ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-26 08:00:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge commit 'e91ee8578cc9e493f12ee01055a35a405571effc' into develop

Browse Source
This commit is contained in:
assistant-librarian[bot]
2025-11-18 19:12:13 +00:00
parent 86a4127e31
commit 751e5d85a6
330 changed files with 424 additions and 425 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
include/ck_tile/ops/flatmm/kernel/flatmm_kernel.hpp
View File

@@ -363,8 +363,8 @@ struct FlatmmKernel
template <class KernelArgs>
__device__ SplitKBatchOffset(const KernelArgs& kargs, const std::size_t k_id = blockIdx.z)
{
constexpr auto N1 = TilePartitioner::BlockGemmShape::WarpTile::at(number<1>{});
constexpr auto K1 = TilePartitioner::BlockGemmShape::WarpTile::at(number<2>{});
constexpr auto N1 = BlockGemmShape::WarpTile::at(number<1>{});
constexpr auto K1 = BlockGemmShape::WarpTile::at(number<2>{});
const index_t K_t = kargs.k_batch * K1;
const index_t KRead = (kargs.K + K_t - 1) / K_t * K1;