Skip to content
Snippets Groups Projects
Select Git revision
  • 620be0d6523bf3f2430d6453bad84e89f69bad45
  • develop default protected
  • feature/gmsh-reader
  • feature/reconstruction
  • save_clemence
  • origin/stage/bouguettaia
  • feature/kinetic-schemes
  • feature/local-dt-fsi
  • feature/composite-scheme-sources
  • feature/composite-scheme-other-fluxes
  • feature/serraille
  • feature/variational-hydro
  • feature/composite-scheme
  • hyperplastic
  • feature/polynomials
  • feature/gks
  • feature/implicit-solver-o2
  • feature/coupling_module
  • feature/implicit-solver
  • feature/merge-local-dt-fsi
  • master protected
  • v0.5.0 protected
  • v0.4.1 protected
  • v0.4.0 protected
  • v0.3.0 protected
  • v0.2.0 protected
  • v0.1.0 protected
  • Kidder
  • v0.0.4 protected
  • v0.0.3 protected
  • v0.0.2 protected
  • v0 protected
  • v0.0.1 protected
33 results

CMakeLists.txt

  • Stéphane Del Pino's avatar
    620be0d6
    Remove ASTNode::m_value member · 620be0d6
    Stéphane Del Pino authored
    Values are no more stored into AST nodes. The reason for that is that it was not
    appropriate for multi-threaded evaluation of functions, since multiple
    evaluations could not occur at the same time.
    
    Now expression evaluation do not fill the node value but eventually return the
    computed value.
    
    Also, function are evaluated using a context array of values. Thus each function
    calls is done in a separated context (as it should have always been).
    
    This is a big change in the AST execution strategy. Nonetheless, it leads to a
    seemingly better code, and there will be no more undefined trailing values in
    the AST (for some node ASTNode::m_value would never be initialized/used).
    620be0d6
    History
    Remove ASTNode::m_value member
    Stéphane Del Pino authored
    Values are no more stored into AST nodes. The reason for that is that it was not
    appropriate for multi-threaded evaluation of functions, since multiple
    evaluations could not occur at the same time.
    
    Now expression evaluation do not fill the node value but eventually return the
    computed value.
    
    Also, function are evaluated using a context array of values. Thus each function
    calls is done in a separated context (as it should have always been).
    
    This is a big change in the AST execution strategy. Nonetheless, it leads to a
    seemingly better code, and there will be no more undefined trailing values in
    the AST (for some node ASTNode::m_value would never be initialized/used).