Fix a bug in Interpolate_Alpha

- The interpolation implementation reuses th->V1 by mistake
- th->V1 is updated once SNESTSFormFunction() is called
- -snes_monitor triggers an addition call to SNESTS

Fix a bug in Interpolate_Alpha

- The interpolation implementation reuses th->V1 by mistake
- th->V1 is updated once SNESTSFormFunction() is called
- -snes_monitor triggers an addition call to SNESTSFormFunction() to compute the norm, changing the value of V1

show more ...

bjkokkos: cleanup diagnostics

Merge remote-tracking branch 'origin/release'

Tests: disable tests for sycl

Revert "Landau: temporarily disable it for SYCL due to link errors with Intel OneAPI"

This reverts commit c36a59bee2e66f2e80e3e9e7603aedea4197893c.

This requires some changes to user FORTRAN code

Pass PETSC_NULL_XXX_ARRAY when passing a NULL as an argument that is expecting an array
Pass PETSC_NULL_ENUM when argument returns an enum instead of

This requires some changes to user FORTRAN code

Pass PETSC_NULL_XXX_ARRAY when passing a NULL as an argument that is expecting an array
Pass PETSC_NULL_ENUM when argument returns an enum instead of PETSC_NULL_INTEGER
Pass arrays (and not scalar values) when the argument is expecting an array; this means replace, for example, the argument v with [v]
Use PetscObjectIsNull(obj) to check if the object is NULL, instead of obj == PETSC_NULL_XXX

The compiler will now automatically prevent you from using the wrong argument type for the first three bullets above
This will also require an update sowing with the new sowing branch this MR is using.

show more ...

All constructors should do PetscAssertPointer on output object

Reported-by: Blaise Bourdin

Constructors should not initialize the output to NULL upon entry, serves no purpose

Fix some formatting i

All constructors should do PetscAssertPointer on output object

Reported-by: Blaise Bourdin

Constructors should not initialize the output to NULL upon entry, serves no purpose

Fix some formatting in constructors

show more ...

Merge remote-tracking branch 'origin/release'

TS ex2: Replace swarm ex9 with hamiltonian ex2

- Parallelize initialization and primal implementation
- Add test based output, mpiu vs callmpi, fix global particle number
- Fix parallelization for m

TS ex2: Replace swarm ex9 with hamiltonian ex2

- Parallelize initialization and primal implementation
- Add test based output, mpiu vs callmpi, fix global particle number
- Fix parallelization for mixed form, require superlu_dist for test case

show more ...

TS: BSI should propagate timestep and time to subsolvers

Merge branch 'jolivet/cleanup' into 'main'

Remove unneeded files and update CMake

See merge request petsc/petsc!7585

Update sowing bfort for automatic generation of XXXDestroy() code

Fix some Fortran stub code to work with generation of XXXDestroy()

Add some runtime tests for XXXCreate() calls

Also update sowing

Update sowing bfort for automatic generation of XXXDestroy() code

Fix some Fortran stub code to work with generation of XXXDestroy()

Add some runtime tests for XXXCreate() calls

Also update sowing and fortranimpl.h to use PETSC_NULLPTR instead of NULL

show more ...

Remove unneeded files that don't compile

Shorten tests

TSDIRK: allow non-consistent dual variables

this requires the algebraic IS during the residual callback
add tests

Merge remote-tracking branch 'origin/release'

Docs: fix double backticks in seealso

Update to sowing version that supports PetscViewer and PetscObject

Remove manual Fortran stubs for PetscViewer and PetscObject arguments since now handled by sowing

Also handles the passing of null

Update to sowing version that supports PetscViewer and PetscObject

Remove manual Fortran stubs for PetscViewer and PetscObject arguments since now handled by sowing

Also handles the passing of null objects correctly

show more ...

Merge remote-tracking branch 'origin/release'

Fix gcc-14 -Warray-bounds warnings for hypothetical negative sizes

These warnings arise because we use (signed int32) PetscInt for sizes
that must be nonnegative, then cast to size_t. This leads to

Fix gcc-14 -Warray-bounds warnings for hypothetical negative sizes

These warnings arise because we use (signed int32) PetscInt for sizes
that must be nonnegative, then cast to size_t. This leads to warnings
when the compiler doesn't know that the integers can't be negative.

In function ‘PetscMemzero’,
inlined from ‘TSARKIMEXRegister’ at /home/jed/petsc/src/ts/impls/arkimex/arkimex.c:1182:10:
/home/jed/petsc/include/petscstring.h:758:3: warning: ‘memset’ pointer overflow between offset 0 and size [-17179869184, -8] [-Warray-bounds=]
758 | memset(a, 0, n);
| ^~~~~~~~~~~~~~~

Compare commit 095c51fa7c16ea848964bdaab3c2c2b53b8a3072 from 2023:
Fix gcc-13 -Warray-bounds warnings involving hypothetical negative sizes

show more ...

Update source code removing all unneeded /*@C and associated manual stubs and interfaces

Merge remote-tracking branch 'origin/release'

Update ts.c

Merge remote-tracking branch 'origin/release'

Typos