|
Uni20
Asynchronous tensor and runtime library.
|
Directories | |
| blas | |
| cpu | |
| cublas | |
| cuda | |
| cusolver | |
| direct_gemm | |
| lapack | |
| looped_gemm | |
Files | |
| binary80_common.hpp | |
| Binary80 scalar eligibility and logical host packing for BLAS/LAPACK adapters. | |
| least_squares_common.hpp | |
| linear_solve_common.hpp | |
| lu_common.hpp | |
| mpfr_common.hpp | |
| packed_matrix.hpp | |
| Logical host packing for value-converting BLAS and LAPACK adapters. | |
| reduction_output.hpp | |
| Shared host-output handling for synchronous scalar reductions. | |