Single petscdir.mk
PetscUnlikelyDebug: add useful macro for conditionals that should be skipped in optimized mode
PetscDefined: remove many uses of #ifdef PETSC_USE_DEBUG
fix dangling pointer; add MatProductClear() as suggested by Lisandro
build: add more support for not needing to set PETSC_DIR yet build with PETSc, also support for GNU make to directly use "make test" instead of "make -f gmakefile.test"Still needs -include ../petsc
build: add more support for not needing to set PETSC_DIR yet build with PETSc, also support for GNU make to directly use "make test" instead of "make -f gmakefile.test"Still needs -include ../petscdir.mk in many of the makefilesStill needs the link for ../petscdir.mk in many directoriesGNUMakefile: call legacy makefile recursively instead of including itGNUMakefile: promote default "all" target; "libs" builds libpetsc*GNUMakefile: If gnumake < 4.2 prefer makefile wrapper interface over direct gmakefile interface to avoid the following warningsgmakefile.test:92: arch-ci-linux-cxx-cmplx-pkgs-64idx/tests/testfiles: No such file or directorygmakefile:67: arch-ci-linux-cxx-cmplx-pkgs-64idx/lib/petsc/conf/files: No such file or directoryCommit-type: testing-fix, feature, usabilityThanks-to: Jed Brown jed@jedbrown.org
show more ...
Support selective 64-bit mumpsA standard installation of MUMPS supports selective 64-bit. It uses int64_t for number for nonzeros(i.e., id.nnz, id.nnz_loc), and use int for indices of nonzeros (i.
Support selective 64-bit mumpsA standard installation of MUMPS supports selective 64-bit. It uses int64_t for number for nonzeros(i.e., id.nnz, id.nnz_loc), and use int for indices of nonzeros (i.e., id.irn, id.jcn, id.irn_loc and id.jcn_loc).Previously, one can only build petsc with 32-bit indices to work with mumps.This commit updates MUMPS interface to work with 64-bit-indices petsc.
Merge remote-tracking branch 'origin/maint'
Wrong description
PCView error with MUMPS, GAMG, and LMVM
Changed XXXPinToCPU() to XXXBindToCPU() to prevent confusion.The reason for this change is that we already use the terminology"pinned" to refer to memory that is non-pageable, in the context ofPe
Changed XXXPinToCPU() to XXXBindToCPU() to prevent confusion.The reason for this change is that we already use the terminology"pinned" to refer to memory that is non-pageable, in the context ofPetscSF as well as allocating host memory when GPUs are being employed.
bugfix for MatMumpsGetInverse_MUMPS()Reported-by: Jan Grießer <griesser.jan@googlemail.com>
MATSBAIJ: Fix Hermitian MatMult for Seq and MPI code pathswe can reuse the same MatMult operation with no impact for non-complex buildsRemove improper check for the hermitian complex case from Mat
MATSBAIJ: Fix Hermitian MatMult for Seq and MPI code pathswe can reuse the same MatMult operation with no impact for non-complex buildsRemove improper check for the hermitian complex case from MatGetFactor and move it where it actually belongsFix conversion routines from SBAIJ when the hermitian flag is set
Merge branch 'hzhang/mumps-errorcheck/maint' into 'master'Hzhang/mumps errorcheck/maintSee merge request petsc/petsc!2177
Wrong number of nz in MatGetInfo
Enable MatMatSolve(F,RHS,RHS) for some packagesRequested-by: Jaysaval, Piyoosh <Piyoosh.Jaysaval@pnnl.gov>
Propogate MUMPS return code within omp_comm. Also fixed the datatype and count in MPI_Bcast
Add error checking in mumps.cReported-by: Barry Smith
Rename valid_GPU_array/matrix to offloadmask
MATSOLVERMUMPS: use accessors routinesthis allows to use MUMPS with up-to-date VIENNACL matrices
minor
Tentative fix for SBAIJ MUMPS with bs=1.
Also handling MPISBAIJ matrices.
12345678910>>...25