ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-05 14:11:29 +00:00
Code Issues Packages Projects Releases Wiki Activity

merge flatmm -scale

Browse Source
This commit is contained in:
lalala-sh
2025-07-24 08:46:51 +00:00
parent 4066454483 b908f5e803
commit 1239d8a546
8 changed files with 1265 additions and 654 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
example/ck_tile/18_flatmm/flatmm_basic.hpp
View File

@@ -83,10 +83,10 @@ struct FlatmmConfig16
template <typename DataType>
struct FlatmmConfig16_950 : public FlatmmConfig16<DataType>
{
static constexpr ck_tile::index_t N_Tile = 256;
static constexpr ck_tile::index_t K_Tile = 256 / sizeof(DataType);
static constexpr ck_tile::index_t N_Tile = 256;
static constexpr ck_tile::index_t K_Tile = 256 / sizeof(DataType);
static constexpr ck_tile::index_t K_Warp_Tile = sizeof(DataType) == 2 ? 32 : 128;
static constexpr int kBlockPerCu = 2;
static constexpr int kBlockPerCu = 1;
};
template <typename ADataType>
@@ -177,6 +177,9 @@ template <typename FlatmmConfig,
typename BLayout,
typename DsLayout,
typename ELayout,
typename ScaleM,
typename ScaleN,
bool persistent,
typename CDEElementWise>
float flatmm_calc(const ck_tile::FlatmmHostArgs<>& args, const ck_tile::stream_config& s);
float flatmm_calc(const ck_tile::ScaleFlatmmHostArgs<ScaleM, ScaleN>& args,
const ck_tile::stream_config& s);