Skip to content
Snippets Groups Projects
  1. Nov 03, 2015
    • Timo Koch's avatar
      [fix][amg] Include dune version header · 058e2921
      Timo Koch authored
      058e2921
    • Martin Beck's avatar
      Merge branch 'feature/FS300-fix-AMGBackend' into 'master' · c1714f30
      Martin Beck authored
      [AMGBackend] fix AMGBackend for sequential grids and HAVE_MPI == 1
      
      So far, the AMGBackend only worked for sequential grids if HAVE_MPI
      was 0. This was due to the fact that AmgTraits distinguished between
      parallel and sequential by evaluating HAVE_MPI. If HAVE_MPI was 1
      but the Grid was sequential, this lead to compiler errors for the
      constructor calls of the Communicator, LinearOperator and
      ScalarProduct.
      
      Do this properly now by using Dune::Capabilities for distinction
      between the sequential and parallel case. For Dune 2.4, the
      capability isParallel has to be used, while for Dune 3.0, it is
      canCommunicate. The correct AmgTraits and constructor calls are
      chosen by template specialization.
      
      This solves FS#300.
      
      See merge request !16
      c1714f30
    • Bernd Flemisch's avatar
      [AMGBackend] fix AMGBackend for sequential grids and HAVE_MPI == 1 · 21c2b3d8
      Bernd Flemisch authored
      So far, the AMGBackend only worked for sequential grids if HAVE_MPI
      was 0. This was due to the fact that AmgTraits distinguished between
      parallel and sequential by evaluating HAVE_MPI. If HAVE_MPI was 1
      but the Grid was sequential, this lead to compiler errors for the
      constructor calls of the Communicator, LinearOperator and
      ScalarProduct.
      
      Do this properly now by using Dune::Capabilities for distinction
      between the sequential and parallel case. For Dune 2.4, the
      capability isParallel has to be used, while for Dune 3.0, it is
      canCommunicate. The correct AmgTraits and constructor calls are
      chosen by template specialization.
      
      This solves FS#300.
      21c2b3d8
    • Gabi Seitz's avatar
      [fluidstate] add fluidstate property · b9e5d439
      Gabi Seitz authored
      Add fluidstate property for all implicit porous media models
      if not yet defined
      
      Conflicts:
      	dumux/implicit/2pncmin/2pncminvolumevariables.hh
      b9e5d439
    • Bernd Flemisch's avatar
      Merge branch 'feature/whitespace-cleanup-and-hook' into 'master' · 1288fcfa
      Bernd Flemisch authored
      Feature/whitespace cleanup and hook
      
      Bitte reviewen, asap, bitte.
      
      See merge request !15
      1288fcfa
    • Christoph Grüninger's avatar
      Enable whitespace-hook. · 402039ac
      Christoph Grüninger authored
      Now all tabs and trailing whitespaces are rejected in commits.
      See FS#272 for more details.
      402039ac
    • Christoph Grüninger's avatar
      75888c39
    • Christoph Grüninger's avatar
      f94a3e16
    • Timo Koch's avatar
      [parameters] Don't check consistent type for parameter macro · 04894362
      Timo Koch authored
      The check was turned on for debug purposes and exited with an error
      if the same parameter was read into object of different type. This
      is not an error. For reading vectors it is even an important feature
      that it's possible to read it into different types of containers
      depending on the usage.
      
      Reviewed by martins
      04894362
  2. Oct 29, 2015
    • Thomas Fetzer's avatar
      Merge branch 'feature/crossproduct-update' into 'master' · 8cb87079
      Thomas Fetzer authored
      [math] (Re-)implement cross product in 2d/3d
      
      This makes the old cross-product more efficient. The old version
      initialized a FieldVector with zeros and then filled it. The
      new version builds it's directly using consise bracket notation.
      The result can be moved without creating a copy.
      
      Implement a cross product in 2d that returns a scalar.
      Not mathematically clean but still useful.
      
      2d and 3d cross products now explicitly only work in 2d and 3d.
      Other dimensions will trigger a compiler error.
      
      See merge request !14
      8cb87079
    • Bernd Flemisch's avatar
      Merge branch 'feature/externalscript' into 'master' · df0eee34
      Bernd Flemisch authored
      Feature/externalscript
      
      Unfortunately this contains some other commits which should have gone directly to master. I apologize for this mistake.
      
      See merge request !12
      df0eee34
    • Timo Koch's avatar
      [math] (Re-)implement cross product in 2d/3d · 979452c1
      Timo Koch authored
      This makes the old cross-product more efficient. The old version
      initialized a FieldVector with zeros and then filled it. The
      new version builds it's directly using consise bracket notation.
      The result can be moved without creating a copy.
      
      Implement a cross product in 2d that returns a scalar.
      Not mathematically clean but still useful.
      
      2d and 3d cross products now explicitly only work in 2d and 3d.
      Other dimensions will trigger a compiler error.
      979452c1
    • Thomas Fetzer's avatar
      Merge branch 'fix/gnuplotterminal' into 'master' · 6333c1bf
      Thomas Fetzer authored
      [gnuplot] Do not set terminal as default
      
      Gnuplot will pick a default terminal itself. If you
      change the number of plotting windows a terminal has to be
      specified. If not specified by the user the default is x11.
      
      See merge request !13
      6333c1bf
    • Thomas Fetzer's avatar
      4c21ca71
    • Timo Koch's avatar
      [gnuplot] Do not set terminal as default · 205f9cd1
      Timo Koch authored
      Gnuplot will pick a default terminal itself. If you
      change the number of plotting windows a terminal has to be
      specified. If not specified by the user the default is x11.
      205f9cd1
    • Thomas Fetzer's avatar
  3. Oct 28, 2015
  4. Oct 22, 2015
  5. Oct 21, 2015
  6. Oct 19, 2015
  7. Oct 16, 2015
Loading