phynexis-deps/mfem
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
17,143 Commits 455 Branches 0 Tags
dcc80a5572ea2ae2bac2dae4643bdb8564c2c7e5
Commit Graph
408 Commits
Author SHA1 Message Date
Tzanio Kolev 2f6eb8838f Merge pull request #3007 from mfem/bugfix/rwa/coarsen-L2-proj 
Use local L2 projection for coarsening L2 spaces on non-conforming AMR meshes [bugfix/rwa/coarsen-L2-proj]
 2023-01-27 14:30:06 -08:00
Tzanio Kolev fb21cca26e Merge branch 'master' into bugfix/rwa/coarsen-L2-proj 2022-12-28 11:03:05 -08:00
Will Pazner 26073f789d Deprecate GetFaceGeometryType and GetFaceBaseGeometry 
Now both of these functions call GetFaceGeometry.

GetFaceGeometry has been updated to handle ghost faces.
 2022-12-06 12:21:12 -08:00
Robert W. Anderson 1504729937 Merge branch 'master' into bugfix/rwa/coarsen-L2-proj 2022-11-09 19:23:44 -08:00
Will Pazner 24825a5d26 Replace sprintf with snprintf 2022-11-05 13:38:42 -07:00
Veselin Dobrev 1482864193 Remove the need to explicitly call EnsureMultTranspose in the 
classes SparseMatrix and HypreParMatrix.

Added new methood HypreParMatrix::ResetTranspose.

Added some crarifying doxygen documentation.
 2022-10-20 16:03:23 -07:00
luz paz 476c855bd7 Fix various typos 
Found via `codespell -q 3 -L allright,ba,equil,esy,fo,hda,lod,nd,ned,numer,ot,pres,ro,seh,shat,solfes,strat,tbe,te,warmup`
 2022-08-05 20:50:51 -04:00
Veselin Dobrev d70bf954ae Merge pull request #3035 from mfem/pmesh-to-serial-mesh 
Save a ParMesh as a serial mesh
 2022-08-02 11:40:22 -07:00
Veselin Dobrev 9592ee5ab6 Fix the HYPRE version where hypre_ParCSRMatrixLocalTranspose() was 
introduced.

Add a method HypreParMatrix::EnsureMultTranspose() that calls
hypre_ParCSRMatrixLocalTranspose() when HYPRE is built for GPUs.

Use HypreParMatrix::EnsureMultTranspose() in the implementation of
class ParFiniteElementSpace when constructing the parallel
prolongation operator as a HypreParMatrix.
 2022-07-27 16:16:44 -07:00
Veselin Dobrev f2f489a36e Updated ParMesh::GetSerialMesh() with some overflow checks and 
some small tweaks.

Replace some uses of 'long' with 'long long' to better support
Win64 builds where 'long' is 32-bit and 'long long' is 64-bit.
On Linux and MacOS, both types are typically 64-bit.

Updated various "MemoryUsage" methods to return 'std::size_t'
instead of 'long' since the latter is 32-bit in Win64 builds.

Updated CHANGELOG.
 2022-07-26 16:09:12 -07:00
Arturo Vargas 0aaf837463 add version guards 2022-06-24 08:56:50 -07:00
Arturo Vargas aee9379f34 disable hypre_ParCSRMatrixBooleanMatvecT and make style 2022-06-10 13:31:27 -07:00
Arturo Vargas d539dc6fb3 optimize At application 2022-05-27 17:38:44 -07:00
Robert W. Anderson 649346bdce minimize diff 2022-05-19 20:28:40 -07:00
Robert W. Anderson d2bba7e1bb fuse DG and CG loops in parallel restriction assembly 2022-05-19 20:27:00 -07:00
Robert W. Anderson a9e0a32888 small cleanups and make style 2022-05-19 16:16:45 -07:00
Robert W. Anderson ef2fb0dfd2 create loop for offdiag parts of discontinuous case in parallel coarsening matrix assembly 2022-05-19 16:12:44 -07:00
Robert W. Anderson 73a8e67efc make style 2022-05-19 11:56:50 -07:00
Robert W. Anderson 1224c30c06 split parallel global assembly of the diagonal part of the derefinement matrix into continous and discontinuous cases, analogous to fespace.cpp; one processor case now works 2022-05-19 11:53:29 -07:00
Will Pazner 4d4015785d LOR bugfix when restriction matrix is non-NULL 
Ensure R and P are built when calling CopyProlongationAndRestriction
 2022-03-28 22:53:59 -07:00
Will Pazner 892d896c06 Remove unneeded UseDevice call 2022-03-12 15:29:44 -08:00
Tzanio 6d2530775b Extended copyright to 2022 2022-02-25 10:24:50 -08:00
Tzanio Kolev 232d46ac77 Merge pull request #2731 from mfem/shadow-out-dev 
Fixing shadow warnings about "out/err" globals [shadow-out-dev]
 2022-02-23 11:55:25 -08:00
Stowell, Mark L f07e137d89 Merge remote-tracking branch 'origin/master' into shadow-fem-dev 
# Conflicts:
#	fem/prestriction.cpp
#	fem/restriction.cpp
 2022-02-18 09:55:58 -08:00
Will Pazner 14b48da4d2 Merge remote-tracking branch 'origin/master' into shadow-out-dev 2022-02-18 08:20:52 -08:00
Veselin Dobrev 3da0d9fc2f Merge branch 'master' into yohann/pa-dg-ncmesh (#2195) 2022-01-14 19:45:55 -08:00
Stowell, Mark L bc76377bda Changes based on feedback from @vladotomov 2021-12-30 10:57:05 -08:00
Stowell, Mark L 00622b78fc Fixing shadow variables in fem classes 2021-12-23 16:32:09 -08:00
Stowell, Mark L 512a889556 Missed two instances of "out" 2021-12-22 15:25:25 -08:00
Veselin Dobrev 8b0262bd29 Fix an issue where some boundary dofs are not properly marked 
as essential in ParFiniteElementSpace::GetEssentialVDofs() when
using a parallel non-conforming mesh.

This should address GitHub issue #2559.
 2021-12-16 02:53:09 -08:00
Yohann 301d0c3b39 Merge branch 'master' into yohann/pa-dg-ncmesh 2021-11-11 19:12:32 -08:00
Yohann Dudouit 46e21def4e Merge branch 'master' into yohann/pa-dg-ncmesh 2021-10-06 15:29:42 -07:00
Tzanio f491fe30c3 Merge branch 'master' into lor-solvers-p-ref 2021-09-28 16:31:10 -07:00
Sebastian Grimberg cd9d131acd Merge branch 'master' into hypre-bigint-fixes 2021-09-21 13:47:11 -07:00
Jakub Červený 2b90ec7c07 Add a check to prevent bad results on parallel shared faces in case 
of Nedelec NC tets of order >= 2.
 2021-09-10 13:24:11 +02:00
Yohann Dudouit 18df4d9a31 Add class ParNCH1FaceRestriction. 2021-09-02 16:56:58 -07:00
Yohann Dudouit a34fcc4beb Merge branch 'master' into yohann/pa-dg-ncmesh 2021-08-30 13:12:11 -07:00
Sebastian Grimberg af21f7258e Fix HYPRE_BigInt bugs for HYPRE_BIGINT and HYPRE_MIXEDINT 2021-08-25 22:01:49 -07:00
Tzanio 7feb27958b Merge branch 'master' into hypre-fix 
Conflicts:
	CHANGELOG
 2021-08-20 16:58:41 -07:00
victor 1361e3907d Update with new fix based on hypre's version 2021-08-19 17:18:21 -07:00
Will Pazner ca01ef44bf Merge branch 'ads-cuda-use-jacobi' into lor-solvers-p-ref 2021-08-18 17:28:48 -07:00
victor baf09ea081 Update calls according to latest changes in hypre 2021-08-18 13:27:20 -07:00
Stowell, Mark L d13f3903e6 Fixing memory leak in ParFiniteElementSpace::Update 2021-08-07 13:45:14 -07:00
Stowell, Mark L ef1ddbceea Merge remote-tracking branch 'origin/master' into no-tet-reorder-dev 2021-07-30 11:06:20 -07:00
Will Pazner 74710220a1 Remove need (mostly) for LOR DOF permutations 
Build the permutation into the P and R operators of the LOR space, so that the
true DOF numbering corresponds to the same true DOF number of the high-order
space. This means the LORSolver does not need to perform any permutation, since
it is incorporated into the RAP.
 2021-06-30 18:30:59 -07:00
Will Pazner 0a9b0c681d Handle LOR permutation construction in variable-order case 2021-06-30 18:30:59 -07:00
Stowell, Mark L 6c687ff50c Merge remote-tracking branch 'origin/master' into no-tet-reorder-dev 2021-06-29 13:31:19 -07:00
Veselin Dobrev 26c2dd70f9 Merge branch 'master' into hypre-cuda-dev 2021-06-25 12:36:31 -07:00
Veselin Dobrev e0a65be94b Fix serial build. 
Fix some issues seen when running 'punit_tests'.
 2021-06-18 12:33:56 -07:00
Yohann Dudouit c869398f60 Add a FaceRestriction interface. 2021-06-11 15:49:49 -07:00