ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-16 19:09:59 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge commit '753b6227c507c45f4838971067f3cf158a392d55' into develop

Browse Source
This commit is contained in:
assistant-librarian[bot]
2025-08-13 23:12:01 +00:00
parent ffae627353
commit 9f8adac9f9
17 changed files with 239 additions and 257 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
example/ck_tile/21_elementwise/elementwise_example_transpose.cpp
View File

@@ -73,7 +73,7 @@ bool run(const ck_tile::ArgParser& arg_parser)
using BlockWarps = ck_tile::sequence<8>;
using WarpTile = ck_tile::sequence<64>;
using Shape = ck_tile::ElementWiseShape<BlockWarps, BlockTile, WarpTile, ComputeDataType>;
using Shape = ck_tile::ElementWiseShape<BlockWarps, BlockTile, WarpTile, XDataType>;
// Problem definition for a single input tensor
using Problem = ck_tile::ElementWisePipelineProblem<XDataType,
@@ -86,7 +86,8 @@ bool run(const ck_tile::ArgParser& arg_parser)
ck_tile::index_t total_elements = M * N;
constexpr ck_tile::index_t kBlockSize = 64 * BlockWarps::at(ck_tile::number<0>{});
constexpr ck_tile::index_t kBlockSize =
ck_tile::get_warp_size() * BlockWarps::at(ck_tile::number<0>{});
constexpr ck_tile::index_t kBlockPerCu = 1;
constexpr ck_tile::index_t elements_per_block = BlockTile::at(ck_tile::number<0>{});
ck_tile::index_t kGridSize = (total_elements + elements_per_block - 1) / elements_per_block;