AMReX_DistributionMapping.cpp
Src/Base/AMReX_FabArrayBase.H
Src/Base/AMReX_FabArrayBase.cpp
commit 99d2227371a6a47e40b8175ee0f73f97ecf3aa59
Author: Lucas Esclapez <13371051+esclapez@users.noreply.github.com>
Date: Tue Aug 4 08:41:54 2020 -0700
Feature extrapolator gpu (#1222)
## Summary
Move the Amr/Extrapolater used in IAMR/PeleLM to C++/GPU compliant.
## Additional background
The Extrapolater allows to fill ghost cells with first order extrapolation from interior cells.
## Checklist
The C++ version does not introduce any changes to the results and passed CI compilation tests.
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [X] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Amr/AMReX_Extrapolater.cpp
Src/Amr/AMReX_extrapolater_1D_K.H
Src/Amr/AMReX_extrapolater_1d.f90
Src/Amr/AMReX_extrapolater_2D_K.H
Src/Amr/AMReX_extrapolater_2d.f90
Src/Amr/AMReX_extrapolater_3D_K.H
Src/Amr/AMReX_extrapolater_3d.f90
Src/Amr/AMReX_extrapolater_K.H
Src/Amr/CMakeLists.txt
Src/Amr/Make.package
commit 294435b79bddc3106249a15c64bd7b4322c19024
Author: Andrew Myers
Date: Mon Aug 3 14:11:17 2020 -0700
Less surprising behavior for the ok() method of FabArray. (#1223)
This makes the `ok()` method of FabArray return `false` instead of crashing if the `define()` method has yet to be called.
We allow FabArray and its derived classes to be default constructed, which places them in a unusable state until the the `define` method has been called. We also provide an `ok()` method, which according to its doxygen strings returns whether the FabArray is well-defined. Currently, this method will crash with a not-particularly instructive error message if it is called on a default-constructed-but-not-defined FabArray. This pull request changes it to return `false` instead, which I'd argue is more intuitive - if it's legal to construct a FabArray and define it later, then it should be legal to test whether that has been done or not. I have also updated the doxygen string for the method in question.
Even if we want this to crash instead of returning `false`, it should do so with its own assertion and a clear error message.
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [x] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_FabArray.H
commit a918cdabf2bd5f5991fe69494593634f1fced2b4
Author: WeiqunZhang
Date: Mon Aug 3 13:16:49 2020 -0700
Optimization of regrid (#1199)
## Summary
New BoxList::maxSize function that keeps Boxes after chopping in an order that is easy to merge them
back into original Box.
New BoxList::order_simplify function that assumes the BoxList is in the order that we only need to
search the next Box in the list.
Cache simplified BoxList inside BoxArray.
Use simplified BoxArray in FPinfo.
This commit will make most regression tests fail because it changes the grids. But if the comparison is
done with `fcompare....ex -a`, it should only change answers at roundoff level unless the test is very
sensitive.
Src/AmrCore/AMReX_AmrMesh.cpp
Src/Base/AMReX_BoxArray.H
Src/Base/AMReX_BoxArray.cpp
Src/Base/AMReX_BoxList.H
Src/Base/AMReX_BoxList.cpp
Src/Base/AMReX_FabArrayBase.cpp
commit 4c50993ce86e8f6c54c50b93cbbb1aab92d42edd
Author: Brandon Runnels
Date: Mon Aug 3 12:29:18 2020 -0600
Fillboundary (#1171)
This pull request fixes the problem of edge nodes not getting filled properly when filling multiple layers of ghost nodes. This occurs when there are overlapping ghost nodes on a coarse/fine boundary. (See issue #568)
Summary of changes to AMreX:
- Added a `BoxArray::complementIn` function that takes a box _list_ and returns the complement box list.
- Added an additional boolean argument `multi_ghost` to all `FillBoundary` functions that gets passed (eventually) to `define_fb`. In all cases, a default value of `false` is supplied.
- Modified `define_fb` to include the missing ghost nodes in the list of `send_tags`. (This uses the new `complementIn` function.)
Summery of changes to `LinearSolver/MultiComponent` tutorial:
- Removed the `realFillBoundary` function, replaced with `FillBoundary(false,true)
- Modified `main.cpp` so that it can be compiled and run (non-trivially) in 2D as well as 3D.
Diagram of the problem that this PR fixes:

Src/Base/AMReX_FabArrayBase.H
Src/Base/AMReX_FabArrayBase.cpp
Tutorials/LinearSolvers/MultiComponent/MCNodalLinOp.H
Tutorials/LinearSolvers/MultiComponent/MCNodalLinOp.cpp
Tutorials/LinearSolvers/MultiComponent/inputs
Tutorials/LinearSolvers/MultiComponent/main.cpp
commit 28ceb92eae33e7b7797831bed85ad6242fdcf317
Author: Brian Friesen
Date: Mon Aug 3 11:25:41 2020 -0700
Tools: add NVIDIA HPC SDK as a compiler (#1131)
HPC SDK replaced PGI after PGI 20.4, so we should get it working now.
Tools/GNUMake/Make.defs
Tools/GNUMake/Make.rules
Tools/GNUMake/comps/hpcsdk.mak
commit 273cf0cc0c2b26d1f81147a5a00ceb26bd17184c
Author: asalmgren
Date: Sun Aug 2 14:09:09 2020 -0700
The average_down_faces calls between AMR levels in the linear solvers need to see periodicity (#1221)
The average_down_faces calls between AMR levels need to use the interfaces that sees the periodicity. This requires adding that functionality to EB_average_down_faces (it has already been added to non-EB average_down_faces.)
## Summary
Recently we added the functionality for an average_down_faces call to average fine data from one side to average not only onto the face under it but also onto the face on the other side of the domain if periodic in that direction. This PR 1) extends that functionality to the EB_average_down_faces call and 2) uses this new functionality when averaging down face-based coefficients between AMR levels in the cell-centered linear solvers.
## Checklist
The proposed changes:
- [X] fix a bug or incorrect behavior in AMReX
- [X] add new capabilities to AMReX
- [X] changes answers in the test suite to more than roundoff level
- [X] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/EB/AMReX_EBMultiFabUtil.H
Src/EB/AMReX_EBMultiFabUtil.cpp
Src/LinearSolvers/MLMG/AMReX_MLABecLaplacian.cpp
Src/LinearSolvers/MLMG/AMReX_MLEBABecLap.cpp
Src/LinearSolvers/MLMG/AMReX_MLEBTensorOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLTensorOp.cpp
commit 9cd538802ea5b3ab7e6db10d6046933e937aa449
Author: OscarAntepara <52221614+OscarAntepara@users.noreply.github.com>
Date: Sun Aug 2 14:00:12 2020 -0700
Changing computation at outflow for the EBTensor (#1187)
Computation at outflow boundaries for the EBTensor has been changed to not take into account cells outside the domain.
Src/LinearSolvers/MLMG/AMReX_MLEBTensorOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLEBTensor_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLEBTensor_3D_K.H
commit ccaebccf710600e2e28510f8ba43b760a10a918b
Author: asalmgren
Date: Sun Aug 2 13:16:56 2020 -0700
Fix the calculation of divu at inflow face in nodal projection so that it does not use tangential velocities on an inflow face (#1219)
## Summary
Mathematically the nodal projection should only see the normal velocity at inflow faces, but it was coded so that it would use non-zero tangential velocities at inflow in the calculation of divu. Most users never set tangential velocities to anything but zero, but if they did, this would give an incorrect divergence.
## Additional background
## Checklist
The proposed changes:
- [X] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [X] changes answers in the test suite to more than roundoff level
- [X] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_1D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLaplacian.cpp
commit 9ed2833df8a110f0ea43e901da96a93f0686a031
Author: Marc Day
Date: Sat Aug 1 18:48:08 2020 -0700
Remove instructions....not needed and clutters Slack notices (#1218)
## Summary
Although the instructions were commented out in the formatted PR text, they clutter up the Slack notices, and aren't really necessary anyway.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
.github/pull_request_template.md
commit 20d9b4a451bc2ee8a9133fa618e46ffd479c6807
Author: WeiqunZhang
Date: Sat Aug 1 12:58:03 2020 -0700
20.08 (#1220)
## Summary
Update CHANGES for 20.08
## Additional background
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
CHANGES
commit 2dad6e2975931107be98205086eb11969e937ed1
Author: WeiqunZhang
Date: Thu Jul 30 22:53:09 2020 -0700
Average down faces (#1216)
New average_down_faces functions that take coarse Geometry and average down periodically.
Src/Base/AMReX_MultiFabUtil.H
Src/Base/AMReX_MultiFabUtil.cpp
commit 1476ea3ef4a3dc6bae80b241d85657c0fc5d9d31
Author: WeiqunZhang
Date: Wed Jul 29 18:30:43 2020 -0700
fix angle brackets pr template (#1215)
## Summary
The `-->` in the html comment part of the PR template closes the comment and leaves the words after that and before the actual closing `-->` visible. This is fixed.
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
.github/pull_request_template.md
commit d8cdfd9eda11126954ac549c57eec8723e97f21b
Author: WeiqunZhang
Date: Wed Jul 29 18:26:27 2020 -0700
Fix warnings for DPC++ (#1214)
## Summary
- Add more warnings flags to dpcpp.
- Fix warnings for DPC++.
- Fix a bug in DPC++ version of ReduceLogicalOr.
## Checklist
The proposed changes:
- [x] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
Src/Base/AMReX_BlockMutex.H
Src/Base/AMReX_BlockMutex.cpp
Src/Base/AMReX_FabArrayUtility.H
Src/Base/AMReX_GpuAtomic.H
Src/Base/AMReX_GpuDevice.cpp
Src/Base/AMReX_GpuUtility.H
Src/Base/AMReX_Random.cpp
Src/Base/AMReX_Reduce.H
Src/Base/AMReX_parstream.cpp
Tools/GNUMake/comps/dpcpp.mak
Tools/GNUMake/comps/llvm.mak
commit 2dc11c2ffd58a4e3f50d691a856446db0dfde8c5
Author: kngott <30483578+kngott@users.noreply.github.com>
Date: Wed Jul 29 09:01:50 2020 -0700
Update the documentation for NSight tools. (#1163)
Created a skeleton in External Profiling for documenting AMReX specific implementations of the NSight package.
Also updated the run.corigpu script with new data from the Hackathon.
Docs/sphinx_documentation/source/External_Profiling_Tools.rst
Docs/sphinx_documentation/source/External_Profiling_Tools_Chapter.rst
Tutorials/GPU/run.corigpu
Tutorials/GPU/run.script
Tutorials/GPU/run.summit
commit 94961adbaf07c228eaae7c3ee3536f1ee338a3ce
Author: Houjun
Date: Tue Jul 28 17:12:15 2020 -0700
Fix a bug with HDF5 boxcenter data (#1213)
* Fix a bug that causes wrong value assignment for the boxcenter dataset
Co-authored-by: Houjun Tang
Src/Base/AMReX_PlotFileUtil.cpp
commit 1a4780d8ef5fb2fa7948ab926c942ea551690e60
Author: Marc Day
Date: Tue Jul 28 14:25:29 2020 -0700
Add PR template file (#1211)
## Sumary
Following CASTRO's lead, this template for PRs will pre-fill PR messages with suggested text. Hitting "Preview" in the PR window will format the message nicely.
## Checklist
The proposed changes:
- [ ] fix a bug or incorrect behavior in AMReX
- [ ] add new capabilities to AMReX
- [ ] changes answers in the test suite to more than roundoff level
- [ ] are likely to significantly affect the results of downstream AMReX users
- [ ] are described in the proposed changes to the AMReX documentation, if appropriate
.github/pull_request_template.md
commit 97a6fa2f59f2985a4f67adbd8d431a50f8a35590
Author: Andrew Myers
Date: Tue Jul 28 13:14:02 2020 -0700
Turn off CI for DPC++ right now, since it's currently broken. (#1210)
* turn off the DPC++ CI workflow for now, as it is broken in beta 8
* note that this is borked right now
.github/workflows/linux.yml
commit e4da66c8e01efa09257e6362688c8a7c4e3bd1dd
Author: asalmgren
Date: Mon Jul 27 17:27:16 2020 -0700
Docs hypre fix (#1207)
Docs/sphinx_documentation/source/LinearSolvers.rst
commit 9db726812c87a2138ee2b2a831d3363824f1c633
Author: Axel Huebl
Date: Mon Jul 27 12:59:33 2020 -0700
CMake: NVCC/cudafe Error Numbers (#1206)
Activate this fancy, undocumented flag to show error numbers in NVCC.
This is great if one wants to suppress (or report) them and should be
standard.
Tools/CMake/AMReX_SetupCUDA.cmake
commit ffe8d2e06bf276cbe382b658223ea3b4d78b89e7
Author: WeiqunZhang
Date: Mon Jul 27 11:16:17 2020 -0700
fix periodic boundary bug in #1204 (#1205)
Src/AmrCore/AMReX_TagBox.cpp
commit d731f8ba3a26c74febfae7d42862617e01112534
Author: WeiqunZhang
Date: Sun Jul 26 18:05:50 2020 -0700
EB outside domain (#1201)
* EB outside domain
Add a runtime parameter `eb2.extend_domain_face` to control EB generation outside the domain. The
default is false. That is the implicit function will be used for cells outside the domain. This
preserves the current behavior. If it is set to true, the geometry information is then extrapolated
from domain faces.
* fix fillFab for EB_levelset
Src/Base/AMReX_Algorithm.H
Src/EB/AMReX_EB2.H
Src/EB/AMReX_EB2.cpp
Src/EB/AMReX_EB2_GeometryShop.H
Src/EB/AMReX_EB2_IndexSpaceI.H
Src/EB/AMReX_EB2_Level.H
Src/EB/AMReX_EB_levelset.H
commit 931ac6fb4a03faed44918731ecb383368ba0cfe2
Author: WeiqunZhang
Date: Sun Jul 26 15:06:25 2020 -0700
Fix bug in #1156 (#1204)
In #1156, newly generated grids were not always properly nested because the tags in ghost cells were
discarded. This is fixed and an assertion is added.
Src/AmrCore/AMReX_AmrMesh.cpp
Src/AmrCore/AMReX_TagBox.cpp
Src/Base/AMReX_iMultiFab.H
Src/Base/AMReX_iMultiFab.cpp
commit 6812ea39f6739eae11d4cfaf9ac28940e77df0ae
Author: Max Katz
Date: Sun Jul 26 09:22:22 2020 -0700
Fix CellConservativeProtected bug from #1184 (#1203)
* Fix CellConservativeProtected bug from #1184
* Fix comment
Src/AmrCore/AMReX_Interpolater.cpp
commit 1dbf5d7a5939e2b9bdb41627716fe392f168c5e0
Author: WeiqunZhang
Date: Sat Jul 25 20:43:52 2020 -0700
GNU Make option to turn warnings into errors (#1198)
Add `WARN_ERROR` (with default of `FALSE`) to the GNU Make system to turn warnings into errors for
GCC and Clang. Add `WARN_ERROR` to the 2d and 3d CIs using GNU Make.
.github/workflows/linux.yml
Docs/sphinx_documentation/source/BuildingAMReX.rst
Tools/GNUMake/Make.defs
Tools/GNUMake/comps/gnu.mak
Tools/GNUMake/comps/llvm.mak
commit b9b0196728b93353a34f1152337f88a701ff6fde
Author: WeiqunZhang
Date: Sat Jul 25 16:14:04 2020 -0700
udpate branch name due to regression_testing branch name change (#1200)
Docs/Doxygen/doxygen.conf
Docs/sphinx_documentation/source/Testing.rst
commit 8ce4756814f2064eae003a5c414e914f60a69960
Author: Andrew Myers
Date: Fri Jul 24 19:12:29 2020 -0700
Fix some more clang warnings (#1197)
Src/AmrCore/AMReX_ErrorList.cpp
Src/AmrCore/AMReX_FillPatchUtil.cpp
Src/Base/AMReX_Array.H
Src/Base/AMReX_Box.H
Src/Base/AMReX_CoordSys.H
Src/Base/AMReX_CoordSys.cpp
Src/Base/AMReX_Geometry.H
Src/Base/AMReX_Geometry.cpp
Src/Base/AMReX_Machine.cpp
Src/Base/AMReX_MultiFabUtil.cpp
Src/Base/AMReX_ParallelReduce.H
Src/Base/AMReX_PhysBCFunct.H
Src/Base/AMReX_PlotFileDataImpl.cpp
Src/Boundary/AMReX_YAFluxRegister.cpp
Src/F_Interfaces/AmrCore/AMReX_FlashFluxRegister.cpp
Src/F_Interfaces/AmrCore/AMReX_fluxregister_fi.cpp
Src/F_Interfaces/LinearSolvers/AMReX_abeclaplacian_fi.cpp
Src/F_Interfaces/LinearSolvers/AMReX_linop_fi.cpp
Src/LinearSolvers/MLMG/AMReX_MLABecLaplacian.cpp
Src/LinearSolvers/MLMG/AMReX_MLALaplacian.H
Src/LinearSolvers/MLMG/AMReX_MLCellLinOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLLinOp.H
Src/LinearSolvers/MLMG/AMReX_MLLinOp_K.H
Src/LinearSolvers/MLMG/AMReX_MLMGBndry.cpp
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_K.H
Src/LinearSolvers/MLMG/AMReX_MLPoisson.H
Src/Particle/AMReX_ParticleCommunication.H
Src/Particle/AMReX_TracerParticles.cpp
commit c013c29ad1e9e7da23431a6f4b5c1ab47cc059a4
Author: WeiqunZhang
Date: Fri Jul 24 18:27:53 2020 -0700
Gcc7 no array bounds (#1196)
* Wno-array-bounds for gcc 7
* [[gnu::fallthrought]] is available for gcc >= 7 only. Test the result of dynamic_cast.
Src/Base/AMReX_Extension.H
Src/LinearSolvers/MLMG/AMReX_MLMG.cpp
Tools/GNUMake/comps/gnu.mak
commit c9ca46b45b96ce1507e74a0bcafe545de802e5c0
Author: Andrew Myers
Date: Fri Jul 24 16:24:55 2020 -0700
Remove anonymous struct, which is not standard. Note that this is a b… (#1191)
Src/Particle/AMReX_Particle.H
commit 660da8bb839c47a6aa530b9611006c1d3fb8c50f
Author: WeiqunZhang
Date: Fri Jul 24 14:42:28 2020 -0700
fix a wrong clang flag (#1194)
* fix a wrong clang flag
* fix a warning
Src/LinearSolvers/MLMG/AMReX_MLNodeLaplacian.cpp
Tools/GNUMake/comps/llvm.mak
commit 572648806d3aeaac06406637b13802c975171b24
Author: WeiqunZhang
Date: Fri Jul 24 12:45:45 2020 -0700
fix clang warnings (#1193)
Src/Amr/AMReX_Amr.H
Src/Amr/AMReX_AmrLevel.H
Src/Amr/AMReX_StateDescriptor.cpp
Src/AmrCore/AMReX_FillPatchUtil_I.H
Src/AmrCore/AMReX_Interpolater.H
Src/Base/AMReX_Array.H
Src/Base/AMReX_Extension.H
Src/Base/AMReX_FabArray.H
Src/Base/AMReX_GpuRange.H
Src/Base/AMReX_PODVector.H
Src/Base/AMReX_ParallelDescriptor.H
Src/Base/AMReX_Tuple.H
Src/Boundary/AMReX_Mask.H
Src/LinearSolvers/MLMG/AMReX_MLCGSolver.cpp
Src/LinearSolvers/Projections/AMReX_MacProjector.H
Src/LinearSolvers/Projections/AMReX_NodalProjector.H
Src/Particle/AMReX_BinIterator.H
Src/Particle/AMReX_ParticleCommunication.H
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_ParticleInit.H
Src/Particle/AMReX_Particles.H
Tools/GNUMake/comps/llvm.mak
commit a4947448f775cbc027a9b0b6a639e70a74caae9f
Author: Andrew Myers
Date: Fri Jul 24 11:49:35 2020 -0700
fix a parameter shadowing warning coming from VisMF::RemoveFiles with gcc (#1192)
Src/Base/AMReX_VisMF.cpp
commit 1fe53cb2852b162a9d7f37a2147e4e4e651d1c2f
Author: WeiqunZhang
Date: Fri Jul 24 11:07:38 2020 -0700
fix more warnings and add CI for building 3d with configure (#1184)
* fix more warnings and add CI for building 3d with configure
* make ignore_unused constexpr and force inline
* add AMREX_FALLTHROUGH
* fix more warnings
* enable xsdk in ci and test no_fortran
* remove constexpr from ignore_unused for C++11
* use Clang for 2d ci
* disable mpi for the 2d ci with clang
.github/workflows/linux.yml
Src/Amr/AMReX_Amr.H
Src/Amr/AMReX_Amr.cpp
Src/Amr/AMReX_AmrLevel.H
Src/Amr/AMReX_AmrLevel.cpp
Src/Amr/AMReX_StateDescriptor.cpp
Src/AmrCore/AMReX_FillPatchUtil.H
Src/AmrCore/AMReX_Interpolater.cpp
Src/AmrCore/AMReX_TagBox.cpp
Src/Base/AMReX.H
Src/Base/AMReX_Array4.H
Src/Base/AMReX_BLProfiler.H
Src/Base/AMReX_BLProfiler.cpp
Src/Base/AMReX_BlockMutex.H
Src/Base/AMReX_Box.cpp
Src/Base/AMReX_CArena.cpp
Src/Base/AMReX_CoordSys.cpp
Src/Base/AMReX_DistributionMapping.cpp
Src/Base/AMReX_Extension.H
Src/Base/AMReX_FACopyDescriptor.H
Src/Base/AMReX_FabArrayBase.cpp
Src/Base/AMReX_FabConv.cpp
Src/Base/AMReX_FilCC_1D_C.H
Src/Base/AMReX_FilCC_2D_C.H
Src/Base/AMReX_Machine.cpp
Src/Base/AMReX_MultiFab.cpp
Src/Base/AMReX_MultiFabUtil.cpp
Src/Base/AMReX_MultiFabUtil_1D_C.H
Src/Base/AMReX_NFiles.cpp
Src/Base/AMReX_ParallelContext.cpp
Src/Base/AMReX_ParallelDescriptor.H
Src/Base/AMReX_ParallelDescriptor.cpp
Src/Base/AMReX_ParallelReduce.H
Src/Base/AMReX_ParmParse.cpp
Src/Base/AMReX_PhysBCFunct.H
Src/Base/AMReX_Reduce.H
Src/Base/AMReX_Utility.H
Src/Base/AMReX_VisMF.cpp
Src/Base/AMReX_iMultiFab.cpp
Src/Boundary/AMReX_InterpBndryData.H
Src/Boundary/AMReX_InterpBndryData_1D_K.H
Src/Boundary/AMReX_YAFluxRegister_1D_K.H
Src/EB/AMReX_EB2_2D_C.cpp
Src/EB/AMReX_EB2_3D_C.cpp
Src/EB/AMReX_EB2_GeometryShop.H
Src/EB/AMReX_EB2_IF_AllRegular.H
Src/EB/AMReX_EB2_Level.H
Src/EB/AMReX_EB2_Level.cpp
Src/EB/AMReX_EBFluxRegister.cpp
Src/EB/AMReX_EBFluxRegister_2D_C.H
Src/EB/AMReX_EBMultiFabUtil.cpp
Src/EB/AMReX_EBMultiFabUtil_2D_C.H
Src/EB/AMReX_EBMultiFabUtil_3D_C.H
Src/EB/AMReX_EB_LSCoreBase.cpp
Src/EB/AMReX_EB_levelset.cpp
Src/EB/AMReX_EB_utils.cpp
Src/EB/AMReX_algoim.cpp
Src/EB/AMReX_distFcnElement.cpp
Src/F_Interfaces/AmrCore/AMReX_FlashFluxRegister.cpp
Src/F_Interfaces/Base/AMReX_FPhysBC.cpp
Src/F_Interfaces/Base/AMReX_boxarray_fi.cpp
Src/F_Interfaces/Base/AMReX_init_fi.cpp
Src/F_Interfaces/Base/AMReX_parallel_mod.F90
Src/F_Interfaces/Octree/AMReX_octree_fi.cpp
Src/F_Interfaces/Octree/AMReX_octree_mod.F90
Src/LinearSolvers/MLMG/AMReX_MLABecLap_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLEBABecLap.H
Src/LinearSolvers/MLMG/AMReX_MLEBTensorOp.H
Src/LinearSolvers/MLMG/AMReX_MLEBTensorOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLLinOp.H
Src/LinearSolvers/MLMG/AMReX_MLLinOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_1D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeTensorLap_1D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeTensorLap_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeTensorLap_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeTensorLaplacian.cpp
Src/LinearSolvers/MLMG/AMReX_MLPoisson.cpp
Src/LinearSolvers/Projections/AMReX_MacProjector.cpp
Src/LinearSolvers/Projections/AMReX_NodalProjector.cpp
Src/Particle/AMReX_ParticleCommunication.cpp
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_ParticleIO.H
commit 5dbcd14861f7656df4e2b2d5a588a00b7ad5c8ef
Author: Andrew Myers
Date: Fri Jul 24 11:03:38 2020 -0700
rewrite these loops in a way that silences a gcc warning about potential null deferences (#1190)
Src/Particle/AMReX_ParticleTile.H
commit fb413521ab2ebd9fa7614c1434a4c913a3516b06
Author: Andrew Myers
Date: Fri Jul 24 09:38:49 2020 -0700
Make aggregation type and buffer static member data of ParticleContainer. This silences a warning about unused variables. (#1189)
Src/Particle/AMReX_Particle.H
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_Particles.H
commit 2acea22a9390f3a691d7c2b721aa0b03c5f44129
Author: asalmgren
Date: Fri Jul 24 09:38:04 2020 -0700
D_DECL --> AMREX_D_DECL (#1188)
Src/AmrCore/AMReX_ErrorList.cpp
commit 0c180badf2333e18101fefbe900794fb24964e51
Author: Axel Huebl
Date: Fri Jul 24 09:32:02 2020 -0700
CI: Clang 6 C++17 SP NOMPI (#1181)
Add a Clang build to CI that also covers single precision (SP).
.github/workflows/dependencies/dependencies_clang6.sh
.github/workflows/linux.yml
commit 92418c3a0256c66a365a6ee4f61510a4308348a9
Author: WeiqunZhang
Date: Thu Jul 23 22:23:50 2020 -0700
Communication buffers (#1186)
Use The_Cpu_Arena (BArena) instead of The_Pinned_Arena (CArena) for communication buffers to avoid
the CArena fragmentation issue with the type of problems that allocate increasingly larger memory in
a contiguous chunk during simulation. CArena then cannot effective use previous allocation chunks
because they are slightly smaller. The fix is for CPU build only. We still need to have a strategy
for this for GPU build.
Src/Base/AMReX_FabArrayBase.cpp
commit 9f29a86876fb1c8ada03cb121a10869211ea9bba
Author: Mu-Hua Chien
Date: Fri Jul 24 00:42:19 2020 -0400
Fix a bug for Nodal semicoarsening (#1185)
* set up starting point
* Fix the problem of semicoarsening in yz direction and restore MyTest.cpp and input-rt
* fix a bug in 2d
* commit for debuggin failure in tridiagonal solve
* fix a bug when doing semicoarsening with short x
Co-authored-by: Kan Bok-Hua
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLaplacian.cpp
Tutorials/LinearSolvers/NodalPoisson/MyTest.cpp
commit 84276f6176ca64a8df617b33c9e82a5915d72396
Author: WeiqunZhang
Date: Thu Jul 23 20:28:15 2020 -0700
Add some supports for nghost vector in Fortran multifab (#1099)
* Support nghost vector in Fortran multifab
* nghost vector version of amrex_multifab%copy
Src/F_Interfaces/Base/AMReX_multifab_fi.cpp
Src/F_Interfaces/Base/AMReX_multifab_mod.F90
commit e010fa881d837acd0824156323c1ff8aa44d67c5
Author: Axel Huebl
Date: Thu Jul 23 19:55:28 2020 -0700
Wreorder: ErrorList (#1182)
Fix:
```
AMReX_ErrorList.H:434:47: warning: field 'm_ngrow' will be initialized after field 'm_info' [-Wreorder]
```
Src/AmrCore/AMReX_ErrorList.H
commit 370bccc3f3b148bde0ec9377697d5db31b4bb946
Author: Axel Huebl
Date: Thu Jul 23 15:21:08 2020 -0700
Fix c++11 float narrowing (#1180)
Fix a compile issue with DPC++ (SP) and AppleClang (DP):
```
Src/AmrCore/AMReX_ErrorList.cpp:318:52: error: non-constant-expression cannot be narrowed from type 'double' to 'float' in initializer list [-Wc++11-narrowing]
```
Src/AmrCore/AMReX_ErrorList.cpp
commit 3b0aec5d0286b533835d1a245acd4afbc0a3067a
Author: Andrew Myers
Date: Thu Jul 23 14:06:11 2020 -0700
reimplement SortParticlesByCell() in terms of SortParticlesByBin(), reducing code duplication. (#1172)
Src/Particle/AMReX_ParticleContainerI.H
commit e1b7a288e50553150610baeabca3d6fe12e29d58
Author: Andrew Myers
Date: Thu Jul 23 14:05:10 2020 -0700
Test for ghost and virtual particles (#1179)
Tests/Particles/GhostsAndVirtuals/GNUmakefile
Tests/Particles/GhostsAndVirtuals/Make.package
Tests/Particles/GhostsAndVirtuals/inputs
Tests/Particles/GhostsAndVirtuals/main.cpp
commit bba8d3fa1c0bba29289caeb314ec19d6188ab84e
Author: Max Katz
Date: Thu Jul 23 13:41:09 2020 -0700
Switch to nvtxRangePush() and nvtxRangePop() in TinyProfiler (#1177)
Src/Base/AMReX_TinyProfiler.H
Src/Base/AMReX_TinyProfiler.cpp
commit 2abcbb8bd30239dc4278cf8bbd717e1af5ab13d1
Author: Andrew Myers
Date: Thu Jul 23 13:35:27 2020 -0700
fix enable_if and typedef (#1178)
Src/Particle/AMReX_ParticleTile.H
commit f031488e07d10d409f7fbffe52f073de055eafc9
Author: WeiqunZhang
Date: Thu Jul 23 12:00:46 2020 -0700
Fused GPU kernel launches (#1135)
* First pass of fused launches
* fuse 4D kernels too
* fuse 1D kernels too
* consolidate cudaMemcpy
* Fuser singleton
* add a test for fuse
* add more assertion on not in omp parallel region
* move the varidic version of launch_global back to GpuLaunch.H because it depends on other funcitons not in GpuLaunchGlobal.H
* fix comment
* fix a couple of bugs
* do the not-in-omp-parallel assertion only if there are fused kernels to launch
* add recommended threshold for kernel fusing
* missing include guard
Src/Base/AMReX.cpp
Src/Base/AMReX_Arena.H
Src/Base/AMReX_Gpu.H
Src/Base/AMReX_GpuFuse.H
Src/Base/AMReX_GpuFuse.cpp
Src/Base/AMReX_GpuLaunch.H
Src/Base/AMReX_GpuLaunchGlobal.H
Src/Base/CMakeLists.txt
Src/Base/Make.package
Tests/GPU/Fuse/GNUmakefile
Tests/GPU/Fuse/Make.package
Tests/GPU/Fuse/main.cpp
commit 685886a14096b5180dbb9f3e039f0d44272c8be2
Author: WeiqunZhang
Date: Thu Jul 23 11:56:17 2020 -0700
update DPC++ wishlist (#1175)
Docs/Notes/DPCPPWishlist.md
commit 411ea9b9467f9443b8d455a98deca43b1f7233ce
Author: WeiqunZhang
Date: Thu Jul 23 11:55:49 2020 -0700
WARN_ALL (#1176)
* WARN_ALL
Add WARN_ALL option to GNUMake system to turn on more compiler warnings for gcc and clang. The
default is FALSE. Fix a couple of bugs found by the warning and a number of warnings.
* add WARN_ALL to documentation
Docs/sphinx_documentation/source/BuildingAMReX.rst
Src/AmrCore/AMReX_Cluster.cpp
Src/AmrCore/AMReX_FluxRegister.cpp
Src/AmrCore/AMReX_Interp_3D_C.H
Src/Base/AMReX_BCUtil.cpp
Src/Base/AMReX_BaseFab.H
Src/Base/AMReX_CArena.H
Src/Base/AMReX_CoordSys.cpp
Src/Base/AMReX_EArena.H
Src/Base/AMReX_FArrayBox.cpp
Src/Base/AMReX_FBI.H
Src/Base/AMReX_FabArray.H
Src/Base/AMReX_FabArrayBase.cpp
Src/Base/AMReX_FabArrayCommI.H
Src/Base/AMReX_FabArrayUtility.H
Src/Base/AMReX_FabConv.cpp
Src/Base/AMReX_FilCC_C.cpp
Src/Base/AMReX_FilND_C.cpp
Src/Base/AMReX_GpuAsyncArray.cpp
Src/Base/AMReX_GpuDevice.cpp
Src/Base/AMReX_GpuElixir.cpp
Src/Base/AMReX_GpuLaunch.H
Src/Base/AMReX_GpuLaunchFunctsC.H
Src/Base/AMReX_MFCopyDescriptor.cpp
Src/Base/AMReX_MultiFab.cpp
Src/Base/AMReX_MultiFabUtil.cpp
Src/Base/AMReX_PODVector.H
Src/Base/AMReX_ParallelDescriptor.cpp
Src/Base/AMReX_PlotFileUtil.cpp
Src/Base/AMReX_Random.cpp
Src/Base/AMReX_Reduce.H
Src/Base/AMReX_TinyProfiler.cpp
Src/Base/AMReX_VisMF.cpp
Src/Boundary/AMReX_InterpBndryData.H
Src/LinearSolvers/MLMG/AMReX_MLABecLaplacian.H
Src/LinearSolvers/MLMG/AMReX_MLABecLaplacian.cpp
Src/LinearSolvers/MLMG/AMReX_MLALaplacian.H
Src/LinearSolvers/MLMG/AMReX_MLCellABecLap.H
Src/LinearSolvers/MLMG/AMReX_MLCellLinOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLLinOp.H
Src/LinearSolvers/MLMG/AMReX_MLLinOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLLinOp_K.H
Src/LinearSolvers/MLMG/AMReX_MLMG.cpp
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLaplacian.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLinOp.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLinOp.cpp
Src/LinearSolvers/MLMG/AMReX_MLNodeTensorLap_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeTensorLaplacian.cpp
Src/LinearSolvers/MLMG/AMReX_MLPoisson.H
Src/LinearSolvers/MLMG/AMReX_MLPoisson.cpp
Src/LinearSolvers/MLMG/AMReX_MLTensorOp.H
Src/LinearSolvers/MLMG/AMReX_MLTensorOp.cpp
Src/Particle/AMReX_ParticleCommunication.cpp
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_ParticleUtil.H
Tools/GNUMake/Make.defs
Tools/GNUMake/comps/gnu.mak
Tools/GNUMake/comps/llvm.mak
commit 38e3dfb14e30c6ed05a5335185fd746d0b23ba4b
Author: WeiqunZhang
Date: Thu Jul 23 08:58:17 2020 -0700
fix documentation format (#1174)
Docs/sphinx_documentation/source/Basics.rst
commit e1cffba53407ce934add479ea73145e0ee815024
Author: Marc Day
Date: Wed Jul 22 20:49:28 2020 -0700
Add alternative error tagging class (#1166)
* Add alternative error tagging class
* Address PR comments
Src/AmrCore/AMReX_ErrorList.H
Src/AmrCore/AMReX_ErrorList.cpp
commit 8e535d480ab2eb3f180aeb5078d6525ec24d8a21
Author: Andrew Myers
Date: Wed Jul 22 13:41:35 2020 -0700
Make ghost particles support soa data (#1053)
Src/Particle/AMReX_ArrayOfStructs.H
Src/Particle/AMReX_ParIter.H
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_ParticleTile.H
Src/Particle/AMReX_ParticleTransformation.H
Src/Particle/AMReX_Particles.H
commit 5fa8dc1d70d58c04c7170dfbe85a02c954a8c071
Author: jmsexton03
Date: Wed Jul 22 16:07:27 2020 -0400
Fix HDF5 dataset id precision for single precision particles (#1173)
Fixes segfault in amrex/Tests/HDF5Benchmark for
USE_HDF5=TRUE USE_CUDA=TRUE USE_SINGLE_PRECISION_PARTICLES=TRUE
Src/Particle/AMReX_ParticleHDF5.H
commit 39d03b864f3bad3de32988eb5351000447075344
Author: WeiqunZhang
Date: Wed Jul 22 10:49:16 2020 -0700
update CONTRIBUTING.md on how to fix git commits to have a clean history (#1164)
* update CONTRIBUTING.md on how to fix git commits to have a clean history
Co-authored-by: Axel Huebl
CONTRIBUTING.md
commit 5fa55019e580a70aa8c1a8f3f9c908f2d24ed5c2
Author: Mu-Hua Chien
Date: Wed Jul 22 12:24:24 2020 -0400
Semicoarsening nodal (#1158)
Src/LinearSolvers/MLMG/AMReX_MLABecLap_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_1D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_2D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLap_3D_K.H
Src/LinearSolvers/MLMG/AMReX_MLNodeLaplacian.cpp
Tutorials/LinearSolvers/NodalPoisson/MyTest.H
commit ecb5c574f5eebd2c63080b0f561f44db9fd27581
Author: WeiqunZhang
Date: Wed Jul 22 08:06:01 2020 -0700
Fix bug in #1159 (#1168)
Src/Base/AMReX_DistributionMapping.cpp
commit f87409ca66f212fa81315fcb77318f4a9131f63d
Author: mic84
Date: Wed Jul 22 00:58:15 2020 -0700
CMake: implement ability to export build tree (#1149)
* CMake: define C++ required flags as a property of target amrex
* CMake: fix handling of typechecker.
Just found out that:
1) We did not install typechecker
2) We do not need the typechecker path in the config file anymore
* CMake: enable exporting of build-tree
* CMake: linking against Flags_CXX_REQUIRED not needed anymore
* CMake: MSVC preprocessor flags must the public
* CMake: one more instance of Flags_CXX_REQUIRED to get rid of
* CMake: clean-up and add some comments
* Update Tools/CMake/AMReXInstallHelpers.cmake
Co-authored-by: Axel Huebl
Src/CMakeLists.txt
Tools/CMake/AMReXConfig.cmake.in
Tools/CMake/AMReXFlagsTargets.cmake
Tools/CMake/AMReXInstallHelpers.cmake
Tools/CMake/AMReX_Config.cmake
commit cc05b20149176b6260a903d84e89e5c073f49759
Author: WeiqunZhang
Date: Tue Jul 21 19:43:13 2020 -0700
add comments to hash version of RemoveDuplicates (#1162)
Src/Base/AMReX_Vector.H
commit a8db2cdf5be4e62e901b68913f1b192dde6fea8c
Author: Neil Carlson
Date: Tue Jul 21 19:54:57 2020 -0600
Reset amrex_geometry_module initialization state when finalizing (#1165)
Src/F_Interfaces/Base/AMReX_geometry_mod.F90
Src/F_Interfaces/Base/AMReX_init_mod.F90
commit 3d6877fb2dbb81de5ae048f8995065b8ff0fc9e0
Author: WeiqunZhang
Date: Tue Jul 21 15:56:05 2020 -0700
Remove Long version of Gatherv and reimplement AllGatherBoxes (#1161)
Src/Base/AMReX_Box.H
Src/Base/AMReX_Box.cpp
Src/Base/AMReX_IntVect.H
Src/Base/AMReX_ParallelDescriptor.H
commit a44b32f2d3aa92ddd33624361a60b85a2ae6f883
Author: WeiqunZhang
Date: Tue Jul 21 15:16:15 2020 -0700
GatherLayoutDataToVector and DistributionMapping (#1159)
Use int version of gatherv because the long version is not scalable. Use GatherLayoutDataToVector
in DistributionMapping because it is faster than ReduceSum.
Src/Base/AMReX_DistributionMapping.cpp
Src/Base/AMReX_ParallelDescriptor.H
commit e2da3ce40ab41cae399207e21e80df22584303c3
Author: Zach Jibben
Date: Mon Jul 20 14:50:41 2020 -0600
Add interfaces for bottom solve Hypre configuration (#1150)
* Add interfaces for bottom solve Hypre configuration
* Remove preserve_hypre_solver flag
This commit moves the Hypre and Petsc solver reset calls inside the
needsUpdate if statement, removing the need for the flag.
Src/Extern/HYPRE/AMReX_Hypre.H
Src/Extern/HYPRE/AMReX_HypreABecLap3.cpp
Src/LinearSolvers/MLMG/AMReX_MLMG.H
Src/LinearSolvers/MLMG/AMReX_MLMG.cpp
commit 27ac069c768ae5b90bbcdd6394d3070edb9b9c13
Author: WeiqunZhang
Date: Mon Jul 20 13:02:35 2020 -0700
Optimization of TagBoxArray::collate (#1156)
Use a new way to remove duplicated tags. Instead of removing duplicated from vector, we remove
duplicates when tags are still stored in FabArray. Because they are in FabArray with a BoxArray
that has no overlapped valid cells, we can use ParallelAdd to remove duplicates due to the overlap
of valid cells with ghost cells.
Assertion in AmrCore on blocking factor to make sure that there are no overlapped valid cells in
TagBoxArray after coarsening.
Use int version of Gatherv. The long version is not scalable.
Broadcast pre-maxsized BoxList instead of tags. In most cases, this should reduce the amount of
data to be broadcasted.
Src/AmrCore/AMReX_AmrMesh.cpp
Src/AmrCore/AMReX_Cluster.cpp
Src/AmrCore/AMReX_TagBox.H
Src/AmrCore/AMReX_TagBox.cpp
Src/Base/AMReX_BoxList.H
Src/Base/AMReX_BoxList.cpp
commit 9a48873f0addaeeac3fff323c5feb3554bf3e260
Author: Andrew Myers
Date: Sat Jul 18 14:16:36 2020 -0700
Combine two functions for periodically shifting particles into one and re-use (#1152)
* combine periodic shifts into one and re-use
* fix assertion
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_ParticleUtil.H
commit ba534dcd5f7b09285c95ed6bbcd18f49fe3922ea
Author: Andrew Myers
Date: Sat Jul 18 13:47:19 2020 -0700
remove the sumParticleMass function, which has been deprecated in favor of the generic reduction functions that also work on the GPU (#1153)
Src/Particle/AMReX_ParticleContainerI.H
Src/Particle/AMReX_Particles.H
commit d7de3d65aeea07683ecd698228925e505e2b8065
Author: jmsexton03
Date: Fri Jul 17 15:33:54 2020 -0400
Change hdf5 make syntax (#1151)
Tools/GNUMake/packages/Make.hdf5
commit fedc596633f0f2c52f5b9de7349adf30873d0f04
Author: WeiqunZhang
Date: Thu Jul 16 19:55:26 2020 -0700
add github action for 1D and 2D using configure (#1147)
.github/workflows/linux.yml
commit 62bee904a589df82e835da2a41a909955ff78db8
Author: WeiqunZhang
Date: Thu Jul 16 16:54:26 2020 -0700
Error on implicit capture of this by extended lambda by default (#1148)
Tools/GNUMake/Make.defs
Tools/GNUMake/comps/nvcc.mak
commit 824b23f6b26b787d86d6c67a5b55be1d00b574f5
Author: Axel Huebl
Date: Thu Jul 16 11:58:42 2020 -0700
CMake: More NVCC Controls (#1095)
* CMake: More NVCC Controls
Add control for more debug and warning flags with NVCC.
- [x] ptx debug infos: performance neutral and good
- [x] compilation timer: useful for interactions with Nvidia engineers
- [x] keep files: good for inspecting generated PTX code
- [x] warnings: accidental capture of this (CUDA11+)
- [x] device LTO: awesome new feature to try (CUDA11+)
- [x] unknown flags: just forward to host, make build more robust (CUDA10.2+)
* Docs: New CMake CUDA Options
Docs/sphinx_documentation/source/BuildingAMReX.rst
Tools/CMake/AMReX_SetupCUDA.cmake
commit 3efc8a9fe5aed2b82d5199dedf26baf772f40579
Author: WeiqunZhang