Skip to content

New mma_atoms and copy_atoms in bmg_grouped_gemm_fp8 #579

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
wants to merge 1 commit into
base: main
Choose a base branch
from
+1,026 −8
Draft

New mma_atoms and copy_atoms in bmg_grouped_gemm_fp8 #579

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 8 additions & 5 deletions examples/09_bmg_grouped_gemm_f8/09_bmg_grouped_gemm_f8.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -576,19 +576,22 @@ int launcher(Options& options)
using LayoutC = cutlass::layout::RowMajor;
using LayoutD = cutlass::layout::RowMajor;

using GmemTiledCopyA = XE_2D_U8x32x32_LD_V;
using GmemTiledCopyB = XE_2D_U8x32x32_LD_V;
using GmemTiledCopyA = void;
using GmemTiledCopyB = void;

// Workgroup-level tile
using TileShape = Shape<_256, _256, _32>;

using TiledMma =
typename TiledMMAHelper<MMA_Atom<XE_8x16x16_F32F16F16F32_TT>, Layout<TileShape>,
Layout<Shape<_8, _4, _1>, Stride<_4, _1, _0>>>::TiledMMA;
typename TiledMMAHelper<
MMA_Atom<XE_DPAS_TT<8, ElementAccumulator, ElementA>>,
Layout<TileShape>,
Layout<Shape<_8, _4, _1>, Stride<_4, _1, _0>>
>::TiledMMA;

constexpr int PipelineStages = 2;
// Dispatch to grouped gemm algorithm
using GEMMDispatchPolicy = cutlass::gemm::MainloopIntelXeXMX16GroupFP8<PipelineStages>;
using GEMMDispatchPolicy = cutlass::gemm::MainloopXeL1StagedGroup<PipelineStages>;
using EpilogueDispatchPolicy = cutlass::epilogue::IntelXeXMX16Group;

using EpilogueOp = cutlass::epilogue::fusion::LinearCombination<ElementOutput, ElementComputeEpilogue,
Expand Down
Loading
Loading