ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-04 13:41:24 +00:00
Code Issues Packages Projects Releases Wiki Activity

[ck] add gridwise base class for in all xdl kernel (#186) (#3544)

Browse Source 
1. Add base class GridwiseGemm_xdl_cshuffle_base for all gridwise_gemm_xdl classes.
- to select correct LDS layout and epilogue behavior , three additional parameters is added.
- ForceNaiveLdsLayout: disable XOR based LDS layout when it is true
- DirectLoad: pipeline only use directload, we need force naive layout and ignore any padding on gfx9
- IsMxGemm: epilogue has two addtional dimensions
2. Move all LDS descriptor layout related fucntion to base class, including
- GetABlockDescriptor_AK0PerBlock_MPerBlock_AK1
- GetBBlockDescriptor_BK0PerBlock_NPerBlock_BK1
- GetCShuffleBlockDescriptor_MBlock_MPerBlock_NBlock_NPerBlock
3. Move several LDS related helper funtions to base class, including
- GetSharedMemoryNumberOfByte
- GetABlockDescriptor_AKB_AK0PerBlock_MPerBlock_AK1
- GetBBlockDescriptor_BKB_BK0PerBlock_NPerBlock_BK1
- GetCBlockDescriptor_MBlock_NXdlPerWave_MWaveMPerXdl_NBlock_NXdlPerWave_NWaveNPerXdl
4. Move all c epilogue related code to base class, and 4 kind of implementation are provided
- RunEpilogueNoShuffle
- RunEpilogue
- RunMultiDEpilogue
- RunMoeEpilogue
This commit is contained in:
linqunAMD
2026-01-28 04:49:47 +08:00
committed by GitHub
parent b737f1dee5
commit 23cefda140
85 changed files with 9016 additions and 20081 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
include/ck/tensor_operation/gpu/block/thread_group_tensor_slice_transfer_v7r3.hpp
View File

@@ -132,7 +132,7 @@ struct ThreadGroupTensorSliceTransfer_v7r3
}
template <typename T>
using is_tuple = decltype(std::declval<T&>().IsTuple());
using is_tuple = decltype(declval<T&>().IsTuple());
template <typename DstBuffers, index_t ThreadScratchId = 0>
__device__ void RunWrite(const DstDescs& dst_descs,

2
include/ck/tensor_operation/gpu/block/thread_group_tensor_slice_transfer_v7r3_scatter.hpp
View File

@@ -144,7 +144,7 @@ struct ThreadGroupTensorSliceTransfer_v7r3_scatter
}
template <typename T>
using is_tuple = decltype(std::declval<T&>().IsTuple());
using is_tuple = decltype(declval<T&>().IsTuple());
template <typename DstBuffers, index_t ThreadScratchId = 0>
__device__ void RunWrite(const DstDescs& dst_descs,