CUTLASS
CUDA Templates for Linear Algebra Subroutines and Solvers
|
Namespaces | |
device | |
kernel | |
thread | |
threadblock | |
warp | |
Classes | |
struct | BatchedGemmCoord |
struct | GemmCoord |
struct | GemmShape |
Shape of a matrix multiply-add operation. More... | |
Typedefs | |
template<typename Shape > | |
using | GemmShapeTranspose = GemmShape< Shape::kN, Shape::kM, Shape::kK > |
Type alias of the transpose of a GemmShape. More... | |
Enumerations | |
enum | Operand { Operand::kA, Operand::kB, Operand::kC, Operand::kD } |
GEMM operand enumeration: D = A * B + C. More... | |
using cutlass::gemm::GemmShapeTranspose = typedef GemmShape<Shape::kN, Shape::kM, Shape::kK> |
|
strong |