[CK_TILE] Tile loop persistent gemm kernel (#2191)

* Implement tile loop persistent gemm kernel

* Enable timing

* Add tests for persistent gemm

* Fix formatting

* Fix gemm_basic

* Rename True/False to Persistent/NonPersistent

* Use only one set of layouts for persistent tests

* Fix gemm example persistent template parameter

* Fix formatting
This commit is contained in:
Sami Remes
2025-06-04 11:46:28 +03:00
committed by GitHub
parent 52b4860a30
commit ffb52783d0
10 changed files with 232 additions and 18 deletions

View File

@@ -18,9 +18,12 @@ template <typename ADataType,
typename CDataType,
typename ALayout,
typename BLayout,
typename CLayout>
typename CLayout,
bool Persistent>
float gemm_calc(const ck_tile::GemmHostArgs& args, const ck_tile::stream_config& s)
{
if constexpr(Persistent)
std::cout << "WARNING: Ignoring persistent kernel option for basic gemm." << std::endl;
// The kPadM, kPadN, kPadK & kBlockPerCu should also come from the Codegen part.
constexpr bool kPadM = false;
constexpr bool kPadN = false;