ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-03 21:21:22 +00:00
Code Issues Packages Projects Releases Wiki Activity

gemm/Conv xdlops + dlops quantization (#625)

Browse Source 
* Add conv perlayer quantization

* Add gemm_dlops quantization

* Support int8 for innerproduct

* Refine gemm dlops int8 kernel parameter

* Support gfx908(MI100) and gfx90a(MI200)

* clang-format

* Rename example number

* Support different layout for d tensor

* Add conv dlops perchannel quantization example

* Move to example 40

* Extract the common code for different platform (dlops and xdlops)

* Move ot subfolder. Prepare to add other op of quantization

* Refine the quantization instance library

* Add conv dl instances and client example

* Remove unnecessary type

* Add gemm quantization instance

* Add external api and client example

* Refine num_bytes

* Separete different layout to different cpp

* Add more xdl instances

* Revert "Remove unnecessary type"

This reverts commit 820869182f.

* Remove CShuffleDataType in dlops
Let acc and CShuffleDataType be the same in xdlops

---------

Co-authored-by: zjing14 <zhangjing14@gmail.com>
This commit is contained in:
rocking5566
2023-03-16 04:29:40 +08:00
committed by GitHub
parent a2d5ca8e95
commit 16dc18e0f9
60 changed files with 3186 additions and 717 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
include/ck/tensor_operation/gpu/grid/gridwise_gemm_dl_multiple_d.hpp
View File

@@ -185,8 +185,10 @@ struct GridwiseGemmDlMultipleD_km_kn_mn
return b_grid_desc_k0_n0_n1_k1;
}
// E desc for destination in blockwise copy
template <typename CGridDesc_M_N_>
__host__ __device__ static constexpr auto
MakeCGridDescriptor_M0_M10_M11_N0_N10_N11(const CGridDesc_M_N& c_grid_desc_m_n)
MakeCGridDescriptor_M0_M10_M11_N0_N10_N11(const CGridDesc_M_N_& c_grid_desc_m_n)
{
const auto M = c_grid_desc_m_n.GetLength(I0);
const auto N = c_grid_desc_m_n.GetLength(I1);