change cp to ${CP} in a makefile rule for portability

Merge branch 'barry/2025-07-20/no-petsc-options-make-check' into 'main'

Skip PETSC_OPTIONS with make check

See merge request petsc/petsc!8570

Skip PETSC_OPTIONS with make check

Some users may have a complicated PETSC_OPTIONS they use for their code but that results in extra output when running make check that confuses/upsets them and they

Skip PETSC_OPTIONS with make check

Some users may have a complicated PETSC_OPTIONS they use for their code but that results in extra output when running make check that confuses/upsets them and they do not associate the message with PETSC_OPTIONS. So ignore any PETSC_OPTIONS that is set

In order for the .gitlab-ci.yml file to continue to work we use EXTRA_OPTIONS to provide options instead of PETSC_OPTIONS

Yet another example how environmental variables are the work of the devil since they are not obvious when being used

Thanks-to: Satish Balay

show more ...

Merge branch 'barry/2025-07-25/change-makealletags-to-makeetags' into 'main'

Change make alletags to make etags since it always makes all of them

See merge request petsc/petsc!8593

Change make alletags to make etags since it always makes all of them

Merge branch 'barry/2025-07-02/fortran-add-fortitude-linter' into 'main'

fortran add fortitude linter

See merge request petsc/petsc!8522

Add make fortitude to lint the Fortran code.

Thanks-to: Blaise Bourdin

Merge branch 'jose/fix-make-install' into 'main'

makefile: avoid showing verbose info at 'make install'

See merge request petsc/petsc!8523

makefile: avoid showing verbose info at 'make install'

Whenever PETSC_POST_INSTALLS is empty, 'make install' will fall down
to 'make info', printing lots of stuff which is confusing

Merge branch 'barry/2025-05-21/fix-post-build-install' into 'main'

Rework handling in configure/build handling of building/installing of packages...

See merge request petsc/petsc!8434

Rework handling in configure/build handling of building/installing of packages that must be built after PETSc

Add package.addPost() to register compile/install of external software to be done after

Rework handling in configure/build handling of building/installing of packages that must be built after PETSc

Add package.addPost() to register compile/install of external software to be done after PETSc is compiled/installed

Add package.addMakeCheck() to register make check for external software

Can now removed hardwired listing of rules for bamg, hpddm, etc from the PETSc makefiles

Much easier to add new packages that require post compiler/install makes

show more ...

Merge branch 'barry/2025-05-27/docs-build-in-petsc-arch' into 'main'

Build the docs files in $PETSC_ARCH when

See merge request petsc/petsc!8444

Build the docs files in $PETSC_ARCH when

PETSC_ARCH is set and make docs is used

Otherwise the docs continue to be built in the doc directory.

Does not apply to petsc4py docs

Merge branch 'barry/2025-05-03/add-dolphinx-package' into 'main'

add --download-dolfinx

See merge request petsc/petsc!8371

Add support for --download-dolphinx and its needed packages

Also add saving of pythonpath and PYTHONPATH in pkgconfig and petscvariables for users to access

Merge remote-tracking branch 'origin/release'

Merge branch 'balay/ci-u24-migrate' into 'release'

CI: migrate Ubuntu-20.04 jobs to Ubuntu-24.04

See merge request petsc/petsc!8328

CI: migrate Ubuntu-20.04 jobs to Ubuntu-24.04

- require gcovr-5.2 on all jobs [upgrading some jobs to latest v8.3 triggers errors in stage-4 'coverage-analyze' job
- gcov triggers errors - and requi

CI: migrate Ubuntu-20.04 jobs to Ubuntu-24.04

- require gcovr-5.2 on all jobs [upgrading some jobs to latest v8.3 triggers errors in stage-4 'coverage-analyze' job
- gcov triggers errors - and requires --gcov-ignore-parse-errors
- --gcov-ignore-parse-errors prints tons of warnings triggering 'Job's log exceeded limit ...' message in CI. So hack installed gcovr to disable these messages
- linux-gcc-quad-64idx-dbg triggers 'diff-snes_tutorials-ex78_2' - so add in alt output file for this change

show more ...

Merge remote-tracking branch 'origin/release'

Merge branch 'balay/fix-PetscAddLogDouble' into 'release'

fix build error with --with-threadsafety --with-log=0

See merge request petsc/petsc!8304

CI: gcovr: use --exclude '.*/petscsys.h'

Avoids govr errors with this code change (with arch-ci-linux-cuda-uni-pkgs):

upgrading gcovr from v5.2 to v8.3 gives useful messages

AssertionError: Got fu

CI: gcovr: use --exclude '.*/petscsys.h'

Avoids govr errors with this code change (with arch-ci-linux-cuda-uni-pkgs):

upgrading gcovr from v5.2 to v8.3 gives useful messages

AssertionError: Got function PetscStrHash(char const*) in /scratch/balay/petsc.x/include/petscsys.h on multiple lines: 2777, 2778.
You can run gcovr with --merge-mode-functions=MERGE_MODE.

However gcovr v8.3 is causing many other job failures. Hence sticking with v5.2 [via: --exclude '.*/petscsys.h']

show more ...

generatefortranstubs: move out of maint (to config/utils)- as its needed in tarball

Merge branch 'barry/2024-04-29/add-fortran-enums' into 'main'

Major updates to Fortran interface

See merge request petsc/petsc!7517

Fortran 90: fully embrace After 34 years!

- deprecate use of 'F90' in Fortran function names
- use Fortran pointers when appropriate
- the new Fortran API is not backward compatible with previous ve

Fortran 90: fully embrace After 34 years!

- deprecate use of 'F90' in Fortran function names
- use Fortran pointers when appropriate
- the new Fortran API is not backward compatible with previous versions!
- also clean up inconsistent PETSc code detected by new Fortran generation tools
- drop use of bfort
- automatically generate all the Fortran PETSc objects, enums etc from the include files
- generate most of the Fortran interface definitions and functions from the source code
- simplify the number and organization of Fortran modules

Co-authored-by: Jose E. Roman <jroman@dsic.upv.es>

show more ...

Merge remote-tracking branch 'origin/release'