diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_comp_instance.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_comp_instance.hpp index f491474d38..6c0ba2f932 100644 --- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_comp_instance.hpp +++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_comp_instance.hpp @@ -1,9 +1,10 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/ck.hpp" #include "ck/tensor_operation/gpu/device/tensor_layout.hpp" #include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle_v3.hpp" +#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp" #include "ck/tensor_operation/gpu/device/convolution_forward_specialization.hpp" #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp" #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp" @@ -89,7 +90,12 @@ using device_grouped_conv_fwd_xdl_bf16_comp_instances = std::tuple< DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Interwave, BlockGemmPipelineVersion::v1>, DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Intrawave, BlockGemmPipelineVersion::v3>, DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Intrawave, BlockGemmPipelineVersion::v3>, - DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Intrawave, BlockGemmPipelineVersion::v3> + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Intrawave, BlockGemmPipelineVersion::v3>, + // mfma 16x16 + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, BF16, BF16, F32, BF16, DsLayout, BF16, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding, 1,256, 64, 64, 32, 8, 8, 16, 16, 2, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 2, 8, 1, 1, 1, S<1, 32, 1, 4>, 1>, + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, BF16, BF16, F32, BF16, DsLayout, BF16, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding, 1,256, 64, 64, 32, 8, 8, 16, 16, 2, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 2, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, 1, 1, S<1, 32, 1, 4>, 2>, + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, BF16, BF16, F32, BF16, DsLayout, BF16, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding, 1,256, 64, 64, 32, 8, 8, 16, 16, 2, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 8, 1, 1, 1, S<1, 32, 1, 4>, 4>, + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, BF16, BF16, F32, BF16, DsLayout, BF16, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding, 1,256, 64, 64, 32, 8, 8, 16, 16, 2, 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, 4>, 8> // clang-format on >; @@ -140,7 +146,12 @@ using device_grouped_conv_fwd_xdl_f16_comp_instances = std::tuple< //########################################| Spatial| Layout| Layout| Layout| Layout| Type| Type| Type| DataType| DataType| Type| Elementwise| Elementwise| Elementwise| Specialization| Specialization| 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| //########################################| | | | | | | | | | | | 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| //########################################| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Intrawave, BlockGemmPipelineVersion::v4> + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Intrawave, BlockGemmPipelineVersion::v4>, + // mfma 16x16 + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, F16, F16, F32, F16, DsLayout, F16, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding, 1,256, 64, 64, 32, 8, 8, 16, 16, 2, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 2, 8, 1, 1, 1, S<1, 32, 1, 4>, 1>, + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, F16, F16, F32, F16, DsLayout, F16, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding, 1,256, 64, 64, 32, 8, 8, 16, 16, 2, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 2, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, 1, 1, S<1, 32, 1, 4>, 2>, + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, F16, F16, F32, F16, DsLayout, F16, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding, 1,256, 64, 64, 32, 8, 8, 16, 16, 2, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 8, 1, 1, 1, S<1, 32, 1, 4>, 4>, + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, F16, F16, F32, F16, DsLayout, F16, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding, 1,256, 64, 64, 32, 8, 8, 16, 16, 2, 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, 4>, 8> // clang-format on >; @@ -184,7 +195,11 @@ using device_grouped_conv_fwd_xdl_f32_comp_instances = std::tuple< DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 0, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Intrawave, BlockGemmPipelineVersion::v4>, DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Intrawave, BlockGemmPipelineVersion::v3>, DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Intrawave, BlockGemmPipelineVersion::v5>, - DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Interwave, BlockGemmPipelineVersion::v1> + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 0, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 0, 1, 1, S<1, 32, 1, 8>, 8, BlockGemmPipelineScheduler::Interwave, BlockGemmPipelineVersion::v1>, + // mfma 16x16 + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, F32, F32, F32, F32, DsLayout, F32, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding,1,256, 64, 64, 32, 8, 8, 16, 16, 2, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 2, 8, 1, 1, 1, S<1, 32, 1, 4>, 1>, + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, F32, F32, F32, F32, DsLayout, F32, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding,1,256, 64, 64, 32, 8, 8, 16, 16, 2, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 2, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 1, 8, 1, 1, 1, S<1, 32, 1, 4>, 2>, + DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle< NDimSpatial,ALayout,BLayout, DsLayout,ELayout, F32, F32, F32, F32, DsLayout, F32, PassThrough, PassThrough, PassThrough, ConvSpec, GemmMNKPadding,1,256, 64, 64, 32, 8, 8, 16, 16, 2, 2, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 8, 1, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 8, 1, 1, 1, S<1, 32, 1, 4>, 4> // clang-format on >; diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp index 638a3f98a3..d5eed7592e 100644 --- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp +++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp @@ -599,7 +599,6 @@ struct DeviceOperationInstanceFactory>>& instances); -void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_oddc_instances( - std::vector>>& instances); - void add_device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_instances( std::vector>>& instances); -void add_device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_oddc_instances( - std::vector>>& instances); - void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_i8_instances( std::vector>>& instances); -void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_i8_oddc_instances( - std::vector>>& instances); - void add_device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_i8_instances( std::vector>>& instances); -void add_device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_i8_oddc_instances( - std::vector>>& instances); #endif #ifdef CK_ENABLE_FP16 @@ -291,20 +236,6 @@ void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_1x1s1p0_instances( PassThrough, PassThrough>>>& instances); -void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_oddc_instances( - std::vector>>& instances); - void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_instances( std::vector>>& instances); -void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_oddc_instances( - std::vector>>& instances); - void add_device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_f16_instances( std::vector>>& instances); -void add_device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_f16_oddc_instances( - std::vector>>& instances); - void add_device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_instances( std::vector>>& instances); - -void add_device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_oddc_instances( - std::vector>>& instances); #endif } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/CMakeLists.txt b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/CMakeLists.txt index c1790901ec..3a101baac0 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/CMakeLists.txt +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/CMakeLists.txt @@ -87,8 +87,6 @@ add_instance_library(device_grouped_conv2d_fwd_instance wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_1x1p0_instance.cpp wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_1x1s1p0_instance.cpp wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_1x1s1p0_instance.cpp - wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_oddc_instance.cpp - wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_oddc_instance.cpp ## NHWGC, GKYXC, NHWGK wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_instance.cpp wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_instance.cpp @@ -96,6 +94,4 @@ add_instance_library(device_grouped_conv2d_fwd_instance wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_1x1p0_instance.cpp wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_1x1s1p0_instance.cpp wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_1x1s1p0_instance.cpp - wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_oddc_instance.cpp - wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_oddc_instance.cpp ) diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_oddc_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_oddc_instance.cpp deleted file mode 100644 index a8f723dfec..0000000000 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_oddc_instance.cpp +++ /dev/null @@ -1,40 +0,0 @@ -// SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. - -#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" -#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_instance.hpp" - -namespace ck { -namespace tensor_operation { -namespace device { -namespace instance { -// Compilation parameters for in[g, n, hi, wi, c] * wei[g, k, y, x, c] = out[g, n, ho, wo, k] -void add_device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_oddc_instances( - std::vector>>& instances) -{ - add_device_operation_instances(instances, - device_grouped_conv_fwd_wmma_f16_instances<2, - GNHWC, - GKYXC, - Empty_Tuple, - GNHWK, - Empty_Tuple, - PassThrough, - ConvFwdOddC>{}); -} - -} // namespace instance -} // namespace device -} // namespace tensor_operation -} // namespace ck diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_oddc_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_oddc_instance.cpp deleted file mode 100644 index 784a118897..0000000000 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_oddc_instance.cpp +++ /dev/null @@ -1,40 +0,0 @@ -// SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. - -#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" -#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_instance.hpp" - -namespace ck { -namespace tensor_operation { -namespace device { -namespace instance { -// Compilation parameters for in[g, n, hi, wi, c] * wei[g, k, y, x, c] = out[g, n, ho, wo, k] -void add_device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_oddc_instances( - std::vector>>& instances) -{ - add_device_operation_instances(instances, - device_grouped_conv_fwd_wmma_i8_instances<2, - GNHWC, - GKYXC, - Empty_Tuple, - GNHWK, - Empty_Tuple, - PassThrough, - ConvFwdOddC>{}); -} - -} // namespace instance -} // namespace device -} // namespace tensor_operation -} // namespace ck diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_oddc_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_oddc_instance.cpp deleted file mode 100644 index 8c621543a9..0000000000 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_oddc_instance.cpp +++ /dev/null @@ -1,40 +0,0 @@ -// SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. - -#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" -#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_instance.hpp" - -namespace ck { -namespace tensor_operation { -namespace device { -namespace instance { -// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k] -void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_oddc_instances( - std::vector>>& instances) -{ - add_device_operation_instances(instances, - device_grouped_conv_fwd_wmma_f16_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - Empty_Tuple, - PassThrough, - ConvFwdOddC>{}); -} - -} // namespace instance -} // namespace device -} // namespace tensor_operation -} // namespace ck diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_oddc_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_oddc_instance.cpp deleted file mode 100644 index 5cb313b3ca..0000000000 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_oddc_instance.cpp +++ /dev/null @@ -1,40 +0,0 @@ -// SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. - -#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" -#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_instance.hpp" - -namespace ck { -namespace tensor_operation { -namespace device { -namespace instance { -// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k] -void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_oddc_instances( - std::vector>>& instances) -{ - add_device_operation_instances(instances, - device_grouped_conv_fwd_wmma_i8_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - Empty_Tuple, - PassThrough, - ConvFwdOddC>{}); -} - -} // namespace instance -} // namespace device -} // namespace tensor_operation -} // namespace ck diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_2x_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_2x_instance.cpp index c078f8ed04..f5df7278d0 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_2x_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_2x_instance.cpp @@ -52,15 +52,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_2x_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances( - instances, - device_grouped_conv_fwd_xdl_bf16_comp_instances_2x<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp index a67b11f1cf..db048679bd 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp @@ -49,14 +49,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_bf16_comp_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_part2_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_part2_instance.cpp index 5c0391a25f..ee9507a80a 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_part2_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_part2_instance.cpp @@ -52,15 +52,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_comp_part2_instanc Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances( - instances, - device_grouped_conv_fwd_xdl_bf16_comp_instances_part2<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_2x_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_2x_instance.cpp index 726276c461..132d3c8411 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_2x_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_2x_instance.cpp @@ -52,15 +52,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_2x_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances( - instances, - device_grouped_conv_fwd_xdl_f16_comp_instances_2x<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_instance.cpp index 8b7bdec2a8..a7deb969ba 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_instance.cpp @@ -49,14 +49,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_f16_comp_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_part2_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_part2_instance.cpp index c66114b9a3..d2732547fa 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_part2_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_part2_instance.cpp @@ -52,15 +52,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_comp_part2_instance Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances( - instances, - device_grouped_conv_fwd_xdl_f16_comp_instances_part2<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_comp_instance.cpp index 93e07e08fb..8a0caebc9f 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_comp_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_comp_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_comp_instance.hpp" @@ -48,14 +48,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_comp_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_f32_comp_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_comp_instance.cpp index 6acbb7475c..e45df1e107 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_comp_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/comp/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_comp_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2024-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_comp_instance.hpp" @@ -50,14 +50,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_comp_instances( NHWGK, ConvFwd1x1S1P0>{}); - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_int8_comp_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); - if(ck::get_device_name() != "gfx950") { add_device_operation_instances( @@ -86,15 +78,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_comp_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances( - instances, - device_grouped_conv_fwd_xdl_int8_comp_instances_part2<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } if(ck::get_device_name() == "gfx950") @@ -125,15 +108,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_comp_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances( - instances, - device_grouped_conv_fwd_xdl_int8_comp_instances_2x<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp index 2afbfdc386..078221f89f 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_instance.hpp" @@ -46,14 +46,6 @@ void add_device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instances( Empty_Tuple, GNHWK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_bf16_instances<2, - GNHWC, - GKYXC, - Empty_Tuple, - GNHWK, - ConvFwdOddC>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp index 822ef51e00..3a481dd204 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_instance.hpp" @@ -46,14 +46,6 @@ void add_device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instances( Empty_Tuple, GNHWK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_f16_instances<2, - GNHWC, - GKYXC, - Empty_Tuple, - GNHWK, - ConvFwdOddC>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp index 79a1fb99a8..5add0f8add 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_instance.hpp" @@ -46,14 +46,6 @@ void add_device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instances( Empty_Tuple, GNHWK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_f32_instances<2, - GNHWC, - GKYXC, - Empty_Tuple, - GNHWK, - ConvFwdOddC>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp index e567c0df75..0257c7d315 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_instance.hpp" @@ -46,14 +46,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_bf16_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp index 3e42184996..2715506fe2 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_instance.hpp" @@ -46,14 +46,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_f16_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp index c035d4c3da..8d3e4d91b1 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_instance.hpp" @@ -46,14 +46,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_f32_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_instance.cpp index 5c425effd8..465fa927a5 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2024-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_instance.hpp" @@ -46,14 +46,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_instances( Empty_Tuple, NHWGK, ConvFwd1x1S1P0>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_int8_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp index e8a763c527..87423801cb 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_mem_instance.hpp" @@ -49,15 +49,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance NHWGK, ConvFwd1x1S1P0, Interwave>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_bf16_mem_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC, - Interwave>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp index 3ae3fb5186..ebb213461a 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_mem_instance.hpp" @@ -49,15 +49,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance NHWGK, ConvFwd1x1S1P0, Intrawave>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_bf16_mem_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC, - Intrawave>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_mem_inter_instance.cpp index cb7e912936..c2c8a099b2 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_mem_inter_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_mem_inter_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_mem_instance.hpp" @@ -49,15 +49,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_mem_inter_instances NHWGK, ConvFwd1x1S1P0, Interwave>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_f16_mem_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC, - Interwave>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_mem_intra_instance.cpp index d787f4b048..11cb853f0d 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_mem_intra_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_mem_intra_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_mem_instance.hpp" @@ -49,15 +49,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_mem_intra_instances NHWGK, ConvFwd1x1S1P0, Intrawave>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_f16_mem_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC, - Intrawave>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_inter_instance.cpp index 5644289790..1992d7f7c1 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_inter_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_inter_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_mem_instance.hpp" @@ -49,15 +49,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_inter_instances NHWGK, ConvFwd1x1S1P0, Interwave>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_f32_mem_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC, - Interwave>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_intra_instance.cpp index 5b12dad5a3..2b8fd3d9db 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_intra_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_intra_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_mem_instance.hpp" @@ -49,15 +49,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_mem_intra_instances NHWGK, ConvFwd1x1S1P0, Intrawave>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_f32_mem_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC, - Intrawave>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_inter_instance.cpp index f667481fa4..5579ec62cc 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_inter_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_inter_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_mem_instance.hpp" @@ -49,15 +49,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_inter_instance NHWGK, ConvFwd1x1S1P0, Interwave>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_int8_mem_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC, - Interwave>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_intra_instance.cpp index 2ff2c7f51f..77f3df2c11 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_intra_instance.cpp +++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/mem/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_intra_instance.cpp @@ -1,5 +1,5 @@ // SPDX-License-Identifier: MIT -// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved. +// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. #include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" #include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_mem_instance.hpp" @@ -49,15 +49,6 @@ void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_int8_mem_intra_instance NHWGK, ConvFwd1x1S1P0, Intrawave>{}); - - add_device_operation_instances(instances, - device_grouped_conv_fwd_xdl_int8_mem_instances<2, - NHWGC, - GKYXC, - Empty_Tuple, - NHWGK, - ConvFwdOddC, - Intrawave>{}); } } // namespace instance diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/CMakeLists.txt b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/CMakeLists.txt index 7b9ccf6609..eeea4aae6d 100644 --- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/CMakeLists.txt +++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/CMakeLists.txt @@ -60,10 +60,6 @@ set(GROUPED_CONV3D_FWD wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_i8_1x1s1p0_instance.cpp wmma/device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_f16_1x1s1p0_instance.cpp wmma/device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_i8_1x1s1p0_instance.cpp - wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_oddc_instance.cpp - wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_i8_oddc_instance.cpp - wmma/device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_f16_oddc_instance.cpp - wmma/device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_i8_oddc_instance.cpp ) if((DTYPES MATCHES "fp8" AND DTYPES MATCHES "fp16") OR NOT DEFINED DTYPES) diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_oddc_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_oddc_instance.cpp deleted file mode 100644 index fa378af1ee..0000000000 --- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_oddc_instance.cpp +++ /dev/null @@ -1,41 +0,0 @@ -// SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. - -#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" -#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_instance.hpp" - -namespace ck { -namespace tensor_operation { -namespace device { -namespace instance { -// Compilation parameters for in[g, n, di, hi, wi, c] * wei[g, k, z, y, x, c] = out[g, n, do, ho, -// wo, k] -void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_oddc_instances( - std::vector>>& instances) -{ - add_device_operation_instances(instances, - device_grouped_conv_fwd_wmma_f16_instances<3, - GNDHWC, - GKZYXC, - Empty_Tuple, - GNDHWK, - Empty_Tuple, - PassThrough, - ConvFwdOddC>{}); -} - -} // namespace instance -} // namespace device -} // namespace tensor_operation -} // namespace ck diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_i8_oddc_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_i8_oddc_instance.cpp deleted file mode 100644 index d41416fd4a..0000000000 --- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_i8_oddc_instance.cpp +++ /dev/null @@ -1,41 +0,0 @@ -// SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. - -#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" -#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_instance.hpp" - -namespace ck { -namespace tensor_operation { -namespace device { -namespace instance { -// Compilation parameters for in[g, n, di, hi, wi, c] * wei[g, k, z, y, x, c] = out[g, n, do, ho, -// wo, k] -void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_i8_oddc_instances( - std::vector>>& instances) -{ - add_device_operation_instances(instances, - device_grouped_conv_fwd_wmma_i8_instances<3, - GNDHWC, - GKZYXC, - Empty_Tuple, - GNDHWK, - Empty_Tuple, - PassThrough, - ConvFwdOddC>{}); -} - -} // namespace instance -} // namespace device -} // namespace tensor_operation -} // namespace ck diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_f16_oddc_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_f16_oddc_instance.cpp deleted file mode 100644 index 8a7bc26178..0000000000 --- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_f16_oddc_instance.cpp +++ /dev/null @@ -1,41 +0,0 @@ -// SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. - -#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" -#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_instance.hpp" - -namespace ck { -namespace tensor_operation { -namespace device { -namespace instance { -// Compilation parameters for in[n, di, hi, wi, g, c] * wei[g, k, z, y, x, c] = out[n, do, ho, wo, -// g, k] -void add_device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_f16_oddc_instances( - std::vector>>& instances) -{ - add_device_operation_instances(instances, - device_grouped_conv_fwd_wmma_f16_instances<3, - NDHWGC, - GKZYXC, - Empty_Tuple, - NDHWGK, - Empty_Tuple, - PassThrough, - ConvFwdOddC>{}); -} - -} // namespace instance -} // namespace device -} // namespace tensor_operation -} // namespace ck diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_i8_oddc_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_i8_oddc_instance.cpp deleted file mode 100644 index 7649f86971..0000000000 --- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_i8_oddc_instance.cpp +++ /dev/null @@ -1,41 +0,0 @@ -// SPDX-License-Identifier: MIT -// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved. - -#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp" -#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_instance.hpp" - -namespace ck { -namespace tensor_operation { -namespace device { -namespace instance { -// Compilation parameters for in[n, di, hi, wi, g, c] * wei[g, k, z, y, x, c] = out[n, do, ho, wo, -// g, k] -void add_device_grouped_conv3d_fwd_wmma_ndhwgc_gkzyxc_ndhwgk_i8_oddc_instances( - std::vector>>& instances) -{ - add_device_operation_instances(instances, - device_grouped_conv_fwd_wmma_i8_instances<3, - NDHWGC, - GKZYXC, - Empty_Tuple, - NDHWGK, - Empty_Tuple, - PassThrough, - ConvFwdOddC>{}); -} - -} // namespace instance -} // namespace device -} // namespace tensor_operation -} // namespace ck