Skip to content

GitLab

Commit 993ec45c authored by ltqin's avatar ltqin
Browse files

fix input parameters

parent c3de33fa
Hide whitespace changes
Inline Side-by-side
Showing with 7 additions and 7 deletions
example/01_gemm/gemm_xdl_skip_b_lds_fp16.cpp
View file @ 993ec45c
...@@ -56,7 +56,7 @@ using AccDataType = float; ...@@ -56,7 +56,7 @@ using AccDataType = float;
#else #else
// < F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 64, 16, 4, 4, 16, 16, 1, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 4, 7, 1>; // < F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 64, 16, 4, 4, 16, 16, 1, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 4, 7, 1>;
// < F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 128, 32, 4, 4, 32, 32, 1, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 4, 7, 1>; // < F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 128, 32, 4, 4, 32, 32, 1, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 4, 7, 1>;
< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 64, 64, 4, 4, 32, 32, 1, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 4, 7, 1>; < F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 4, 7, 1>;
using ADataType = float; using ADataType = float;
using BDataType = float; using BDataType = float;
using CDataType = float; using CDataType = float;
...@@ -69,9 +69,9 @@ using DeviceGemmInstance = ck::tensor_operation::device::DeviceGemmXdl ...@@ -69,9 +69,9 @@ using DeviceGemmInstance = ck::tensor_operation::device::DeviceGemmXdl
//###########| Type| Type| Type| Type| | | | Elementwise| Elementwise| Elementwise|Spacialization| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| SrcDstVectorDim| DstScalar| //###########| Type| Type| Type| Type| | | | Elementwise| Elementwise| Elementwise|Spacialization| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| SrcDstVectorDim| DstScalar|
//###########| | | | | | | | Operation| Operation| Operation| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | | PerVector| //###########| | | | | | | | Operation| Operation| Operation| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | | PerVector|
//###########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | //###########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
// < F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 64, 16, 16, 4, 4, 16, 16, 1, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 7, 1, 2>; // < F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 64, 16, 4, 4, 16, 16, 1, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 7, 1, 2>;
// < F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 128, 32, 4, 4, 32, 32, 1, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 7, 1, 2>; // < F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 128, 32, 4, 4, 32, 32, 1, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 7, 1, 2>;
< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 64, 128, 4, 4, 32, 32, 2, 1, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 7, 1, 2>; < F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 128, 64, 4, 4, 32, 32, 1, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, true, 7, 1, 2>;
using ADataType = float; using ADataType = float;
using BDataType = float; using BDataType = float;
using CDataType = float; using CDataType = float;
...@@ -107,13 +107,13 @@ int main(int argc, char* argv[]) ...@@ -107,13 +107,13 @@ int main(int argc, char* argv[])
// GEMM shape // GEMM shape
#if 1 #if 1
ck::index_t M = 64; ck::index_t M = 128 * 360;
ck::index_t N = 4096; ck::index_t N = 64;
ck::index_t K = 4096; ck::index_t K = 4096;
ck::index_t StrideA = 4096; ck::index_t StrideA = 4096;
ck::index_t StrideB = 4096; ck::index_t StrideB = 4096;
ck::index_t StrideC = 4096; ck::index_t StrideC = N;
#else #else
ck::index_t M = 16; ck::index_t M = 16;
ck::index_t N = 16; ck::index_t N = 16;
......
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