febasic: fix logging flops to avoid int32 overflow and use CHKERRQReported-by: Valeria Barra <valeria.barra@colorado.edu>Reported-by: Jeremy Thompson <jeremy.thompson@colorado.edu>
febasic: perform change of basis using GEMM* Adds LAPACKREAL and BLASREAL getrf, getri, gemm* This reduces setup time for my example 3D mesh using Q_6 hexes from 160 seconds to 6 seconds.
Doc: Text fixes
FE+DS: Introduce PetscTabulation to hold tabulated function data- This was created to prevent mismatches in sizes I was getting by justpassing around arrays- FE+FV+DS now hold tabulations- GetTab
FE+DS: Introduce PetscTabulation to hold tabulated function data- This was created to prevent mismatches in sizes I was getting by justpassing around arrays- FE+FV+DS now hold tabulations- GetTabulation() --> CreateTabulation() and it creates an object ratherthan using DMGetWorkArray(). I put in ComputeTabulation() to reusespace, but there still may be performance regression.- Default tabulations are now "cell tabulations"- The replicas in the tabulation are used to index faces in a facetabulation
show more ...
PetscFE: Init memory properly for boundary integral- Also check for missing pointfunctions
Merge branch 'balay/glci-werror' into 'master'gitlab-ci: use -Werror with more testsSee merge request petsc/petsc!2075
fix compile warning from linux-viennacl/sandbox/balay/petsc/src/dm/dt/fe/impls/opencl/feopencl.c:492:10: warning: ‘y’ may be used uninitialized in this function [-Wmaybe-uninitialized] if (*x *
fix compile warning from linux-viennacl/sandbox/balay/petsc/src/dm/dt/fe/impls/opencl/feopencl.c:492:10: warning: ‘y’ may be used uninitialized in this function [-Wmaybe-uninitialized] if (*x * *y != Nblocks) SETERRQ2(PETSC_COMM_SELF, PETSC_ERR_ARG_SIZ, "Could not find partition for %d with block size %d", N, blockSize); ^
Fix bug introduced in testing, cannot rebase it away because some giant merge is in the wayCommit-type: bug-fixReported-by: gitlab-ci
Check return code of LAPACK calls
PetscFE: These are needed by the OpenCL implementation
PetscFE: Fix documentation
PetscFEOpenCLLogResidual: fix stage declaration
Add PetscArraycmp(), PetscArraycpy(), PetscArrayzero(), PetscArraymove(), update code to use new formsReplace a couple multi-mallocs with a single mallocReplace backwards loops in MatSetValues_*()
Add PetscArraycmp(), PetscArraycpy(), PetscArrayzero(), PetscArraymove(), update code to use new formsReplace a couple multi-mallocs with a single mallocReplace backwards loops in MatSetValues_*() with PetscArraymove()Commit-type: style-fix, feature
fix -Wmaybe-uninitialized warnings with arch-c-exodus-dbg-builder
Plex+FE: Fixed improper reference for coordinates
PetscFE+Plex: Use cell and face geometry in boundary integration
FE+OpenCL: Updated to new API
Plex+FE: Moved internal functions from header to fe.cSuggested-by: Jed Brown <jed@jedbrown.org>
Plex: Implemented pullback/pushforward - Added k to PetscDualSpace. It indicates what kind of transformation we use. - DualSpace does pullback and FE integration routines do pushforward - All integra
Plex: Implemented pullback/pushforward - Added k to PetscDualSpace. It indicates what kind of transformation we use. - DualSpace does pullback and FE integration routines do pushforward - All integration now takes a PetscFEGeom argument and pushes forward correctly - Added PetscDualSpaceTransformType - Removed TransformF() because we now transform the basis function instead - Chaged workspace management in DS - Updated L2 diff functions - Updated interpolation functions - Added way to output exact solution in DMSNESCheck()
Plex: Small fixes
PetscFE: Initialize variable to shut up idiot compilers without a proper use/disuse chain
PetscFE: If the aux field is on a boundary mesh, so we are not using face quadrature, index tabulation by only the quad point number
PetscFE: Better output for ASCII viewer
doc: fix LOCDIR wrt broken URL from MatHYPRESetPreallocation.htmlReported-by: Fande Kong <fdkong.jd@gmail.com>Tested with:make ACTION=sourcelist BASE_DIR=$PWD alltree_makefilels `cat files` > /
doc: fix LOCDIR wrt broken URL from MatHYPRESetPreallocation.htmlReported-by: Fande Kong <fdkong.jd@gmail.com>Tested with:make ACTION=sourcelist BASE_DIR=$PWD alltree_makefilels `cat files` > /dev/null
Merge branch 'master' into knepley/feature-pc-patch* master: (131 commits) fix mpiuni test PETSCFEOPENCL: declare PetscFEGetTabulation_Basic as external PETSCFEOPENCL: declare PetscFESetUp_Ba
Merge branch 'master' into knepley/feature-pc-patch* master: (131 commits) fix mpiuni test PETSCFEOPENCL: declare PetscFEGetTabulation_Basic as external PETSCFEOPENCL: declare PetscFESetUp_Basic as external one more fix jenkins: run 'cleantest' after examples are run - to recover diskspace logical order bug in diagbrdn update fixed, tests updated again replace new -petscspace_order instances with -petscspace_degree PETSCSPACETENSOR: clear up naming confusion of subspaces regression.py: remove configure code related to old test suite - now that SLEPc no longer needs it PetscSpaceSetup_Poly(): make idempotent Plex ex3: update tests Lift changes to defunct dtfe.c from [PR #1074] PetscSpace_Tensor: clarification of vector-valued PetscSpace_Tensor: allow height subspaces, fixes tabulation initialization PetscSpace: add a dimension member PetscSpaceSetDegree(): now take maxDegree argument -petscspacepoly_tensor -> -petscspace_type tensor PetscSpaceEvaluate_Polynomial(): fix temp array dependencies PETSCSPACETENSOR: A generic tensor product of spaces. PetscSpace: -petscspace_order -> -petscspace_degree ...Conflicts: src/dm/dt/interface/dtfe.c src/snes/examples/tutorials/ex13.c src/snes/examples/tutorials/ex17.c src/snes/examples/tutorials/ex62.c src/snes/examples/tutorials/output/ex62_bddc_quad.out src/snes/examples/tutorials/output/ex62_fetidp_2d_quad.out src/snes/examples/tutorials/output/ex62_fetidp_2d_tri.out src/snes/examples/tutorials/output/ex62_fetidp_3d_hex.out src/snes/examples/tutorials/output/ex62_fetidp_3d_tet.out
1234567