ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-14 10:09:41 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
0cc91c73d8d42e0c8b1476008d171a67d915e275
composable_kernel/include/ck/tensor_operation/gpu/block
History
Anthony Chang 7d211aa652 Hotfix LDS data hazard in fused attention (#360) 
* avoid LDS data hazard in gemm_softmax_gemm pipeline

* trivial refactors

* comments

* shrink blockwise gemm v2 thread buffer size

* reclaim A block lds space when during 2nd gemm

* amend

* amend

[ROCm/composable_kernel commit: c961ce9226]
2022-08-15 12:04:20 -05:00
..
blockwise_gemm_dl_v2r3.hpp
add license in file (#303)
2022-06-24 23:32:43 -05:00
blockwise_gemm_dlops_v2r2.hpp
add license in file (#303)
2022-06-24 23:32:43 -05:00
blockwise_gemm_dlops_v3.hpp
add license in file (#303)
2022-06-24 23:32:43 -05:00
blockwise_gemm_xdlops_skip_b_lds.hpp
Fused GEMM+GEMM (#351)
2022-08-13 09:18:58 -05:00
blockwise_gemm_xdlops.hpp
Hotfix LDS data hazard in fused attention (#360)
2022-08-15 12:04:20 -05:00
blockwise_softmax.hpp
Fused attention (#345)
2022-08-13 00:16:14 -05:00
blockwise_tensor_slice_transfer_v5r1.hpp
add license in file (#303)
2022-06-24 23:32:43 -05:00
blockwise_welford.hpp
Layernorm welford (#346)
2022-08-13 09:43:18 -05:00
reduction_functions_blockwise.hpp
Fused attention (#345)
2022-08-13 00:16:14 -05:00
thread_group_tensor_slice_transfer_v4r1.hpp
add license in file (#303)
2022-06-24 23:32:43 -05:00
thread_group_tensor_slice_transfer_v6r1.hpp
add license in file (#303)
2022-06-24 23:32:43 -05:00
thread_group_tensor_slice_transfer_v6r2.hpp
add license in file (#303)
2022-06-24 23:32:43 -05:00
thread_group_tensor_slice_transfer_v6r3.hpp
add license in file (#303)
2022-06-24 23:32:43 -05:00
thread_group_tensor_slice_transfer_v7.hpp
add license in file (#303)
2022-06-24 23:32:43 -05:00