ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-04 13:41:24 +00:00
Code Issues Packages Projects Releases Wiki Activity

supporting prefill shapes for preshuffle block scale gemm (#2975)

Browse Source 
* debugging

* debugging for prefill shapes

* comment unused code

* fix for prefill shapes

* clearing up the code

* add int4 to universal gemm example

* clang formatted

* adding test for prefill shapes in block scale gemm

* lil improv on the block pipeline

* Address Review Comment

---------

Co-authored-by: ThomasNing <thomas.ning@amd.com>
This commit is contained in:
Khushbu Agarwal
2025-10-10 15:36:24 -07:00
committed by GitHub
parent 9d060d3e3c
commit 3c39d279ab
10 changed files with 137 additions and 89 deletions
Download Patch File Download Diff File Expand all files Collapse all files

41
test/ck_tile/gemm_block_scale/test_gemm_quant_fixtures.hpp
View File

@@ -57,26 +57,10 @@ struct GemmConfigPreshuffleQuantTransposeC : public GemmConfigBase
static constexpr bool TransposeC = true;
};
struct GemmConfigPreshuffleB
struct GemmConfigPreshuffleBDecode : public GemmConfigBase
{
static constexpr bool kPadM = false;
static constexpr bool kPadN = false;
static constexpr bool kPadK = false;
static constexpr bool PermuteA = false;
static constexpr bool PermuteB = false;
static constexpr bool TransposeC = false;
static constexpr bool UseStructuredSparsity = false;
static constexpr int kBlockPerCu = 1;
static constexpr ck_tile::index_t TileParitionerGroupNum = 8;
static constexpr ck_tile::index_t TileParitionerM01 = 4;
static constexpr auto Scheduler = ck_tile::GemmPipelineScheduler::Intrawave;
static constexpr ck_tile::index_t NumWaveGroups = 1;
static constexpr bool PreshuffleQuant = false;
static constexpr bool PreshuffleB = true;
static constexpr bool DoubleSmemBuffer = true;
static constexpr bool PreshuffleB = true;
static constexpr bool DoubleSmemBuffer = true;
// Default GEMM tile sizes for tests
static constexpr ck_tile::index_t M_Tile = 16;
@@ -92,6 +76,25 @@ struct GemmConfigPreshuffleB
static constexpr ck_tile::index_t K_Warp_Tile = 64;
};
struct GemmConfigPreshuffleBPrefill : public GemmConfigBase
{
static constexpr bool PreshuffleB = true;
static constexpr bool DoubleSmemBuffer = true;
// Default GEMM tile sizes for tests
static constexpr ck_tile::index_t M_Tile = 128;
static constexpr ck_tile::index_t N_Tile = 128;
static constexpr ck_tile::index_t K_Tile = 128;
static constexpr ck_tile::index_t M_Warp = 1;
static constexpr ck_tile::index_t N_Warp = 4;
static constexpr ck_tile::index_t K_Warp = 1;
static constexpr ck_tile::index_t M_Warp_Tile = 16;
static constexpr ck_tile::index_t N_Warp_Tile = 16;
static constexpr ck_tile::index_t K_Warp_Tile = 64;
};
template <typename Tuple>
class TestCkTileGemmAQuant : public TestCkTileGemmQuantBase<Tuple, TestCkTileGemmAQuant<Tuple>>
{

13
test/ck_tile/gemm_block_scale/test_gemm_quant_typed.cpp
View File

@@ -62,10 +62,15 @@ using BQuantTypes = ::testing::Types<
// clang-format off
using BPreshuffleBQuantTypes = ::testing::Types<
std::tuple<RowMajor, ColumnMajor, RowMajor, FP8, FP8, float, Half, BQuantGrouped, GemmConfigPreshuffleB, GroupSize>,
std::tuple<RowMajor, ColumnMajor, RowMajor, BF8, BF8, float, Half, BQuantGrouped, GemmConfigPreshuffleB, GroupSize>,
std::tuple<RowMajor, ColumnMajor, RowMajor, FP8, PkInt4, FP8, Half, BQuantGrouped, GemmConfigPreshuffleB, GroupSize>,
std::tuple<RowMajor, ColumnMajor, RowMajor, BF8, PkInt4, BF8, Half, BQuantGrouped, GemmConfigPreshuffleB, GroupSize>
std::tuple<RowMajor, ColumnMajor, RowMajor, FP8, FP8, float, Half, BQuantGrouped, GemmConfigPreshuffleBDecode, GroupSize>,
std::tuple<RowMajor, ColumnMajor, RowMajor, BF8, BF8, float, Half, BQuantGrouped, GemmConfigPreshuffleBDecode, GroupSize>,
std::tuple<RowMajor, ColumnMajor, RowMajor, FP8, PkInt4, FP8, Half, BQuantGrouped, GemmConfigPreshuffleBDecode, GroupSize>,
std::tuple<RowMajor, ColumnMajor, RowMajor, BF8, PkInt4, BF8, Half, BQuantGrouped, GemmConfigPreshuffleBDecode, GroupSize>,
std::tuple<RowMajor, ColumnMajor, RowMajor, FP8, FP8, float, Half, BQuantGrouped, GemmConfigPreshuffleBPrefill, GroupSize>,
std::tuple<RowMajor, ColumnMajor, RowMajor, BF8, BF8, float, Half, BQuantGrouped, GemmConfigPreshuffleBPrefill, GroupSize>,
std::tuple<RowMajor, ColumnMajor, RowMajor, FP8, PkInt4, FP8, Half, BQuantGrouped, GemmConfigPreshuffleBPrefill, GroupSize>,
std::tuple<RowMajor, ColumnMajor, RowMajor, BF8, PkInt4, BF8, Half, BQuantGrouped, GemmConfigPreshuffleBPrefill, GroupSize>
>;
// clang-format off