Tao: Replace 0 -> NULL for pointers
Single petscdir.mk
Merge remote-tracking branch 'origin/maint'
gcc/gfortran-10: fix errors/warningsThe primary issue now is - gfortran defaults to flagging argument mismatch as error.We can work-around this with '-fallow-argument-mismatch' option - however th
gcc/gfortran-10: fix errors/warningsThe primary issue now is - gfortran defaults to flagging argument mismatch as error.We can work-around this with '-fallow-argument-mismatch' option - however this switches from errors to warnings.And the only way to disable this warning is '-w' i.e disable all warnings.Fixes involve adding interface definitions for all prototypes that get used. [and work-around mpich by using a(1) for scalars]
show more ...
Fix TaoConvergedReasons
CI: fixes for arch-ci-linux-cuda-double-64idx test suite
Fix broken PARDISO --with-precision=single and add a test to CI
Merge branch 'maint'
Typos
DM: Change examples to use DMAddBoundary() instead of PetscDSAddBoundary()- Now you have to call DMCopyDS() after labels have been established on the coarse DM- Also, p4est setup has to have happen
DM: Change examples to use DMAddBoundary() instead of PetscDSAddBoundary()- Now you have to call DMCopyDS() after labels have been established on the coarse DM- Also, p4est setup has to have happened before label completion
PC: ASM, Block Jacobi: when viewing, note local solver on rank 0 is shownThe KSPView() call shows the (sub-)KSP on rank 0, which should have the sametype, PC, parameters, etc., but will not be the
PC: ASM, Block Jacobi: when viewing, note local solver on rank 0 is shownThe KSPView() call shows the (sub-)KSP on rank 0, which should have the sametype, PC, parameters, etc., but will not be the identical solver,as the size of the local subdomains and the entries of the localsubmatrices will be different. For instance, the output only showsthe local Mat for rank 0, which might confuse a user into thinking thatthe subdomains are all of the same size when they are not. Similarly,a PC like LU would show different factorization details on differentranks, even though the solver settings are the same.Make the viewer output more precise by noting explicitly that isdisplaying information for the rank 0 subsolver.Update test output with the new string.
PetscOptionsBegin: Title argument made obligatoryto avoid (null) headers in -help output.Additionally* PetscOptionsBegin() inarg validation* fix a few examples which break this* fix TaoMatGetS
PetscOptionsBegin: Title argument made obligatoryto avoid (null) headers in -help output.Additionally* PetscOptionsBegin() inarg validation* fix a few examples which break this* fix TaoMatGetSubMat() to use PetscObjectOptionsBegin() instead of PetscOptionsBegin()* fix weird position of PetscOptionsEnd() in TAO examples
TAOPDIPM: Add primal-dual interior point method solver in TAO
DM+Plex: Example Fixes for cell type rewrite
docs: fix Input/Output Parameters list formatReported-by: Pierre Jolivet <pierre.jolivet@enseeiht.fr>
docs: fix sowing list syntax/incomplete man page
fixing manpages for Tao IS utilities
refactored LMVM matrix names for Broyden-class methods
SNESQN harmonization with MATLMVM objects
Merge branch 'hongzh/fix-tstrajectory-examples' into 'master'Adjust the TSTrajectory setup in adjoint examplesSee merge request petsc/petsc!2609
Provide support for not setting PETSC_DIR when gnumake is being used; this will prevent some petsc-maint/petsc-users email threadsCommit-type: error-checking, testing-fix, feature, exampleThanks-t
Provide support for not setting PETSC_DIR when gnumake is being used; this will prevent some petsc-maint/petsc-users email threadsCommit-type: error-checking, testing-fix, feature, exampleThanks-to: Jed Brown <jed@jedbrown.org>Thanks-to: Lisandro Dalcin <dalcinl@gmail.com>
Merge branch 'jed/promote-examples-tests-tutorials' [petsc/petsc!2610]* jed/promote-examples-tests-tutorials: Promote examples/{tests,tutorials}/ to {tests,tutorials}/
1...<<21222324252627282930>>...62