zjing14 b6eaf3eb7e Pass gemm_descs for grouped gemm via __constant__ buff (#232) ...

* moved gemm_descs_args into const buff

* use CK_CONSTANT_ADDRESS_SPACE instead of global constant

* clean

* moved hipMemAlloc outside of deviceOp

* add SetWorkSpacePointer

* fix ignore

2022-05-31 17:00:43 -05:00

..

batched_gemm

Common forward convolution utility refactor. (#141)

2022-04-05 15:16:59 -05:00

batched_gemm_reduce

Add host API (#220)

2022-05-12 09:21:01 -05:00

block_to_ctile_map

Tensile-style block to C tile map (#239)

2022-05-24 21:55:22 -05:00

client_app

Add host API (#220)

2022-05-12 09:21:01 -05:00

conv2d_bwd_data

Unified implementation of 1d/2d/3d conv bwd-data. fp32/fp16/bfp16/int8 (#134)

2022-03-29 10:52:25 -05:00

conv2d_bwd_weight

Add host API (#220)

2022-05-12 09:21:01 -05:00

conv_util

Post PR183 review fixes. (#224)

2022-05-10 15:41:29 -05:00

convnd_bwd_data

Add host API (#220)

2022-05-12 09:21:01 -05:00

convnd_fwd

Post PR183 review fixes. (#224)

2022-05-10 15:41:29 -05:00

gemm

Add FP64 XDL GEMM built-in function (#199)

2022-05-26 14:48:57 -05:00

gemm_reduce

Add host API (#220)

2022-05-12 09:21:01 -05:00

gemm_split_k

Add host API (#220)

2022-05-12 09:21:01 -05:00

grouped_gemm

Pass gemm_descs for grouped gemm via __constant__ buff (#232)

2022-05-31 17:00:43 -05:00

magic_number_division

Common forward convolution utility refactor. (#141)

2022-04-05 15:16:59 -05:00

reduce

Overhaul to Reducton and its dependants (#237)

2022-05-24 12:19:12 -05:00

reference_conv_fwd

Post PR183 review fixes. (#224)

2022-05-10 15:41:29 -05:00

space_filling_curve

ctest of batched_gemm returns 0 or 1 (#149)

2022-03-24 19:38:02 -05:00

CMakeLists.txt

Navi21 gemm (#197)

2022-05-24 12:19:27 -05:00