Skip to content

GitLab

Commit 809d7dfb authored by Jing Zhang's avatar Jing Zhang
Browse files

format

parent 2724c519
Hide whitespace changes
Inline Side-by-side
Showing with 43 additions and 43 deletions
example/02_gemm_bilinear/gemm_bilinear_wmma_fp16.cpp
View file @ 809d7dfb
...@@ -65,49 +65,49 @@ using CDEElementOp = AlphaBetaAdd; ...@@ -65,49 +65,49 @@ using CDEElementOp = AlphaBetaAdd;
static constexpr auto GemmSpec = ck::tensor_operation::device::GemmSpecialization::MNKPadding; static constexpr auto GemmSpec = ck::tensor_operation::device::GemmSpecialization::MNKPadding;
using DeviceOpInstance = using DeviceOpInstance = ck::tensor_operation::device::DeviceGemmMultipleD_Wmma_CShuffle<
ck::tensor_operation::device::DeviceGemmMultipleD_Wmma_CShuffle<ALayout, ALayout,
BLayout, BLayout,
ck::Tuple<DLayout>, ck::Tuple<DLayout>,
ELayout, ELayout,
ADataType, ADataType,
BDataType, BDataType,
AccDataType, AccDataType,
CShuffleDataType, CShuffleDataType,
ck::Tuple<DDataType>, ck::Tuple<DDataType>,
EDataType, EDataType,
AElementOp, AElementOp,
BElementOp, BElementOp,
CDEElementOp, CDEElementOp,
GemmSpec, GemmSpec,
1, 2, // Prefetch stage
128, 128, // BlockSize
64, 128, // MPerBlock
64, 64, // NPerBlock
64, 64, // KPerBlock
4, 8, // K1
16, 16, // MPerWmma
16, 16, // NPerWmma
1, 4, // M-Repeat // M-PerWmma / M-Repeat = M-Wave
4, 2, // N-Repeat // N-PerWmma / N-Repeat = N-Wave
S<4, 32, 1>, S<4, 32, 1>,
S<1, 0, 2>, S<1, 0, 2>,
S<1, 0, 2>, S<1, 0, 2>,
2, 2,
4, 8,
4, 8,
true, true,
S<4, 32, 1>, S<4, 32, 1>,
S<1, 0, 2>, S<1, 0, 2>,
S<1, 0, 2>, S<1, 0, 2>,
2, 2,
4, 8,
4, 8,
true, true,
1, 1, // C shuffle (M Repeat) Per store
1, 1, // C shuffle (N Repeat) Per store
S<1, 64, 1, 2>, S<1, 32, 1, 4>,
8>; 8>;
int main(int argc, char* argv[]) int main(int argc, char* argv[])
{ {
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment