|
template<int nColor, typename sumType , typename real > |
__device__ __host__ void | quda::colorInnerProduct (complex< sumType > &dot, int i, complex< real > v[nColor], complex< real > w[nColor]) |
|
template<int nColor, typename sumType , typename real > |
__device__ __host__ void | quda::colorNorm (sumType &nrm, complex< real > v[nColor]) |
|
template<typename real , int nColor> |
__device__ __host__ void | quda::colorScaleSubtract (complex< real > v[nColor], complex< real > a, complex< real > w[nColor]) |
|
template<typename real , int nColor> |
__device__ __host__ void | quda::colorScale (complex< real > v[nColor], real a) |
|
template<typename sumFloat , typename Float , int nSpin, int spinBlockSize, int nColor, int coarseSpin, int nVec, typename Arg > |
void | quda::blockOrthoCPU (Arg &arg) |
|
template<int block_size, typename sumFloat , typename Float , int nSpin, int spinBlockSize, int nColor, int coarseSpin, int nVec, typename Arg > |
| quda::__launch_bounds__ (2 *block_size) __global__ void blockOrthoGPU(Arg arg) |
|