Loading src/Benchmarks/DistSpMV/tnl-benchmark-distributed-spmv.h +1 −1 Original line number Diff line number Diff line Loading @@ -239,7 +239,7 @@ struct SpmvBenchmark DistributedRowLengths distributedRowLengths( localRange, matrix.getRows(), group ); for( IndexType i = 0; i < distributedMatrix.getLocalMatrix().getRows(); i++ ) { const auto gi = distributedMatrix.getLocalRowRange().getGlobalIndex( i ); distributedRowLengths[ gi ] = matrix.getRowLength( gi ); distributedRowLengths[ gi ] = matrix.getRowCapacity( gi ); } distributedMatrix.setCompressedRowLengths( distributedRowLengths ); Loading src/Benchmarks/LinearSolvers/tnl-benchmark-linear-solvers.h +1 −1 Original line number Diff line number Diff line Loading @@ -461,7 +461,7 @@ struct LinearSolversBenchmark DistributedRowLengths distributedRowLengths( localRange, matrixPointer->getRows(), group ); for( IndexType i = 0; i < distMatrixPointer->getLocalMatrix().getRows(); i++ ) { const auto gi = distMatrixPointer->getLocalRowRange().getGlobalIndex( i ); distributedRowLengths[ gi ] = matrixPointer->getRowLength( gi ); distributedRowLengths[ gi ] = matrixPointer->getRowCapacity( gi ); } distMatrixPointer->setCompressedRowLengths( distributedRowLengths ); Loading src/TNL/Algorithms/MemoryOperations.h +2 −0 Original line number Diff line number Diff line Loading @@ -132,10 +132,12 @@ template<> struct MemoryOperations< Devices::Cuda > { template< typename Element > __cuda_callable__ static void setElement( Element* data, const Element& value ); template< typename Element > __cuda_callable__ static Element getElement( const Element* data ); template< typename Element, typename Index > Loading src/TNL/Algorithms/MemoryOperationsCuda.hpp +2 −2 Original line number Diff line number Diff line Loading @@ -24,7 +24,7 @@ namespace TNL { namespace Algorithms { template< typename Element > void __cuda_callable__ void MemoryOperations< Devices::Cuda >:: setElement( Element* data, const Element& value ) Loading @@ -48,7 +48,7 @@ setElement( Element* data, } template< typename Element > Element __cuda_callable__ Element MemoryOperations< Devices::Cuda >:: getElement( const Element* data ) { Loading src/TNL/Matrices/DistributedMatrix.h +2 −1 Original line number Diff line number Diff line Loading @@ -104,7 +104,8 @@ public: void setCompressedRowLengths( const CompressedRowLengthsVector& rowLengths ); void getCompressedRowLengths( CompressedRowLengthsVector& rowLengths ) const; template< typename Vector > void getCompressedRowLengths( Vector& rowLengths ) const; IndexType getRowLength( IndexType row ) const; Loading Loading
src/Benchmarks/DistSpMV/tnl-benchmark-distributed-spmv.h +1 −1 Original line number Diff line number Diff line Loading @@ -239,7 +239,7 @@ struct SpmvBenchmark DistributedRowLengths distributedRowLengths( localRange, matrix.getRows(), group ); for( IndexType i = 0; i < distributedMatrix.getLocalMatrix().getRows(); i++ ) { const auto gi = distributedMatrix.getLocalRowRange().getGlobalIndex( i ); distributedRowLengths[ gi ] = matrix.getRowLength( gi ); distributedRowLengths[ gi ] = matrix.getRowCapacity( gi ); } distributedMatrix.setCompressedRowLengths( distributedRowLengths ); Loading
src/Benchmarks/LinearSolvers/tnl-benchmark-linear-solvers.h +1 −1 Original line number Diff line number Diff line Loading @@ -461,7 +461,7 @@ struct LinearSolversBenchmark DistributedRowLengths distributedRowLengths( localRange, matrixPointer->getRows(), group ); for( IndexType i = 0; i < distMatrixPointer->getLocalMatrix().getRows(); i++ ) { const auto gi = distMatrixPointer->getLocalRowRange().getGlobalIndex( i ); distributedRowLengths[ gi ] = matrixPointer->getRowLength( gi ); distributedRowLengths[ gi ] = matrixPointer->getRowCapacity( gi ); } distMatrixPointer->setCompressedRowLengths( distributedRowLengths ); Loading
src/TNL/Algorithms/MemoryOperations.h +2 −0 Original line number Diff line number Diff line Loading @@ -132,10 +132,12 @@ template<> struct MemoryOperations< Devices::Cuda > { template< typename Element > __cuda_callable__ static void setElement( Element* data, const Element& value ); template< typename Element > __cuda_callable__ static Element getElement( const Element* data ); template< typename Element, typename Index > Loading
src/TNL/Algorithms/MemoryOperationsCuda.hpp +2 −2 Original line number Diff line number Diff line Loading @@ -24,7 +24,7 @@ namespace TNL { namespace Algorithms { template< typename Element > void __cuda_callable__ void MemoryOperations< Devices::Cuda >:: setElement( Element* data, const Element& value ) Loading @@ -48,7 +48,7 @@ setElement( Element* data, } template< typename Element > Element __cuda_callable__ Element MemoryOperations< Devices::Cuda >:: getElement( const Element* data ) { Loading
src/TNL/Matrices/DistributedMatrix.h +2 −1 Original line number Diff line number Diff line Loading @@ -104,7 +104,8 @@ public: void setCompressedRowLengths( const CompressedRowLengthsVector& rowLengths ); void getCompressedRowLengths( CompressedRowLengthsVector& rowLengths ) const; template< typename Vector > void getCompressedRowLengths( Vector& rowLengths ) const; IndexType getRowLength( IndexType row ) const; Loading