Commit 2bc2e9b8 authored by Tomáš Oberhuber's avatar Tomáš Oberhuber Committed by Jakub Klinkovský
Browse files

Fixes after rebase.

parent ca680ebd

Documentation/Examples/CMakeLists.txt

+4 −2
Original line number Diff line number Diff line
@@ -3,7 +3,9 @@ ADD_SUBDIRECTORY( Containers ) 
ADD_SUBDIRECTORY( Pointers )

ADD_SUBDIRECTORY( Matrices )



set( COMMON_EXAMPLES

set( COMMON_EXAMPLES )



set( CUDA_EXAMPLES

   FileExampleCuda

)
@@ -24,7 +26,7 @@ set( HOST_EXAMPLES 
   TimerExampleLogger )



if( BUILD_CUDA )

   foreach( target IN ITEMS ${COMMON_EXAMPLES} )

   foreach( target IN ITEMS ${COMMON_EXAMPLES} ${CUDA_EXAMPLES} )

      cuda_add_executable( ${target}-cuda ${target}.cu OPTIONS )

      add_custom_command( COMMAND ${target}-cuda > ${TNL_DOCUMENTATION_OUTPUT_SNIPPETS_PATH}/${target}.out OUTPUT ${target}.out )

      set( CUDA_OUTPUTS ${CUDA_OUTPUTS} ${target}.out )

Documentation/Tutorials/Pointers/CMakeLists.txt

+4 −4
Original line number Diff line number Diff line 
IF( BUILD_CUDA )

   CUDA_ADD_EXECUTABLE( UniquePointerExample UniquePointerExample.cu )

   ADD_CUSTOM_COMMAND( COMMAND UniquePointerExample > ${TNL_DOCUMENTATION_OUTPUT_SNIPPETS_PATH}/UniquePointerExample.out OUTPUT UniquePointerExample.out )

   CUDA_ADD_EXECUTABLE( UniquePointerExample_ UniquePointerExample.cu )

   ADD_CUSTOM_COMMAND( COMMAND UniquePointerExample_ > ${TNL_DOCUMENTATION_OUTPUT_SNIPPETS_PATH}/UniquePointerExample.out OUTPUT UniquePointerExample.out )

   CUDA_ADD_EXECUTABLE( SharedPointerExample SharedPointerExample.cu )

   ADD_CUSTOM_COMMAND( COMMAND SharedPointerExample > ${TNL_DOCUMENTATION_OUTPUT_SNIPPETS_PATH}/SharedPointerExample.out OUTPUT SharedPointerExample.out )

   CUDA_ADD_EXECUTABLE( DevicePointerExample DevicePointerExample.cu )

   ADD_CUSTOM_COMMAND( COMMAND DevicePointerExample > ${TNL_DOCUMENTATION_OUTPUT_SNIPPETS_PATH}/DevicePointerExample.out OUTPUT DevicePointerExample.out )

ELSE()

   ADD_EXECUTABLE( UniquePointerExample UniquePointerExample.cpp )

   ADD_CUSTOM_COMMAND( COMMAND UniquePointerExample > ${TNL_DOCUMENTATION_OUTPUT_SNIPPETS_PATH}/UniquePointerExample.out OUTPUT UniquePointerExample.out )

   ADD_EXECUTABLE( UniquePointerExample_ UniquePointerExample.cpp )

   ADD_CUSTOM_COMMAND( COMMAND UniquePointerExample_ > ${TNL_DOCUMENTATION_OUTPUT_SNIPPETS_PATH}/UniquePointerExample.out OUTPUT UniquePointerExample.out )

ENDIF()



ADD_EXECUTABLE( UniquePointerHostExample UniquePointerHostExample.cpp )

src/Python/pytnl/tnl/SparseMatrix.h

+1 −1
Original line number Diff line number Diff line
@@ -67,7 +67,7 @@ struct export_CSR< Segments, typename TNL::enable_if_type< decltype(Segments{}.g 
   static void e( Scope & s )

   {

      s

         .def("getOffsets", []( const Segments& segments ) -> const typename Segments::OffsetsHolder& {

         .def("getOffsets", []( const Segments& segments ) -> const typename Segments::OffsetsContainer& {

                  return segments.getOffsets();

            }, py::return_value_policy::reference_internal)

      ;

src/TNL/Algorithms/Segments/BiEllpackView.h

+2 −2
Original line number Diff line number Diff line
@@ -15,7 +15,7 @@ 
#include <TNL/Containers/Vector.h>

#include <TNL/Algorithms/Segments/ElementsOrganization.h>

#include <TNL/Algorithms/Segments/BiEllpackSegmentView.h>

#include <TNL/Algorithms/Segments/details/BiEllpack.h>

#include <TNL/Algorithms/Segments/detail/BiEllpack.h>

#include <TNL/Algorithms/Segments/SegmentsPrinting.h>



namespace TNL {
@@ -205,7 +205,7 @@ class BiEllpackView 
                                             Real_ zero );



      template< typename Index_, typename Fetch_, int BlockDim_, int WarpSize_, bool B_ >

      friend struct details::BiEllpackreduceSegmentsDispatcher;

      friend struct detail::BiEllpackreduceSegmentsDispatcher;

#endif

};

src/TNL/Algorithms/Segments/BiEllpackView.hpp

+1 −1
Original line number Diff line number Diff line
@@ -425,7 +425,7 @@ reduceSegments( IndexType first, IndexType last, Fetch& fetch, const Reduction& 
         dim3 cudaGridSize = Cuda::getMaxGridSize();

         if( gridIdx == cudaGrids - 1 )

            cudaGridSize.x = cudaBlocks % Cuda::getMaxGridSize();

         details::BiEllpackreduceSegmentsKernel< ViewType, IndexType, Fetch, Reduction, ResultKeeper, Real, BlockDim  >

         detail::BiEllpackreduceSegmentsKernel< ViewType, IndexType, Fetch, Reduction, ResultKeeper, Real, BlockDim  >

            <<< cudaGridSize, cudaBlockSize, sharedMemory >>>

            ( *this, gridIdx, first, last, fetch, reduction, keeper, zero );

         cudaThreadSynchronize();