1# Changes: 3.17 2 3% STYLE GUIDELINES: 4% * Capitalize sentences 5% * Use imperative, e.g., Add, Improve, Change, etc. 6% * Don't use a period (.) at the end of entries 7% * If multiple sentences are needed, use a period or semicolon to divide sentences, but not at the end of the final sentence 8% * Use full function names, for ease of searching and so that man pages links are generated 9 10```{rubric} General: 11``` 12 13- PETSc now requires a C99 compliant C compiler in all cases. Previously C99 was only required when building PETSc, but this now extends to public interfaces and header-files 14- PETSc now requires a C++11 compliant C++ compiler. Note this requirement is only enforced if C++ is used; it is acceptable to have a compiler that does not support C++11 if you only ever build C source 15- PETSc now requires at least Microsoft Visual Studio 2015 when using the Microsoft Visual C/C++ Compiler 16- Change `MPIU_Allreduce()` to always returns a PETSc error code that should be checked with `CHKERRQ(ierr)` or `PetscCall()` 17 18```{rubric} Configure/Build: 19``` 20 21- Change minimum value of `--with-cxx-dialect` argument from "03" to "11" 22- C++ dialect will now also be inferred from compiler flags, although users will be warned that they should let PETSc auto-detect the flag when setting the dialect this way 23- Change C++ dialect flag option to be consistent with compiler flags; `--with-cxx-dialect=gnu++14` means you want `-std=gnu++14`, no more, no less 24- Fix for requesting no C++ dialect flag via `--with-cxx-dialect=0`. Previously `configure` would bail out immediately without running the tests and therefore wouldn't set any of the capability defines. `configure` now runs all tests, just doesn't add the flag in the end 25- Fix a number of corner-cases when handling C++ dialect detection 26- Remove deprecated `PETSC_VERSION_PATCH` so as to not have confusion with patch releases where the subminor version changes 27- Change `PETSC_HAVE_MKL` to `PETSC_HAVE_MKL_LIBS` 28- Add `PETSC_HAVE_MKL_INCLUDES` 29- Enable HYPRE GPU for 64bit indices build (using HYPRE's mixed-int configuration) 30- Reuse PETSc-installed hwloc when installing Open MPI 31 32```{rubric} Sys: 33``` 34 35- Add `MPI_Comm_get_name()` and `MPI_Comm_set_name()` to MPIUNI 36- Remove `petsccublas.h` and `petschipblas.h` 37- Remove `-petsc_use_default_null_stream` and `-[cuda|hip]_synchronize` options 38- Remove `PetscCUDASynchronize` and `PetscHIPSynchronize`. Their operation is now managed by `PetscDeviceContext` via its `PetscStreamType` attribute 39- Remove `PetscCUDAInitialize()`, `PetscCUDAInitializeCheck()`, `PetscHIPInitialize()`, and `PetscHIPInitializeCheck()`. Their function is now handled by `PetscDeviceInitialize()` and `PetscDeviceInitialized()` 40- Remove `PetscCUBLASInitializeHandle()`, `PetscCUSOLVERDnInitializeHandle()`, `PetscHIPBLASInitializeHandle()`, and `PetscHIPSOLVERInitializeHandle()`. Their function is now handled implicitly by `PetscDeviceContext` 41- Remove `petsc_gputimer_begin` and `petsc_gputimer_begin` 42- Add `-device_enable`, `-device_select` and `-device_view` startup-options to control coarse-grained device initialization strategy 43- Replace `-[cuda|hip]_device` with split options `-device_enable_[cuda|hip]` and `-device_select_[cuda|hip]` to enable fine-grained control of device selection and initialization strategy 44- Replace `-[cuda|hip]_view` with `-device_view_[cuda|hip]` 45- Add `PetscDeviceInitType` to enumerate PETSc device initialization strategies 46- Add `PetscDeviceInitialize()` to eagerly initialize a `PetscDeviceType`, and `PetscDeviceInitialized()` to query the corresponding initialization state 47- Change `PetscDeviceCreate()` to also accept a `PetscInt devid`, to create a `PetscDevice` for a specific device 48- Add `PetscDeviceView()` 49- Move `PetscInt64_FMT` and `MPIU_INT64` definitions to `petscsystypes.h` 50- Add `PetscBLASInt_FMT`, `PETSC_MPI_COMM_FMT`, and `PETSC_MPI_WIN_FMT` format specifiers 51- Add `petscmacros.h` header to house common PETSc preprocessor macros 52- Add `PetscUnreachable()` to indicate unreachable code section to compiler 53- Add `PetscHasAttribute()` macro to query for existence of an `__attribute__` specifier 54- Add `PetscCommGetComm()` and `PetscCommRestoreComm()` to allow reuse of MPI communicator with external packages, as some MPI implementations have broken `MPI_Comm_free()` 55- Add `PetscExpand()`, `PetscConcat()`, `PetscCompl()`, and `PetscExpandToNothing()` 56- Add `PETSC_CONSTEXPR_14`, `PETSC_NULLPTR`, and `PETSC_NODISCARD` 57- Add `PetscSizeT` as a language-agnostic equivalent of `size_t` from `<stddef.h>` 58- Add `PetscCount` as a signed datatype for counts, equivalent to `ptrdiff_t` from `<stddef.h>`. 59- Add `PetscCountCast`, `PetscSortIntWithCountArray()`, and `PetscSortIntWithIntCountArrayPair()` 60- Deprecate `SETERRQ1()` - `SETERRQ9()` in favor of `SETERRQ()` which is now variadic 61- Deprecate `PetscInfo1()` - `PetscInfo9()` in favor of `PetscInfo()` which is now variadic 62- Deprecate `PETSC_INLINE`, `inline` is a standard keyword since C99 and C++11 63- Deprecate `PETSC_STATIC_INLINE`, as both `static` and `inline` are standard keywords since C99 and C++11 64- Remove `PETSC_C_RESTRICT`, `restrict` is a standard keyword since C99 65- Change `SETERRMPI()` to be variadic 66- Change `SETERRABORT()` to be variadic 67- Add `PetscCheck()` and `PetscAssert()` for checking a boolean condition is true. The former is always enabled, while the latter is enabled only in debug builds. 68- `PetscDevice` initialization for CUDA and HIP will now respect `CUDA_VISIBILE_DEVICES` and `HIP_VISIBLE_DEVICES` environment variables respectively 69- Add `PETSC_ATTRIBUTE_COLD` to inform compilers that a function is unlikely to be called 70- Add `PetscCall()`, `PetscCallVoid()`, `PetscCallMPI()`, `PetscCallAbort()`, `PetscCallContinue()`, `PetscCallThrow()`, and `PetscCallCXX()`. These supersede `CHKERRQ()`, `CHKERRV()`, `CHKERRMPI()`, `CHKERRABORT()`, `CHKERRCONTINUE()`, `CHKERRXX()`, and `CHKERRCXX()` respectively 71- Add `PetscCallCUDA()`, `PetscCallCUBLAS()`, `PetscCallCUSPARSE()`, `PetscCallCUSOLVER()`, `PetscCallCUFFT()`, and `PetscCallCURAND()`. These supersede `CHKERRCUDA()`, `CHKERRCUBLAS()`, `CHKERRCUSPARSE()`, `CHKERRCUSOLVER()`, `CHKERRCUFFT()`, and `CHKERRCURAND()` respectively 72- Add `PetscCallHIP()`, `PetscCallHIPBLAS()`, and `PetscCallHIPSOLVER()`. These supersede `CHKERRHIP()`, `CHKERRHIPBLAS()`, and `CHKERRHIPSOLVER()` respectively 73- Add `PetscCallCEED()` which supersedes `CHKERRQ_CEED()` 74- Soft-deprecate `CHKERR` variants listed above in C/C++ sources. New code should prefer the `PetscCall` variants though no compiler diagnostics will be emitted if the old versions are used. One may use `${PETSC_DIR}/share/petsc/chkerrconvert.py` to perform the conversions automatically, though users should note that this script is considered "beta" software. It was used to facilitate the `CHKERR` changes during development and is being exposed to users in order to smooth the transition -- while every reasonable effort was made to ensure correctness, we provide no guarantee of it. 75 76```{rubric} PetscViewer: 77``` 78 79- Add `PetscViewerHDF5SetDefaultTimestepping()` and `PetscViewerHDF5SetDefaultTimestepping()` to deal with HDF5 files missing the timestepping attribute 80 81```{rubric} PetscDraw: 82``` 83 84- Add `PetscDrawSPAddPointColorized()` to change scatter point color based on third input value 85 86```{rubric} AO: 87``` 88 89```{rubric} IS: 90``` 91 92- `ISLocalToGlobalMappingCreateSF()`: allow passing `start = PETSC_DECIDE` 93- Add `ISGeneralSetIndicesFromMask()` 94- Add `ISSetLayout()` 95- Add `PetscSectionSymDistribute()` 96- Add `ISLocalToGlobalMappingGetType()` 97 98```{rubric} VecScatter / PetscSF: 99``` 100 101- Add MPI-4.0 large count support. With an MPI-4.0 compliant MPI implementation and 64-bit indices, one can now pass over 2 billion elements in a single message in either VecScatter or PetscSF 102- Add `PetscSFFetchAndOpWithMemTypeBegin()`, which is similar to `PetscSFFetchAndOpBegin()`, but with explicit memory types 103- Change `PetscSFSetGraph()` and `PetscSFSetGraphLayout()` to sort leaves, remove `const` from `ilocal` and `iremote` arguments 104- Add `PetscSFConcatenate()` 105 106```{rubric} PF: 107``` 108 109```{rubric} Vec: 110``` 111 112- Change `VecTaggerComputeBoxes()` and `VecTaggerComputeIS()` to return a boolean whose value is true if the list was created 113- Add `-vec_bind_below` option for specifying size threshold below which GPU is not used for `Vec` operations 114- Add `VecSetBindingPropagates()` 115- Add `VecGetBindingPropagates()` 116- For CUDA and ViennaCL and HIP GPU vectors, `VecCreate()` no longer allocates the array on CPU eagerly, it is only allocated if it is needed 117- `VecGetArrayAndMemType()` and `VecGetArrayReadAndMemType()` now always return a device pointer (copying the data to the device if needed) for the standard CUDA, HIP, and CUDA/HIP Kokkos vectors. Previously, they did so only when the device had the latest data 118- Add `VecGetArrayWriteAndMemType()` and `VecRestoreArrayWriteAndMemType()`, which are similar to the `VecGetArrayReadAndMemType()` family, but only write to the vector on device 119 120```{rubric} PetscPartitioner: 121``` 122 123```{rubric} Mat: 124``` 125 126- Add `-mat_bind_below` option for specifying size threshold below which GPU is not used for `Mat` operations 127- Add `MatSetBindingPropagates()` 128- Add `MatGetBindingPropagates()` 129- Add `MatSeqAIJGetArrayWrite()` and `MatSeqAIJRestoreArrayWrite()` to get write-access to the value array of `MatSeqAIJ` on CPU 130- Add `MatCUSPARSESetUseCPUSolve()` Use CPU solve with cuSparse for LU factorization that are on the CPU. 131- Change `MatCreateIS()` behavior when NULL is passed for the mappings. Now a NULL map implies matching local and global spaces 132- Add support of `MatSetValuesCOO()` and `MatSetPreallocationCOO()` for matrix type AIJKOKKOS. Additionally, for AIJKOKKOS, they support negative indices and remote entries 133- Add `MatMultHermitianTransposeEqual()` and `MatMultHermitianTransposeAddEqual()` 134- Add `MatSetPreallocationCOOLocal()` to set preallocation for matrices using a coordinate format of the entries with local indices 135- Change `MatStructures` enumeration to avoid spaces and match capitalization of other enumerations 136- Change size argument of `MatSetPreallocationCOO()` to `PetscCount` 137- Add `MATORDERINGMETISND` use METIS for nested dissection ordering of `MatSeqAIJ`, with options `nseps`, `niter`, `ufactor` and `pfactor` under the common prefix `-mat_ordering_metisnd_` 138- Change options `-matproduct_<product_type>_via` to `-mat_product_algorithm` 139- Add `-mat_superlu_dist_3d` and `-mat_superlu_dist_d <n>` to support using SuperLU_DIST's version 7.2 3d decomposition algorithms 140- `MATIS` now supports negative and repeated indices in the local-to-global map and the COO assembly routines. This allows for better integration with libceed 141- Add `MatISGetLocalToGlobalMapping()` to retrieve the logical map for assembled subdomain problem 142 143```{rubric} PC: 144``` 145 146- Add MG option `-pc_mg_galerkin_mat_product_algorithm [cusparse|hypre]` and `PCMGGalerkinSetMatProductAlgorithm()` to use cuSparse or hypre's SpGEMM for Galerkin products in hypre 147- Add PC type `PCBJKOKKOS` a new, experimental batch Kokkos solver `-pc_type bjkokkos -pc_bjkokkos_ksp_type [tfqmr|bicg] -pc_bjkokkos_pc_type jacobi -ksp_type preonly` 148- Add -pc_svd_monitor ::all option to print all the singular values instead of a maximum of ten 149 150```{rubric} PCMG: 151``` 152 153- Add `PCMGGetGridComplexity()` to get operator and grid complexity of MG hierarchy 154- Change `PCGAMG` default to use `PCJACOBI` smoothing instead of `PCSOR`. This also allows the default configuration to use GPUs effectively, and to deliver equivalent convergence. For the old default, use `-mg_levels_pc_type sor`. 155- Change `PCGAMG` eigenvalue estimation to use `KSPCG` when `MAT_SPD` has been set (see `MatSetOption()`) and `KSPCR` when `MAT_SYMMETRIC` or `MAT_HERMITIAN` has been set. These are usually somewhat more accurate and reliable than the previous default of `KSPGMRES`, and in tune with `KSPCHEBYSHEV`. Note that Chebyshev will generally not be a suitable smoother for indefinite matrices. 156- Change `PCGAMG` to use `PCGAMGSetUseSAEstEig()` by default when the smoother uses Jacobi preconditioning. 157 158```{rubric} KSP: 159``` 160 161- Outer most `KSPSolve()` will error if `KSP_DIVERGED_ITS` and `KSPSetErrorIfNotConverged()` is used 162- Add `KSPQMRCGS` to support qmrcgstab with right preconditioning 163- Add `KSPGuessSetTolerance()` 164- Add a new model option to `KSPGuessFischerSetModel()` 165 166```{rubric} SNES: 167``` 168 169- Add `SNESNewtonTRDCGetRhoFlag()`, `SNESNewtonTRDCSetPreCheck()`, `SNESNewtonTRDCGetPreCheck()`, `SNESNewtonTRDCSetPostCheck()`, `SNESNewtonTRDCGetPostCheck()` 170 171```{rubric} SNESLineSearch: 172``` 173 174```{rubric} TS: 175``` 176 177- Add `TSSundialsSetUseDense()` and options database option `-ts_sundials_use_dense` to use a dense linear solver (serial only) within CVODE, instead of the default iterative solve 178- Change timestepper type `TSDISCGRAD` to include additional conservation terms based on formulation from [Gonzalez 1996] for Hamiltonian systems: 179 \- Add `TSDiscGradIsGonzalez()` to check flag for whether to use additional conservative terms in discrete gradient formulation 180 \- Add `TSDiscGradUseGonzalez()` to set discrete gradient formulation with or without additional conservative terms. Without flag, the discrete gradients timestepper is just backwards euler 181- Add `TSRemoveTrajectory` to destroy and remove the internal TSTrajectory object from TS 182- Change `TSMonitorSPSwarmSolution()` to have correct axes labels and bounds 183 184```{rubric} Tao: 185``` 186 187- Add `TaoGetGradient()`, `TaoGetObjectiveAndGradient()` and `TaoGetHessian()` 188- Deprecate `TaoSetInitialVector()` in favor of `TaoSetSolution()` 189- Deprecate `TaoGetSolutionVector()` in favor of `TaoGetSolution()` 190- Deprecate `TaoGetGradientVector()` in favor of `TaoGetGradient()` 191- Deprecate `TaoSetObjectiveRoutine()` in favor of `TaoSetObjective()` 192- Deprecate `TaoSetGradientRoutine()` in favor of `TaoSetGradient()` 193- Deprecate `TaoSetObjectiveAndGradientRoutine()` in favor of `TaoSetObjectiveAndGradient()` 194- Deprecate `TaoSetHessianRoutine()` in favor of `TaoSetHessian()` 195- Change `TaoGetObjective()`. Use `TaoGetSolutionStatus(tao,NULL,&fct,NULL,NULL,NULL,NULL)` instead 196 197```{rubric} DM/DA: 198``` 199 200- Add `DMLabelGetNonEmptyStratumValuesIS()`, similar to `DMLabelGetValueIS()` but counts only nonempty strata 201- Add `DMLabelCompare()` for `DMLabel` comparison 202- Add `DMCompareLabels()` comparing `DMLabel`s of two `DM`s 203- `DMCopyLabels()` now takes DMCopyLabelsMode argument determining duplicity handling 204- Add `-dm_bind_below` option for specifying size threshold below which GPU is not used for `Vec` and `Mat` objects associated with a DM 205- Add `DMCreateMassMatrixLumped()` to support explicit timestepping, also add `DMTSCreateRHSMassMatrix()`, `DMTSCreateRHSMassMatrixLumped()`, and `DMTSDestroyRHSMassMatrix()` 206- Promote `DMGetFirstLabelEntry()` to public API and rename 207- Add bias vector argument to `DMGetDefaultConstraints()` and `DMSetDefaultConstraints()`. Passing `bias=NULL` recovers prior behavior. 208- Change `DMGetAuxiliaryVec()`, `DMSetAuxiliaryVec()`, and `DMGetAuxiliaryLabels()` to take an equation part number 209 210```{rubric} DMSwarm: 211``` 212 213- Add `DMSwarmGetNumSpecies()` and `DMSwarmSetNumSpecies()` to support PIC 214- Add `DMSwarmComputeLocalSize()`, `DMSwarmComputeLocalSizeFromOptions()`, `DMSwarmInitializeCoordinates()`, `DMSwarmInitializeVelocities()`, `DMSwarmInitializeVelocitiesFromOptions()` to assist initialization of PIC methods 215 216```{rubric} DMPlex: 217``` 218 219- Add `DMExtrude()` which now the default extrusion 220- Change `DMPlexExtrude()` to use DMPlexTransform underneath 221- Add `DMPlexTransformExtrudeSetNormalFunction()` to allow computed normals 222- Add `DMGetNaturalSF()` and `DMSetNaturalSF()` 223- Change `-dm_plex_csr_via_mat` to `-dm_plex_csr_alg` which takes a DMPlexCSRAlgorithm name 224- Add public API for metric-based mesh adaptation: 225 : - Move `DMPlexMetricCtx` from public to private and give it to `DMPlex` 226 - Add `DMPlexMetricSetFromOptions()` to assign values to `DMPlexMetricCtx` 227 - Add `DMPlexMetricSetIsotropic()` for declaring whether a metric is isotropic 228 - Add `DMPlexMetricIsIsotropic()` for determining whether a metric is isotropic 229 - Add `DMPlexMetricSetUniform()` for declaring whether a metric is uniform 230 - Add `DMPlexMetricIsUniform()` for determining whether a metric is uniform 231 - Add `DMPlexMetricSetRestrictAnisotropyFirst()` for declaring whether anisotropy should be restricted before normalization 232 - Add `DMPlexMetricRestrictAnisotropyFirst()` for determining whether anisotropy should be restricted before normalization 233 - Add `DMPlexMetricSetMinimumMagnitude()` for specifying the minimum tolerated metric magnitude 234 - Add `DMPlexMetricGetMinimumMagnitude()` for retrieving the minimum tolerated metric magnitude 235 - Add `DMPlexMetricSetMaximumMagnitude()` for specifying the maximum tolerated metric magnitude 236 - Add `DMPlexMetricGetMaximumMagnitude()` for retrieving the maximum tolerated metric magnitude 237 - Add `DMPlexMetricSetMaximumAnisotropy()` for specifying the maximum tolerated metric anisostropy 238 - Add `DMPlexMetricGetMaximumAnisotropy()` for retrieving the maximum tolerated metric anisotropy 239 - Add `DMPlexMetricSetTargetComplexity()` for specifying the target metric complexity 240 - Add `DMPlexMetricGetTargetComplexity()` for retrieving the target metric complexity 241 - Add `DMPlexMetricSetNormalizationOrder()` for specifying the order of L-p normalization 242 - Add `DMPlexMetricGetNormalizationOrder()` for retrieving the order of L-p normalization 243 - Change `DMPlexMetricCtx` so that it is only instantiated when one of the above routines are called 244 - Change `DMPlexMetricEnforceSPD()` to have more arguments: 245 : - control whether anisotropy is restricted 246 - output the modified metric, rather than modifying the input 247 - output the determinant 248 - Change `DMPlexMetricNormalize()` to have another argument, for controlling whether anisotropy is restricted 249- Change `DMAdaptor` so that its `-adaptor_refinement_h_min/h_max/a_max/p` command line arguments become `-dm_plex_metric_h_min/h_max/a_max/p` 250- Add 2D and 3D mesh adaptation interface to Mmg and 3D mesh adaptation interface to ParMmg. Mmg/ParMmg specific changes: 251 : - Change `DMPlexBuildFromCellListParallel()` to have another argument, for the connectivity 252 - Change `DMPlexCreateFromCellListParallelPetsc()` to have another argument, for the connectivity 253 - Add `DMPlexMetricSetVerbosity()` for setting the verbosity of the metric-based mesh adaptation package 254 - Add `DMPlexMetricGetVerbosity()` for getting the verbosity of the metric-based mesh adaptation package 255 - Add `DMPlexMetricSetNoInsertion()` to turn off node insertion and deletion for (Par)Mmg 256 - Add `DMPlexMetricNoInsertion()` to determine whether node insertion and deletion are turned off for (Par)Mmg 257 - Add `DMPlexMetricSetNoSwapping()` to turn off facet swapping for (Par)Mmg 258 - Add `DMPlexMetricNoSwapping()` to determine whether facet swapping is turned off for (Par)Mmg 259 - Add `DMPlexMetricSetNoMovement()` to turn off node movement for (Par)Mmg 260 - Add `DMPlexMetricNoMovement()` to determine whether node movement is turned off for (Par)Mmg 261 - Add `DMPlexMetricSetNoSurf()` to turn off surface modification for (Par)Mmg 262 - Add `DMPlexMetricNoSurf()` to determine whether surface modification is turned off for (Par)Mmg 263 - Add `DMPlexMetricSetGradationFactor()` to set the metric gradation factor 264 - Add `DMPlexMetricGetGradationFactor()` to get the metric gradation factor 265 - Add `DMPlexMetricSetHausdorffNumber()` to set the metric Hausdorff number 266 - Add `DMPlexMetricGetHausdorffNumber()` to get the metric Hausdorff number 267 - Add `DMPlexMetricSetNumIterations()` to set the number of ParMmg adaptation iterations 268 - Add `DMPlexMetricGetNumIterations()` to get the number of ParMmg adaptation iterations 269- Change `DMPlexCoordinatesLoad()` to take a `PetscSF` as argument 270- Change `DMPlexLabelsLoad()` to take the `PetscSF` argument and load in parallel 271- Change `DMPlexCreateFromFile()` to take the mesh name as argument 272- Change `DMAdaptMetric` so that it takes an additional argument for cell tags 273- Change `DMTransformAdaptLabel` so that it takes an additional argument for cell tags 274- Change `DMGenerateRegister` so that it registers routines that take an additional argument for cell tags 275- Change `DMPlexFindVertices()` to take `Vec` and `IS` arguments instead of arrays 276- Add `DMPlexTSComputeRHSFunctionFEM()` to support explicit timestepping 277- Newly created `DMPlex` will be distributed by default; this previously required `-dm_distribute` or explicit calls to `DMPlexDistribute()` 278- Add `DMPlexDistributeGetDefault()` and `DMPlexDistributeSetDefault()` to determine and set the default for `DMPlex` distribution 279- Add meshing of the Schwarz-P and Gyroid triply periodic minimal surface (see `DMPlexCreateTPSMesh()`). These meshes can be automatically generated using `-dm_plex_shape schwarz_p` or `-dm_plex_shape gyroid`, with optional levels of refinement and extrusion to 3D solids with prescribed thickness. 280- Add `DMCreateFEDefault()` as a convenience method for creating the right element on a mesh 281- Add `DMPlexCreateReferenceCell()` 282- Remove deprecated `DMPlexCreateFromCellList()` and `DMPlexCreateFromCellListParallel()` 283- Add `DMSetMatrixPreallocateSkip()` to save initialization time when `MatSetPreallocationCOO()` will be used. 284- Improve loading performance related to coordinate projection in common cases. 285- Add `DMPlexGetOrdering1D()` for 1D Plex problems 286- Add `DMPlexComputeClementInterpolant()` averaging operator 287 288```{rubric} DMPlexLandau: 289``` 290 291- Add `DMPlexLandauCreateVelocitySpace()` Create DMComposite of DMPlex for Landau collision operator 292- Add `DMPlexLandauDestroyVelocitySpace()` Destroy DMComposite of DMPlex for Landau collision operator 293- Add `DMPlexLandauIFunction()` Landau collision operator 294- Add `DMPlexLandauIJacobian()` Landau collision operator 295 296```{rubric} FE/FV: 297``` 298 299- Deprecate `PetscSpacePolynomialGetSymmetric()` and `PetscSpacePolynomialSetSymmetric()`: symmetric polynomials were never supported and support is no longer planned 300- Remove `PetscSpacePolynomialType` enum and associated array of strings `PetscSpacePolynomialTypes`: other polynomial spaces are now handled by other implementations of `PetscSpace` 301- Add `PETSCSPACEPTRIMMED` that implements trimmed polynomial spaces (also known as the spaces in Nedelec face / edge elements of the first kind) 302- Replace `PetscDSGet/SetHybrid()` with `PetscDSGet/SetCohesive()` 303- Add `PetscDSIsCohesive()`, `PetscDSGetNumCohesive()`, and `PetscDSGetFieldOffsetCohesive()` 304- Add argument to `PetscFEIntegrateHybridJacobian()` to indicate the face for the integration 305- Add `PetscFECreateByCell()` and `PetscFECreateLagrangeByCell()` to create FE spaces on specific cell types 306- Replace `PetscDualSpaceCreateReferenceCell()` with `DMPlexCreateReferenceCell()` 307- Add `PetscDualSpaceEqual()` and `PetscQuadratureEqual()` 308 309```{rubric} DMNetwork: 310``` 311 312- `DMNetworkAddComponent()` now requires a valid component key for each call 313- Add `DMNetworkSharedVertexGetInfo()` 314 315```{rubric} DMStag: 316``` 317 318```{rubric} DT: 319``` 320 321- Add `PetscDTPTrimmedEvalJet()` to evaluate a stable basis for trimmed polynomials, and `PetscDTPTrimmedSize()` for the size of that space 322- Add `PetscDSGetRHSResidual()` and `PetscDSSetRHSResidual()` to support explicit timestepping 323- Add `PetscDTTensorQuadratureCreate()` to combine different quadratures, such as on a prism 324- Add `PetscProbComputeKSStatistic()` to apply the Kolmogorov-Smirnov test 325- Add probability distributions `PetscPDFMaxwellBoltzmann1D()`, `PetscCDFMaxwellBoltzmann1D()`, `PetscPDFMaxwellBoltzmann2D()`, `PetscCDFMaxwellBoltzmann2D()`, `PetscPDFMaxwellBoltzmann3D()`, `PetscCDFMaxwellBoltzmann3D()`, `PetscPDFGaussian1D()`, `PetscCDFGaussian1D()`, `PetscPDFSampleGaussian1D()`, `PetscPDFGaussian2D()`, `PetscPDFSampleGaussian2D()`, `PetscPDFConstant1D()`, `PetscCDFConstant1D()` 326 327```{rubric} Fortran: 328``` 329 330```{rubric} Logging: 331``` 332 333- Add `PetscLogIsActive()` to determine if logging is in progress or not 334