ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-02 20:51:23 +00:00
Code Issues Packages Projects Releases Wiki Activity

[CK_TILE] layernorm support fused-quant/fused-add (#1604)

Browse Source 
* add prenorm/postnorm support, refactor using generate.py

* update README

* update README

* fix format

* update some description and fix format

* update format

* format

* use non-raw for loading

* format and update n4096

* dynamic-quant ready

* update readme

* support fused dynamic-quant

* update fused-quant, with smooth

* update README

* update args

* update some based on comment
This commit is contained in:
carlushuang
2024-10-31 14:54:53 +08:00
committed by GitHub
parent 9a8a52130d
commit c3a4800c5f
61 changed files with 1790 additions and 766 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
include/ck_tile/ops/reduce/block/block_reduce.hpp
View File

@@ -301,7 +301,10 @@ struct BlockReduce2D
.get_static_tile_distribution_encoding(),
ReduceDim{}));
return make_static_distributed_tensor<InDataType>(acc_dstr);
auto dst_ = make_static_distributed_tensor<InDataType>(acc_dstr);
// init acc_tensor
tile_elementwise_inout([&](auto& x_) { x_ = type_convert<InDataType>(reduce_init); }, dst_);
return dst_;
}
// return number of pixels each lane need to reduce