Anthony Chang 6280231f39 use inline asm for 4x4 int8 transposition (#187) ...

[ROCm/composable_kernel commit: 08a979f188]

2022-04-22 15:47:31 -05:00

..

reduction_functions_threadwise.hpp

Improve Reduction kernel api (#152)

2022-04-04 20:31:44 -05:00

threadwise_contraction_dlops.hpp

Reorganize files, Part 1 (#119)

2022-03-08 21:46:36 -06:00

threadwise_gemm_dlops_v3.hpp

Reorganize files, Part 1 (#119)

2022-03-08 21:46:36 -06:00

threadwise_tensor_slice_set.hpp

Reorganize files, Part 1 (#119)

2022-03-08 21:46:36 -06:00

threadwise_tensor_slice_transfer_v3r1.hpp

use inline asm for 4x4 int8 transposition (#187)

2022-04-22 15:47:31 -05:00

threadwise_tensor_slice_transfer_v3r3.hpp

Compile for gfx908 and gfx90a (#130)

2022-03-31 12:33:34 -05:00

threadwise_tensor_slice_transfer_v4r1.hpp

Reorganize files, Part 1 (#119)

2022-03-08 21:46:36 -06:00

threadwise_tensor_slice_transfer_v5r1.hpp

Compile for gfx908 and gfx90a (#130)

2022-03-31 12:33:34 -05:00

threadwise_tensor_slice_transfer_v6r1.hpp

Compile for gfx908 and gfx90a (#130)

2022-03-31 12:33:34 -05:00

threadwise_tensor_slice_transfer_v6r2.hpp

Compile for gfx908 and gfx90a (#130)

2022-03-31 12:33:34 -05:00

threadwise_tensor_slice_transfer_v6r3.hpp

Compile for gfx908 and gfx90a (#130)

2022-03-31 12:33:34 -05:00

threadwise_tensor_slice_transfer.hpp

Compile for gfx908 and gfx90a (#130)

2022-03-31 12:33:34 -05:00