Skip to content

Commit bf6294a

Browse files
AMD-mkulikowAMD-mkulikow
committed
[CK][Examples] Extending support for rdna3/4 in following examples:
-example_gemm_xdl_splitk_reduce_multi_d_fp16 -example_gemm_xdl_splitk_reduce_multi_d_bf16 -example_gemm_xdl_splitk_reduce_bf16A_i8B -example_gemm_xdl_splitk_reduce_bfp16 -example_splitk_gemm_bias_e_permute_xdl_fp32 -example_gemm_add_multiply_xdl_fp16 -example_complex_contraction_bilinear_xdl_fp32 -example_grouped_gemm_lower_triangle_scale_softmax_gemm_permute_xdl_fp16 -example_batched_gemm_bias_e_permute_xdl_fp16 -example_gemm_xdl_fp16 -example_gemm_xdl_fp16_av2 -example_gemm_xdl_wavelet_fp16 -example_gemm_add_add_fastgelu_xdl_bf16 -example_gemm_add_add_fastgelu_xdl_fp16 -example_gemm_add_add_fastgelu_xdl_fp32 -example_grouped_gemm_xdl_fp32 -example_grouped_gemm_xdl_fp16 -example_grouped_gemm_xdl_bf16 -example_cgemm_xdl_bf16 -example_cgemm_xdl_fp16 Signed-off-by: Michal Kulikowski <[email protected]>
1 parent 9292917 commit bf6294a

21 files changed

+64
-64
lines changed

‎example/01_gemm/gemm_xdl_fp16.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
// SPDX-License-Identifier: MIT
2-
// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
2+
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
33

44
#include "common.hpp"
55

@@ -37,7 +37,7 @@ using DeviceGemmInstance1 = ck::tensor_operation::device::DeviceGemm_Xdl_CShuffl
3737
// ######| | | | Type| Type| Type| Type| DataType| Elementwise| Elementwise| Elementwise| Spacialization| Prefetch| Size| Block| Block| Block| | | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave| _MBlock_MWaveMPerXdl| ScalarPerVector|
3838
// ######| | | | | | | | | Operation| Operation| Operation| | Stage| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NWaveNPerXdl| _NWaveNPerXdl|
3939
// ######| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
40-
< ALayout, BLayout, CLayout, ADataType, BDataType, CDataType, AccDataType, CShuffleDataType, AElementOp, BElementOp, CElementOp, GemmDefault, 1, 256, 256, 128, 32, 8, 2, 32, 32, 4, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, S<8, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 2, 0, 1, 2, S<1, 16, 1, 16>, 8, ck::LoopScheduler::Interwave, ck::PipelineVersion::v1>;
40+
< ALayout, BLayout, CLayout, ADataType, BDataType, CDataType, AccDataType, CShuffleDataType, AElementOp, BElementOp, CElementOp, GemmDefault, 1, 256, 256, 128, 32, 8, 2, 16, 16, 8, 4, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, S<8, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 2, 0, 1, 2, S<1, 16, 1, 16>, 4, ck::LoopScheduler::Interwave, ck::PipelineVersion::v1>;
4141
// clang-format on
4242

4343
using DeviceGemmInstance = DeviceGemmInstance1;

‎example/01_gemm/gemm_xdl_fp16_v2.cpp‎

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
// SPDX-License-Identifier: MIT
2-
// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
2+
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
33

44
#include "common.hpp"
55

@@ -33,13 +33,13 @@ using DeviceGemmInstance =
3333
2, 256,
3434
256, 256,
3535
32, 8, 4,
36-
32, 32,
37-
4, 4,
36+
16, 16,
37+
8, 8,
3838
S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>,
3939
2, 8, 8, 0,
4040
S<8, 32, 1>, S<0, 2, 1>, S<0, 2, 1>,
4141
1, 8, 4, 0,
42-
1, 1, S<1, 32, 1, 8>, 8,
42+
1, 1, S<1, 32, 1, 8>, 4,
4343
ck::LoopScheduler::Default, ck::PipelineVersion::v1>;
4444
// clang-format on
4545

‎example/01_gemm/gemm_xdl_wavelet_fp16.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
// SPDX-License-Identifier: MIT
2-
// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
2+
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
33

44
#include "common.hpp"
55

@@ -29,7 +29,7 @@ using DeviceGemmInstance = ck::tensor_operation::device::DeviceGemm_Xdl_WaveletM
2929
// ######| | | | Type| Type| Type| DataType| Type| Elementwise| Elementwise| Elementwise| Spacialization| Prefetch| ThreadGroupSize| ThreadGroupSize| Block| Block| Block| | | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave| _MBlock_MWaveMPerXdl| ScalarPerVector|
3030
// ######| | | | | | | | | Operation| Operation| Operation| | Stage| | | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NWaveNPerXdl| _NWaveNPerXdl|
3131
// ######| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
32-
< ALayout, BLayout, CLayout, ADataType, BDataType, AccDataType, F16, CDataType, AElementOp, BElementOp, CElementOp, GemmDefault, 1, 256, 256, 256, 128, 32, 8, 8, 32, 32, 4, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, 1, 1, S<1, 32, 1,8>, 8>;
32+
< ALayout, BLayout, CLayout, ADataType, BDataType, AccDataType, F16, CDataType, AElementOp, BElementOp, CElementOp, GemmDefault, 1, 256, 256, 256, 128, 32, 8, 8, 16, 16, 8, 4, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, 1, 1, S<1, 32, 1,8>, 4>;
3333
// clang-format on
3434

3535
using DeviceGemmInstance = DeviceGemmInstance;

‎example/04_gemm_add_add_fastgelu/gemm_add_add_fastgelu_xdl_bf16.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
// SPDX-License-Identifier: MIT
2-
// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
2+
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
33

44
#include "common.hpp"
55

@@ -32,7 +32,7 @@ using DeviceOpInstance = ck::tensor_operation::device::DeviceGemmMultipleD_Xdl_C
3232
//######| | | | | Type| Type| Type| DataType| Type| Type| Elementwise| Elementwise| Elementwise| Spacialization| Prefetch| Size| Block| Block| Block| | | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave| _MBlock_MWaveMPerXdl| ScalarPerVector|
3333
//######| | | | | | | | | | | Operation| Operation| Operation| | Stage| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NWaveNPerXdl| _NWaveNPerXdl|
3434
//######| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
35-
< ALayout, BLayout, DsLayout, ELayout, ADataType, BDataType, AccDataType, CShuffleDataType, DsDataType, EDataType, AElementOp, BElementOp, CDEElementOp, GemmDefault, 1, 256, 256, 128, 32, 8, 8, 32, 32, 4, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, 1, 1, S<1, 32, 1, 8>, 8>;
35+
< ALayout, BLayout, DsLayout, ELayout, ADataType, BDataType, AccDataType, CShuffleDataType, DsDataType, EDataType, AElementOp, BElementOp, CDEElementOp, GemmDefault, 1, 256, 256, 128, 32, 8, 8, 16, 16, 8, 4, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, 1, 1, S<1, 32, 1, 8>, 4>;
3636
// clang-format on
3737

3838
using ReferenceGemmInstance = ck::tensor_operation::host::ReferenceGemm<ADataType,

‎example/04_gemm_add_add_fastgelu/gemm_add_add_fastgelu_xdl_fp16.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
// SPDX-License-Identifier: MIT
2-
// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
2+
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
33

44
#include "common.hpp"
55

@@ -32,7 +32,7 @@ using DeviceOpInstance = ck::tensor_operation::device::DeviceGemmMultipleD_Xdl_C
3232
//######| | | | | Type| Type| Type| DataType| Type| Type| Elementwise| Elementwise| Elementwise| Spacialization| Prefetch| Size| Block| Block| Block| | | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave| _MBlock_MWaveMPerXdl| ScalarPerVector|
3333
//######| | | | | | | | | | | Operation| Operation| Operation| | Stage| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NWaveNPerXdl| _NWaveNPerXdl|
3434
//######| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
35-
< ALayout, BLayout, DsLayout, ELayout, ADataType, BDataType, AccDataType, CShuffleDataType, DsDataType, EDataType, AElementOp, BElementOp, CDEElementOp, GemmDefault, 1, 256, 256, 128, 32, 8, 8, 32, 32, 4, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, 1, 1, S<1, 32, 1, 8>, 8>;
35+
< ALayout, BLayout, DsLayout, ELayout, ADataType, BDataType, AccDataType, CShuffleDataType, DsDataType, EDataType, AElementOp, BElementOp, CDEElementOp, GemmDefault, 1, 256, 256, 128, 32, 8, 8, 16, 16, 8, 4, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, 1, 1, S<1, 32, 1, 8>, 4>;
3636
// clang-format on
3737

3838
using ReferenceGemmInstance = ck::tensor_operation::host::ReferenceGemm<ADataType,

‎example/04_gemm_add_add_fastgelu/gemm_add_add_fastgelu_xdl_fp32.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
1+
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
22

33
#include "common.hpp"
44

@@ -31,7 +31,7 @@ using DeviceOpInstance = ck::tensor_operation::device::DeviceGemmMultipleD_Xdl_C
3131
//######| | | | | Type| Type| Type| DataType| Type| Type| Elementwise| Elementwise| Elementwise| Spacialization| Prefetch| Size| Block| Block| Block| | | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave| _MBlock_MWaveMPerXdl| ScalarPerVector|
3232
//######| | | | | | | | | | | Operation| Operation| Operation| | Stage| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NWaveNPerXdl| _NWaveNPerXdl|
3333
//######| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
34-
< ALayout, BLayout, DsLayout, ELayout, ADataType, BDataType, AccDataType, CShuffleDataType, DsDataType, EDataType, AElementOp, BElementOp, CDEElementOp, GemmDefault, 1, 256, 256, 128, 32, 8, 8, 32, 32, 4, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 1, 1, 1, S<1, 32, 1, 8>, 4>;
34+
< ALayout, BLayout, DsLayout, ELayout, ADataType, BDataType, AccDataType, CShuffleDataType, DsDataType, EDataType, AElementOp, BElementOp, CDEElementOp, GemmDefault, 1, 256, 256, 128, 32, 8, 8, 16, 16, 8, 4, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 1, 1, 1, S<1, 32, 1, 8>, 2>;
3535
// clang-format on
3636

3737
using ReferenceGemmInstance = ck::tensor_operation::host::ReferenceGemm<ADataType,

‎example/15_grouped_gemm/grouped_gemm_xdl_bf16.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
// SPDX-License-Identifier: MIT
2-
// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
2+
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
33

44
#include <iostream>
55
#include <numeric>
@@ -54,7 +54,7 @@ using DeviceGemmInstance = ck::tensor_operation::device::DeviceGroupedGemm_Xdl
5454
//######| | | | | Type| Type| Type| DataType| Type| Type| Elementwise| Elementwise| Elementwise| Spacialization| Prefetch| Size| Block| Block| Block| | | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MXdlPerWave| NXdlPerWave| _MBlock_MWaveMPerXdl| ScalarPerVector|
5555
//######| | | | | | | | | | | Operation| Operation| Operation| | Stage| | | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NWaveNPerXdl| _NWaveNPerXdl|
5656
//######| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
57-
< ALayout, BLayout, DsLayout, ELayout, ADataType, BDataType, AccDataType, CShuffleDataType, DsDataType, EDataType, AElementOp, BElementOp, CDEElementOp, GemmDefault, 1, 256, 256, 128, 32, 8, 8, 32, 32, 4, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, 1, 1, S<1, 32, 1, 8>, 8>;
57+
< ALayout, BLayout, DsLayout, ELayout, ADataType, BDataType, AccDataType, CShuffleDataType, DsDataType, EDataType, AElementOp, BElementOp, CDEElementOp, GemmDefault, 1, 256, 256, 128, 32, 8, 8, 16, 16, 8, 4, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 1, 1, 1, S<1, 32, 1, 8>, 4>;
5858
// clang-format on
5959

6060
#include "run_grouped_gemm_example.inc"

0 commit comments

Comments
 (0)