Tuning CUDA traversers.

tests/benchmarks/heat-equation-benchmark/tnl-benchmark-simple-heat-equation-bug.h

@@ -68,9 +68,10 @@ class TestGridEntity
             
 };
 
-__global__ void testKernel( const tnlTestGrid* grid )
+template< typename GridType, typename GridEntity >
+__global__ void testKernel( const GridType* grid )
 {   
-   TestGridEntity entity( *grid );
+   GridEntity entity( *grid );
 }
 
 int main( int argc, char* argv[] )
@@ -81,7 +82,8 @@ int main( int argc, char* argv[] )
    dim3 cudaGridSize( gridXSize / 16 + ( gridXSize % 16 != 0 ),
                       gridYSize / 16 + ( gridYSize % 16 != 0 ) );
       
-   typedef tnlTestGrid GridType;
+   //typedef tnlTestGrid GridType;
+   typedef tnlGrid< 2, double, tnlCuda > GridType;
    typedef typename GridType::VertexType VertexType;
    typedef typename GridType::CoordinatesType CoordinatesType;      
    GridType grid;
@@ -93,7 +95,7 @@ int main( int argc, char* argv[] )
    auto t_start = std::chrono::high_resolution_clock::now();
    while( iteration < 1000 )
    {
-      testKernel<<< cudaGridSize, cudaBlockSize >>>( cudaGrid );
+      testKernel< GridType, typename GridType::Cell ><<< cudaGridSize, cudaBlockSize >>>( cudaGrid );
       cudaThreadSynchronize();
       iteration++;
    }