Commit 797e663a authored by Tomáš Oberhuber's avatar Tomáš Oberhuber
Browse files

Merge branch 'cuda-tuning' of geraldine.fjfi.cvut.cz:/local/projects/tnl/tnl into cuda-tuning

parents 6667520a ddaaa19b

CMakeLists.txt

+5 −1
Original line number Diff line number Diff line
@@ -71,7 +71,7 @@ if( WITH_CUDA STREQUAL "yes" ) 
        # disable false compiler warnings

        #   reference for the -Xcudafe flag: http://stackoverflow.com/questions/14831051/how-to-disable-compiler-warnings-with-nvcc/17095910#17095910

        #   list of possible tokens: http://www.ssl.berkeley.edu/~jimm/grizzly_docs/SSL/opt/intel/cc/9.0/lib/locale/en_US/mcpcom.msg

        set(CUDA_NVCC_FLAGS ${CUDA_NVCC_FLAGS} ; --expt-relaxed-constexpr -Xcudafe "\"--diag_suppress=code_is_unreachable --diag_suppress=implicit_return_from_non_void_function\"")

        set(CUDA_NVCC_FLAGS ${CUDA_NVCC_FLAGS} ; -Wno-deprecated-gpu-targets --expt-relaxed-constexpr -Xcudafe "\"--diag_suppress=code_is_unreachable --diag_suppress=implicit_return_from_non_void_function\"")

        #AddCompilerFlag( "-DHAVE_NOT_CXX11" ) # -U_GLIBCXX_ATOMIC_BUILTINS -U_GLIBCXX_USE_INT128 " )

        set( ALL_CUDA_ARCHS -gencode arch=compute_20,code=sm_20

                            -gencode arch=compute_30,code=sm_30
@@ -322,6 +322,10 @@ find_package( PythonInterp 3 ) 
#   endif()

#endif()



if( OPTIMIZED_VECTOR_HOST_OPERATIONS STREQUAL "yes" )

   AddCompilerFlag( "-DOPTIMIZED_VECTOR_HOST_OPERATIONS " )

endif()



set( CXX_TEST_FLAGS "-fprofile-arcs -ftest-coverage" )

set( LD_TEST_FLAGS "-lgcov -coverage" )

build

+4 −1
Original line number Diff line number Diff line
@@ -13,6 +13,7 @@ INSTANTIATE_INT="yes" 
INSTANTIATE_LONG_DOUBLE="no"

INSTANTIATE_DOUBLE="yes"

INSTANTIATE_FLOAT="no"

OPTIMIZED_VECTOR_HOST_OPERATIONS="no"

CMAKE="cmake"

CMAKE_ONLY="no"

HELP="no"
@@ -42,6 +43,7 @@ do 
                                           INSTANTIATE_DOUBLE="yes"

                                           INSTANTIATE_FLOAT="no"

                                           WITH_CUDA_ARCH="auto" ;;

        --optimize-vector-host-operations=* ) OPTIMIZED_VECTOR_HOST_OPERATIONS="yes" ;;

        --with-cmake=*                   ) CMAKE="${option#*=}" ;;

        --build-jobs=*                   ) BUILD_JOBS="${option#*=}" ;;

        --cmake-only=*                   ) CMAKE_ONLY="${option#*=}" ;;
@@ -97,7 +99,8 @@ ${CMAKE} ${ROOT_DIR} \ 
         -DINSTANTIATE_DOUBLE=${INSTANTIATE_DOUBLE} \

         -DINSTANTIATE_LONG_DOUBLE=${INSTANTIATE_LONG_DOUBLE} \

         -DINSTANTIATE_INT=${INSTANTIATE_INT} \

         -DINSTANTIATE_LONG_INT=${INSTANTIATE_LONG_INT}

         -DINSTANTIATE_LONG_INT=${INSTANTIATE_LONG_INT} \

         -DOPTIMIZED_VECTOR_HOST_OPERATIONS=${OPTIMIZED_VECTOR_HOST_OPERATIONS}



if test $? != 0; then

    echo "Error: cmake exited with error code."

examples/advection/advectionRhs.h

+2 −2
Original line number Diff line number Diff line
@@ -26,7 +26,7 @@ template< typename Mesh, typename Real >class advectionRhs 
         typedef typename MeshEntity::MeshType::VertexType VertexType;

         VertexType v = entity.getCenter();

         return 0.0;

      };

      }

};



} // namespace TNL

examples/heat-equation/tnl-heat-equation-eoc.h

+1 −1
Original line number Diff line number Diff line
@@ -25,7 +25,7 @@ using namespace TNL; 
using namespace TNL::Problems;



//typedef tnlDefaultBuildMeshConfig BuildConfig;

typedef Solvers::tnlFastBuildConfig BuildConfig;

typedef Solvers::FastBuildConfig BuildConfig;



template< typename MeshConfig >

class heatEquationEocConfig

examples/heat-equation/tnl-heat-equation.h

+1 −1
Original line number Diff line number Diff line
@@ -26,7 +26,7 @@ using namespace TNL; 
using namespace TNL::Problems;



//typedef tnlDefaultBuildMeshConfig BuildConfig;

typedef Solvers::tnlFastBuildConfig BuildConfig;

typedef Solvers::FastBuildConfig BuildConfig;



template< typename MeshConfig >

class heatEquationConfig