requires: defined -> requires: defineMost of these tests where deactivated in https://gitlab.com/petsc/petsc/-/merge_requests/3744
Rename -mat_cusparse_transgen to -mat_form_explicit_transpose
Mat ex70: add test from Pierre
CI: arch-ci-linux-gcc-complex-opt.py -> arch-ci-linux-gcc-complex-opt-32bit.pyFix bug in ex4.c Code assumed that floating point product would produce exact result/spend 3hReported-by: Gitlab CI
Print additional information on PetscMallocValidate() errors
Add new mat/tests example for testing performance of BAIJ for different block sizesAdd support for options database controln of which version of MatMult_SeqBAIJ is usedAdd support for block size
Add new mat/tests example for testing performance of BAIJ for different block sizesAdd support for options database controln of which version of MatMult_SeqBAIJ is usedAdd support for block size 12 include AVX (no faster)Fixes for MatMult_SeqBAIJ_12_AVX2()- Code, was cut and paste and not properly fixed, so completely wrong beforeClean up test example for MatMult BAIJ blocksize 12Commit-type: optimization/spend 5.5hReported-by: Asiri Obeysekara <gitlab@mg.gitlab.com>
show more ...
Make QR test optional in mat/tests/ex1.cUntil all necessary implementations are presentDisable QR for viennacl test (not implemented)
Refactor LAPACK factorization routines used by MatSeqDenseCorrect MatSolveTranspose for recent changes in MatSeqDense
Add geqrf QR factorization for sequential dense matricesCorrect orthogonal matrix character code for complex numbers
Merge branch 'release' into main
ex237: PETSC_HAVE_MKL is not enoughReported-by: Eric Chamberland <Eric.Chamberland@giref.ulaval.ca>
Merge branch 'tisaac/feature-mat-permute-via-createsubmatrix' into 'main'Make MatPermute call MatCreateSubMatrix as a fallbackSee merge request petsc/petsc!3636
SF: add an MPI_Op argument to SFBcast
Make MatCreateSubMatrix_SeqDense match current interface of MatCreateSubMatrix and test permutation
PetscSFView() never called viewer for the specific type (bug), hence many output files were incorrect.Added PetscSFView() support for basic MPI implementation and binary viewer saves communication
PetscSFView() never called viewer for the specific type (bug), hence many output files were incorrect.Added PetscSFView() support for basic MPI implementation and binary viewer saves communication data to binary file.This new viewer shows exactly how much data is sent from each rank to the other ranksallowing fiting various performanace models to the data.This uses MatView() to save the PetscSF data, but since Mat is above PetscSF in the software stackthe code will not work with multiple PETSc libraries. How should this be managed if one wishes to preserve thetraditionl PETSc algorithm stack?Commit-type: bug-fix, featureFunded-by: brain projectProject: brain projectspend 1hUpdate sfbasic.c save the transpose of the matrix since that indicates the sends for each rank which is easier to understandEnable PetscSFView_Basic_PatternAndSizes only with single library
CHKERRQ() -> CHKERRMPI()
Mat ex70: restore test
Mat test ex132: Test UNKNOWN_NONZERO_PATTERNReported-by: Jose Roman <jroman@dsic.upv.es>
ci: disable mat_tests-ex70_1_par_cuda likely breakage from 039c6fba and perhaps masked by petscdiff prior to c2eed0ed06not ok diff-mat_tests-ex70_1_par_cuda+local-0_xgpu-0_bgpu-0_A_mat_type-mpiaijc
ci: disable mat_tests-ex70_1_par_cuda likely breakage from 039c6fba and perhaps masked by petscdiff prior to c2eed0ed06not ok diff-mat_tests-ex70_1_par_cuda+local-0_xgpu-0_bgpu-0_A_mat_type-mpiaijcusparse # Error code: 1
MatAXPY: add support for CUSPARSE
Mat ex62: add backend test
Mat ex62: add broken MKL test
MatAIJMKL: enable parallel PtaPWondering if it ever worked?
Mat test ex62: add MPICUSPARSE tests
MatSetValuesCOO: perform addition of repeated entries when INSERT_VALUES is specifieduse temporary data instead of persistent storage in cusparse implementation
1...<<11121314151617181920>>...24