|
Uni20
Asynchronous tensor and runtime library.
|
Checked column-major GEMM wrappers over a leased cuBLAS execution context. More...
#include <uni20/backend/backend.hpp>#include <uni20/backend/cublas/cublas_error.hpp>#include <uni20/backend/cublas/execution.hpp>#include <uni20/common/trace.hpp>#include <uni20/core/types.hpp>#include <cuComplex.h>#include <cublas_v2.h>#include <algorithm>#include <concepts>#include <type_traits>Go to the source code of this file.
Namespaces | |
| namespace | uni20 |
| namespace | uni20::cublas |
Concepts | |
| concept | uni20::cublas::CublasScalar |
| Scalar types supported by the cuBLAS GEMM wrappers. | |
Functions | |
| template<CublasScalar Scalar> | |
| void | uni20::cublas::gemm (ExecutionLease &execution, char lhs_transform, char rhs_transform, int rows, int cols, int inner, Scalar alpha, Scalar const *lhs, int lhs_leading_dimension, Scalar const *rhs, int rhs_leading_dimension, Scalar beta, Scalar *output, int output_leading_dimension) |
| Enqueue column-major GEMM through an exclusive handle/stream lease. | |
Checked column-major GEMM wrappers over a leased cuBLAS execution context.