dumux merge requestshttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests2020-04-18T14:09:30Zhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1992[fluidmatrixinteraction] Use std::clamp for improved readability2020-04-18T14:09:30ZTimo Kochtimokoch@math.uio.no[fluidmatrixinteraction] Use std::clamp for improved readability3.3Dennis GläserDennis Gläserhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1991Feature/2d segment segment intersections2020-05-08T13:36:08ZDennis GläserFeature/2d segment segment intersectionsThis MR adds the intersection algorithm for segment-like intersections of two segments in 2d space.This MR adds the intersection algorithm for segment-like intersections of two segments in 2d space.3.3Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1982[linear][istlsolverfactorybackend] Make usable with dune 2.72020-04-20T09:00:53ZKilian Weishaupt[linear][istlsolverfactorybackend] Make usable with dune 2.7* throw exception when used in parallel because this feature
is only available for >= 2.7.1
should be backported to 3.2* throw exception when used in parallel because this feature
is only available for >= 2.7.1
should be backported to 3.23.3Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1981Feature/deduce boundarytypes2020-06-18T18:44:11ZTimo Kochtimokoch@math.uio.noFeature/deduce boundarytypes* [x] Remove deprecation warnings* [x] Remove deprecation warnings3.3Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1977Feature/cleanup fvproblem dependencies2020-04-14T16:21:11ZTimo Kochtimokoch@math.uio.noFeature/cleanup fvproblem dependencies3.3Dennis GläserDennis Gläserhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1967Feature/data input2020-04-14T07:18:25ZTimo Kochtimokoch@math.uio.noFeature/data input3.3Dennis GläserDennis Gläserhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1936Support newest version of opm-grid for cornerpoint grids2020-04-02T09:07:45ZMartin SchneiderSupport newest version of opm-grid for cornerpoint grids3.2Dennis GläserDennis Gläserhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1928[efftoabs] Add sneToSn for symmetry reasons2020-03-31T09:07:15ZTimo Kochtimokoch@math.uio.no[efftoabs] Add sneToSn for symmetry reasonsTested in !1874 Tested in !1874 3.2Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1927[brookscorey] add regularization for rel. permeability derivatives2020-03-31T07:54:36ZTimo Kochtimokoch@math.uio.no[brookscorey] add regularization for rel. permeability derivatives3.2Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1926[gitignore] Add Python's *.pyc and __pycache__2020-03-31T07:42:37ZTimo Kochtimokoch@math.uio.no[gitignore] Add Python's *.pyc and __pycache__3.2Dennis GläserDennis Gläserhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1921Feature/uzawa 3.22020-05-04T13:04:57ZKilian WeishauptFeature/uzawa 3.2This MR supersedes !1827
- [x] So far, this only works reliably for 1p problems (without energy or mass transport).
Fixing this could be arbitrarily complex and time consuming. Should we assure for 3.2 that only 1p is used or at lea...This MR supersedes !1827
- [x] So far, this only works reliably for 1p problems (without energy or mass transport).
Fixing this could be arbitrarily complex and time consuming. Should we assure for 3.2 that only 1p is used or at least throw
a warning otherwise?3.2Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1904Feature/istl paramtree2020-03-26T20:02:06ZKilian WeishauptFeature/istl paramtree<!--
Thanks for sending a merge request!
If this is your first time, read our [contributing guidelines](/CONTRIBUTING.md)
-->
**What this MR does / why does DuMux need it**:
This MR adds a class for creating a parameter tree nee...<!--
Thanks for sending a merge request!
If this is your first time, read our [contributing guidelines](/CONTRIBUTING.md)
-->
**What this MR does / why does DuMux need it**:
This MR adds a class for creating a parameter tree needed for the dune istl solvers and preconditioners.
At can be used by the `IstlSolverFactoryBackend` and by a `SeqSolverBackend`.
The corresponding code in `IstlSolverFactoryBackend` is removed.
<!--
**Which issue this MR fixes** *(optional - uncomment and add issue)*:
fixes #
-->
**Special notes for your reviewer**:
@timok @bernd
Do we want `LinearSolverTraits` as template parameter?
So far it is only used to extract `dim` for AMG.
The traits are currently not passed to the `SeqSolverBackend`s.
I did not notice any difference between setting setDefaultValuesIsotropic(dim)
and not setting anything.
`BlockDiagAMGBiCGSTABSolver` has `setDefaultValuesIsotropic(3)`hard-coded.3.2Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1893[typetraits][matrix] Add trait for MultiTypeBlockMatrix2020-03-17T18:58:58ZKilian Weishaupt[typetraits][matrix] Add trait for MultiTypeBlockMatrix* use forward declare to remove includes* use forward declare to remove includes3.2Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1891Feature/nonisothermal facet coupling2020-04-07T12:18:46ZDennis GläserFeature/nonisothermal facet couplingWIP because it depends on !1882 and !1684
Fixes #592 WIP because it depends on !1882 and !1684
Fixes #592 3.2Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1882Feature/compositional facet coupling2020-04-07T12:17:10ZDennis GläserFeature/compositional facet couplingThis MR introduces support for compositional models by the facet coupling module.
depends !1684
Together with !1891 it adresses #592This MR introduces support for compositional models by the facet coupling module.
depends !1684
Together with !1891 it adresses #5923.2Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1875Feature/parallel co2 tests2020-03-17T21:13:31ZSimon EmmertFeature/parallel co2 tests<!--
Thanks for sending a merge request!
If this is your first time, read our [contributing guidelines](/CONTRIBUTING.md)
-->
**What this MR does / why does DuMux need it**:
Adds communication for boxVolume output and adds four ne...<!--
Thanks for sending a merge request!
If this is your first time, read our [contributing guidelines](/CONTRIBUTING.md)
-->
**What this MR does / why does DuMux need it**:
Adds communication for boxVolume output and adds four new co2 tests with and unstructured grid with saturation changes over process boundaries.
Additionally increase the injection rate and harmonizes the output by lowering the initial timestepsize and introducing a max TimeStepSize for the tests.
**Which issue this MR fixes**:
fixes #819
**Special notes for your reviewer**: Check on as many machines as possible if the output actually is the same. If i remember correctly the co2-tests have been prone to fail with different compiler versions before.3.2Kilian WeishauptKilian Weishaupthttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1867Feature/linear solver parallel firstcall2020-03-06T16:26:12ZTimo Kochtimokoch@math.uio.noFeature/linear solver parallel firstcallMake it possible to reset firstCall which is needed to compute ghost and owners. Should be possible to reset that if the grid changes.Make it possible to reset firstCall which is needed to compute ghost and owners. Should be possible to reset that if the grid changes.3.2Kilian WeishauptKilian Weishaupthttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1865[newton] Count number of failed linear solves for statistic2020-02-14T08:39:29ZTimo Kochtimokoch@math.uio.no[newton] Count number of failed linear solves for statistic3.2Dennis GläserDennis Gläserhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1855[test][material][components] added new components needed for the biomin example2020-03-24T12:41:57ZJohannes Hommel[test][material][components] added new components needed for the biomin exampleReplaces !1854.Replaces !1854.3.2Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/1850Restructure linearsolver parallel2020-02-14T12:55:02ZTimo Kochtimokoch@math.uio.noRestructure linearsolver parallel* Deduce vector and matrix type from solve
* Actually use sequential solver if we compute on one process (before a "parallel" solver was used)
* Fix "sequential" constructor of the AMG backend
* Rename to AMGBiCGSTABBackend
* LinearS...* Deduce vector and matrix type from solve
* Actually use sequential solver if we compute on one process (before a "parallel" solver was used)
* Fix "sequential" constructor of the AMG backend
* Rename to AMGBiCGSTABBackend
* LinearSolverTraits only need GridGeometry now
* enable box with overlap (big thanks to @blattms for implementing the draft for this (see !1861) and giving the necessary hint on how to do this)
* restructure the LinearSolverTraits
I consider the internal structure of the AMG non-public so we don't have to deprecate things. Opinions?
This is suggestion to fix #805.3.2Kilian WeishauptKilian Weishaupt