ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-20 12:59:49 +00:00
Code Issues Packages Projects Releases Wiki Activity

Conv3d new (#94)

Browse Source 
* conv3d compiles but has memory error

* conv3d works

* fix performance issue by using __builtin_amdgc_readfirstlane

* change MakeBlock2CTileMap to MakeDefaultBlock2CTileMap; change c_blockid_to* to cblockid_to*

* clang-format

* remove CK_EXPERIMENTAL_PASS_TENSOR_DECRIPTOR_BY_*; moved wrapper into DeviceConv3d

* format

* remove useless marc

* add comment

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

[ROCm/composable_kernel commit: 6dfb92bbef]
This commit is contained in:
Jianfeng Yan
2022-02-22 22:45:28 -06:00
committed by GitHub
parent 4dd24cf943
commit c2e3fa5c91
61 changed files with 2255 additions and 1725 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
test/conv2d_fwd/main.cpp
View File

@@ -130,13 +130,13 @@ int main(int argc, char* argv[])
const ck::index_t Ho = (Hi + in_left_pad_h + in_right_pad_h - YEff) / conv_stride_h + 1;
const ck::index_t Wo = (Wi + in_left_pad_w + in_right_pad_w - XEff) / conv_stride_w + 1;
const std::vector<ck::index_t> input_spatial_lengths{{Hi, Wi}};
const std::vector<ck::index_t> filter_spatial_lengths{{Y, X}};
const std::vector<ck::index_t> output_spatial_lengths{{Ho, Wo}};
const std::vector<ck::index_t> conv_filter_strides{{conv_stride_h, conv_stride_w}};
const std::vector<ck::index_t> conv_filter_dilations{{conv_dilation_h, conv_dilation_w}};
const std::vector<ck::index_t> input_left_pads{{in_left_pad_h, in_left_pad_w}};
const std::vector<ck::index_t> input_right_pads{{in_right_pad_h, in_right_pad_w}};
const std::vector<ck::index_t> input_spatial_lengths{Hi, Wi};
const std::vector<ck::index_t> filter_spatial_lengths{Y, X};
const std::vector<ck::index_t> output_spatial_lengths{Ho, Wo};
const std::vector<ck::index_t> conv_filter_strides{conv_stride_h, conv_stride_w};
const std::vector<ck::index_t> conv_filter_dilations{conv_dilation_h, conv_dilation_w};
const std::vector<ck::index_t> input_left_pads{in_left_pad_h, in_left_pad_w};
const std::vector<ck::index_t> input_right_pads{in_right_pad_h, in_right_pad_w};
auto f_host_tensor_descriptor =
[](std::size_t N_, std::size_t C_, std::size_t H, std::size_t W) {