Jianfeng Yan 0619ebf70b Refactor threadwise copy using sfcurve (#101) ...

* add space_filling_curve

* cleanup and move space_filling_curve into test

* WIP: start refactoring threadwise_transfer_v1r3

* threadwise_copy works but needs further refactoring

* add some comments

* add SpaceFillingCurve::GetIndices()

* minor changes

* removed GetIndices; refactored GetDstCoordinateResetStep

* add DynamicBuffer::Transfer, but Add is not tested

* rebased agaist develop

* threadwise_copy_v6r1/v6r2/v6r3 using space-filling curve start to work

* minor changes

* refactored threadcopy v3r1, v2; removed old implementations

* clang-format

* cleanup

* fix a typo in v6r3

* format

Co-authored-by: Chao Liu <chao.liu2@amd.com>

2022-03-04 00:11:50 -06:00

..

1_gemm_xdl

Refactor threadwise copy using sfcurve (#101)

2022-03-04 00:11:50 -06:00

2_gemm_xdl_bias_relu

GEMM+Bias+ReLU+Add (#76)

2022-02-06 22:32:47 -06:00

3_gemm_xdl_bias_relu_add

GEMM+Bias+ReLU+Add (#76)

2022-02-06 22:32:47 -06:00

4_conv2d_fwd_xdl

Add gridwise GEMM pipeline (#89)

2022-02-23 17:23:49 -06:00

5_conv2d_fwd_xdl_bias_relu

Conv3d new (#94)

2022-02-22 22:45:28 -06:00

6_conv2d_fwd_xdl_bias_relu_add

Conv3d new (#94)

2022-02-22 22:45:28 -06:00

7_conv2d_fwd_xdl_bias_relu_atomic_add

Conv3d new (#94)

2022-02-22 22:45:28 -06:00

8_gemm_xdl_alpha_beta

Gemm alpha beta profiler (fp32 & fp16) (#91)

2022-02-21 11:35:21 -06:00

9_conv2d_fwd_xdl_int8

Conv3d new (#94)

2022-02-22 22:45:28 -06:00

10_conv3d_fwd_xdl

Conv3d new (#94)

2022-02-22 22:45:28 -06:00

11_convnd_fwd_xdl

Unify Convolution FWD XDL 1D/2D implementation. (#93)

2022-02-23 10:44:20 -06:00

12_conv2d_bwd_data_xdl

NHWC conv 2d: bwd fp32/fp16/bfp16/int8, Device level tuning and host API (#92)

2022-03-04 00:08:26 -06:00

CMakeLists.txt

NHWC conv 2d: bwd fp32/fp16/bfp16/int8, Device level tuning and host API (#92)

2022-03-04 00:08:26 -06:00