PetscErrorStrings array: fix comments and trim to correct sizeAdd a missing comma, causing concatenation of the (empty) string on thenext line, and fix the numbering afterwards.Remove extra entr
PetscErrorStrings array: fix comments and trim to correct sizeAdd a missing comma, causing concatenation of the (empty) string on thenext line, and fix the numbering afterwards.Remove extra entries at the end of the array.
show more ...
PetscError Man Page: add missing newline for Developer Note section
Sys: Fix 'extern' declarations
Merge branch 'maint'
Fortran: detect gfortran-8 and use size_t for string lengthshttps://gcc.gnu.org/gcc-8/porting_to.html
Merge branch 'knepley/fix-plex-residual-loop'* knepley/fix-plex-residual-loop: Plex+SNES: Protect Destroy Plex+SNES/TS: Moved zeroing of Jacobian out of DMPlexComputeJacobian_Internal() - Shoul
Merge branch 'knepley/fix-plex-residual-loop'* knepley/fix-plex-residual-loop: Plex+SNES: Protect Destroy Plex+SNES/TS: Moved zeroing of Jacobian out of DMPlexComputeJacobian_Internal() - Should probably go in DMLocal versions, but now its in the top-level Plex FEM versions SNES+Plex: Needed cell offset, not cell number Sys: Added a check for gnome-terminal, which never uses -display - Added some docs Plex: Made a public interface for ComputeBdResidualSingle() Plex: The first argument to a pointfunc should be the coordinate dimension
Revert "remove dead code"This reverts commit e42d1bb89c832af7cff7ca5dc681a50206c18b0f.
remove dead codeCommit-type: style-fix
examples: add compile target 'ex%: ex%.o' etc - and remove the corresponding dups
Man pages: add newlines after "Notes:"This allows for proper formatting from sowing.On OS X (using gsed, not the default BSD sed), from the PETSc root directory: find src include -type f \(
Man pages: add newlines after "Notes:"This allows for proper formatting from sowing.On OS X (using gsed, not the default BSD sed), from the PETSc root directory: find src include -type f \( -name "*.c" -or -name "*.h" -or -name "*.cxx" \) | xargs gsed -i 's/Notes\s*:\s*\(\w.*\)/Notes:\n \1/'This adds a newline and 4 spaces whenever "Notes:" is followed by any "word" character, in any .c, .h, or .cxx file in src/ or include/
Sys: Added a check for gnome-terminal, which never uses -display- Added some docs
cleanup PETSC_DESIRE_FEATURE_TEST_MACROSCommit-type: style-fix
Add required PETSC_DESIRE_FEATURE_TEST_MACROS for certain system calls
Merge branch 'maint'* maint: PetscCheckPointerSetIntensity(): document -check_pointer_intensity MatGetRowIJ man page: minor cleanup
PetscCheckPointerSetIntensity(): document -check_pointer_intensity
proper docs for -check_pointer_intensityCommit-type: documentation
Remove handling of error code from PetscPClose() since some systems do not reliably return it.Commit-type: portability-fix
Windows does not have SIGHUP so skip compiling example on such systemsCommit-type: portability-fix, testing-fixReported-by: nightly tests
convert another directory to new test harnessCommit-type: testing-fix
fix runallen_cahn with -fp_trap on arch-linux-gcc-ifc-cmplx> [0]PETSC ERROR: FE_INVALID=0x1 FE_DIVBYZERO=0x4 FE_OVERFLOW=0x8 FE_UNDERFLOW=0x10 FE_INEXACT=0x20> [0]PETSC ERROR: [0] PetscDefaultFPTr
fix runallen_cahn with -fp_trap on arch-linux-gcc-ifc-cmplx> [0]PETSC ERROR: FE_INVALID=0x1 FE_DIVBYZERO=0x4 FE_OVERFLOW=0x8 FE_UNDERFLOW=0x10 FE_INEXACT=0x20> [0]PETSC ERROR: [0] PetscDefaultFPTrap line 357 /sandbox/petsc/petsc.next-2/src/sys/error/fp.c> [0]PETSC ERROR: [0] VecMDot line 1172 /sandbox/petsc/petsc.next-2/src/vec/vec/interface/rvector.c> [0]PETSC ERROR: [0] KSPGMRESClassicalGramSchmidtOrthogonalization line 43 /sandbox/petsc/petsc.next-2/src/ksp/ksp/impls/gmres/borthog2.c
Reported-by: Marco Schauer <m.schauer@tu-braunschweig.de>. Typo in help menu
Updates for checking for incompatible MPI libraries at runtimeRemove false positives due to MPI ABI initiativeNo longer call hard abort() on incompatibilityI welcome constructive criticism on ho
Updates for checking for incompatible MPI libraries at runtimeRemove false positives due to MPI ABI initiativeNo longer call hard abort() on incompatibilityI welcome constructive criticism on how to improve thisCommit-type: portability-fix, testing-fix,style-fix, featureThanks-to: jed@jedbrown.org, dalcinl@gmail.com, knepley@gmail.com
fortran: add SETERRA() and CHKERRA() macros as ifort flags 'return' in 'program main' as error/sandbox/petsc/petsc.clone-2/src/sys/examples/tutorials/ex4f90.F90(42): error #6353: A RETURN statement
fortran: add SETERRA() and CHKERRA() macros as ifort flags 'return' in 'program main' as error/sandbox/petsc/petsc.clone-2/src/sys/examples/tutorials/ex4f90.F90(42): error #6353: A RETURN statement is invalid in the main program. call MPI_Comm_size(PETSC_COMM_WORLD,size,ierr);if (ierr .ne. 0) then ; call PetscError(PETSC_COMM_SELF,ierr,PETSC_ERROR_REPEAT,PETSC_NULL_CHARACTER); return; endifAnd attempt to shorten the macros by using PetscErrorF() subroutineAnd fix all examples to use SETERRA/CHKERRA in main program (instead of SETERRQ/CHKERRQ)Note: CHKERRA are different from CHKERRABORT on the C interface - as it does not take 'comm' option
Give Fortran SETERRQ() and CHKERRQ() same functionality as C versionsThey can now produce proper error messages and stack tracesRequire .F90 files due to line length issuesCommit-type: featureF
Give Fortran SETERRQ() and CHKERRQ() same functionality as C versionsThey can now produce proper error messages and stack tracesRequire .F90 files due to line length issuesCommit-type: featureFunded-by: ECPTime: 4.2 hours
1...<<11121314151617181920>>...23