composable_kernel

mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-14 18:17:44 +00:00

Files

Chao Liu 8791d26e52 FP16 data in-register transpose (#41 )

* start fixing 16bit data packing

* adding StaticTensor

* adding StaticTensor

* adding StaticTensor

* add missing constexpr

* adding static tensor

* adding static tensor

* adding transpose

* add inline asm for transpose 2x2 of half_t

* add general transpose_vectors(), but have unnecessary register initialization using v_mov

* fix unnecessary register initialization in transpose_vector by using more pass-by-reference

* add hardcoded logic for NHWC wrw

* improve asm for v_pack

* make ThreadwiseTensorSliceTransfer_v3r2 support any tensor

* tweak

* reorganize file

[ROCm/composable_kernel commit: b491ebf384]

2021-11-15 10:05:58 -06:00

include

FP16 data in-register transpose (#41 )

2021-11-15 10:05:58 -06:00

device_conv_xdl_instance_f16_f16_f16_nhwc_kyxc_nhwk.cpp

ckProfiler and device-level XDL GEMM operator (#48 )

2021-11-14 11:28:32 -06:00

device_conv_xdl_instance_f32_f32_f32_nhwc_kyxc_nhwk.cpp

ckProfiler and device-level XDL GEMM operator (#48 )

2021-11-14 11:28:32 -06:00

device_gemm_xdl_instance_f16_f16_f16_km_kn_mn.cpp

ckProfiler and device-level XDL GEMM operator (#48 )