From 86ac3213ed1b8d599cd6f9e5df3c3a4ae686e440 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bart=C5=82omiej=20Kocot?= Date: Wed, 11 Mar 2026 23:38:15 +0100 Subject: [PATCH] [CK][CK Tile] Improvements for grouped conv fwd tile profiling (#5114) ## Motivation Improve profiling for grouped convolution forward for better comparison between CK and CK Tile ## Technical Details - Include preprocessing time for ck tile - Add flush cache for conv fwd profiler - Switch configs to builder reflect - Add KPerXdl deduce - Add non-grouped ported instances ## Test Plan test_grouped_convnd_fwd_tile ## Test Result pass ## Submission Checklist - [x] Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests. AICK-786 --- .../ck_tile/builder/testing/conv/ck_tile.hpp | 22 +- .../configs/create_configs.sh | 9 +- .../configs/forward/profiler/ndhwgc_bf16.conf | 503 +++++++-------- .../configs/forward/profiler/ndhwgc_fp16.conf | 485 +++++++------- .../configs/forward/profiler/ndhwgc_fp32.conf | 353 ++++++----- .../configs/forward/profiler/nhwgc_bf16.conf | 593 ++++++++++-------- .../configs/forward/profiler/nhwgc_fp16.conf | 572 +++++++++-------- .../configs/forward/profiler/nhwgc_fp32.conf | 406 ++++++------ .../configs/forward/tests/ndhwgc_bf16.conf | 97 ++- .../configs/forward/tests/ndhwgc_fp16.conf | 96 ++- .../configs/forward/tests/ndhwgc_fp32.conf | 78 ++- .../configs/forward/tests/nhwgc_bf16.conf | 115 ++-- .../configs/forward/tests/nhwgc_fp16.conf | 113 ++-- .../configs/forward/tests/nhwgc_fp32.conf | 88 +-- .../generate_instances.py | 162 ++--- include/ck/host_utility/flush_cache.hpp | 1 + include/ck/host_utility/kernel_launch.hpp | 125 ++++ ...ped_conv_fwd_multiple_abd_xdl_cshuffle.hpp | 199 ++++-- ..._conv_fwd_multiple_abd_xdl_cshuffle_v3.hpp | 43 +- ...d_multiple_d_xdl_large_tensor_cshuffle.hpp | 43 +- include/ck_tile/host/kernel_launch.hpp | 80 ++- .../grouped_convolution_forward_tile_algs.hpp | 10 +- .../profile_grouped_conv_fwd_impl.hpp | 45 +- .../src/profile_grouped_conv_fwd_tile.cpp | 11 +- 24 files changed, 2375 insertions(+), 1874 deletions(-) diff --git a/experimental/builder/include/ck_tile/builder/testing/conv/ck_tile.hpp b/experimental/builder/include/ck_tile/builder/testing/conv/ck_tile.hpp index f383feabf0..da771dfd19 100644 --- a/experimental/builder/include/ck_tile/builder/testing/conv/ck_tile.hpp +++ b/experimental/builder/include/ck_tile/builder/testing/conv/ck_tile.hpp @@ -9,6 +9,7 @@ #include "ck_tile/host/kernel_launch.hpp" #include "ck_tile/ops/gemm.hpp" #include "ck_tile/ops/grouped_convolution.hpp" +#include "ck_tile/host.hpp" #include #include @@ -59,7 +60,8 @@ template ()); - auto preprocess = [&]() { + + auto preprocess = [&]() { if constexpr(ConvDirectionIsBackwardWeight) { if(args.k_batch > 1) @@ -73,10 +75,20 @@ template (conv, grids, blocks, 0, kargs))); + if(s_conf.flush_cache_) + { + return RunResult::from_runtime(ck_tile::launch_kernel_time_mask_flush_cache( + s_conf, + preprocess, + ck_tile::make_kernel(conv, grids, blocks, 0, kargs))); + } + else + { + return RunResult::from_runtime(ck_tile::launch_kernel_time_mask( + s_conf, + preprocess, + ck_tile::make_kernel(conv, grids, blocks, 0, kargs))); + } } } // namespace detail diff --git a/experimental/grouped_convolution_tile_instances/configs/create_configs.sh b/experimental/grouped_convolution_tile_instances/configs/create_configs.sh index b268328d2e..5ddcb6625c 100755 --- a/experimental/grouped_convolution_tile_instances/configs/create_configs.sh +++ b/experimental/grouped_convolution_tile_instances/configs/create_configs.sh @@ -68,7 +68,14 @@ mkdir -p forward/tests mkdir -p backward_weight/tests mkdir -p backward_data/tests -# Do not change the existing fwd test configs +# For FWD, generate new test configs by taking 20% of the profiler configs for each data type and layout +for layout in nhwgc ndhwgc; do + for dtype in fp32 fp16 bf16; do + profiler_config="forward/profiler/${layout}_${dtype}.conf" + test_config="forward/tests/${layout}_${dtype}.conf" + awk 'NR % 5 == 0' $profiler_config > $test_config # 20% of lines in the profiler configs + done +done # For BWD weight, generate new test configs by taking 20% of the profiler configs for each data type and layout for layout in nhwgc ndhwgc; do diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_bf16.conf b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_bf16.conf index e91dd97563..b0452e10d5 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_bf16.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_bf16.conf @@ -1,264 +1,239 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Default, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Default, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<256, 256, 128, 32, Default, 32, 32, 4, 2, 2, 2, 2, 1, 1> -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Default, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Filter1x1Pad0, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Filter1x1Stride1Pad0, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Default, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Filter1x1Pad0, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Filter1x1Stride1Pad0, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 64, 64, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 64, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 64, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 64, 64, Default, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,8),2,bf16,bf16,Default> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,32,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,Interwave,v1,bf16,bf16,false,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,256,128,32,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,Interwave,v1,bf16,bf16,false,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_fp16.conf b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_fp16.conf index dad7cdffd8..b71e88c6b5 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_fp16.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_fp16.conf @@ -1,255 +1,230 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Default, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Default, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<256, 256, 128, 32, Default, 32, 32, 4, 2, 2, 2, 2, 1, 1> -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Default, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Filter1x1Pad0, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Filter1x1Stride1Pad0, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Default, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Filter1x1Pad0, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Filter1x1Stride1Pad0, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> \ No newline at end of file +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,8),2,fp16,fp16,Default> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,32,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,Interwave,v1,fp16,fp16,false,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,256,128,32,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,Interwave,v1,fp16,fp16,false,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_fp32.conf b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_fp32.conf index 4bf7a0b238..abd2701dd9 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_fp32.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/ndhwgc_fp32.conf @@ -1,177 +1,176 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Default, 32, 32, 2, 1, 4, 4, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Default, 32, 32, 2, 2, 1, 1, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 16, Default, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 16, Default, 32, 32, 2, 4, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 16, Default, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Default, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 16, Default, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 16, Default, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Default, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 16, Default, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 16, Default, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 16, Default, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 16, Default, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Default, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 16, Default, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 192, 16, Default, 32, 32, 2, 3, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 16, Filter1x1Pad0, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 16, Filter1x1Pad0, 32, 32, 2, 4, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 16, Filter1x1Pad0, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 16, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 16, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 16, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 16, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 16, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 16, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 16, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 192, 16, Filter1x1Pad0, 32, 32, 2, 3, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 16, Filter1x1Stride1Pad0, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 16, Filter1x1Stride1Pad0, 32, 32, 2, 4, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 16, Filter1x1Stride1Pad0, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 16, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 16, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 16, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 16, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 16, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 192, 16, Filter1x1Stride1Pad0, 32, 32, 2, 3, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 16, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<256, 256, 128, 16, Default, 32, 32, 4, 2, 4, 4, 4, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Default, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> \ No newline at end of file +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,16,4,4,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,16,4,4,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,128,16,4,4,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,64,128,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,128,16,4,4,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,32,16,4,4,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,32,128,16,4,4,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,32,64,16,4,4,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,192,16,4,4,32,32,2,3,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,256,128,16,4,4,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,256,16,4,4,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,128,16,4,4,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,64,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,64,128,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,128,16,4,4,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,32,16,4,4,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,32,128,16,4,4,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,32,64,16,4,4,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,192,16,4,4,32,32,2,3,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,256,128,16,4,4,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,256,16,4,4,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,128,16,4,4,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,64,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,64,128,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,128,16,4,4,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,32,16,4,4,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,32,128,16,4,4,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,32,64,16,4,4,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,192,16,4,4,32,32,2,3,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,16,4,4,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_bf16.conf b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_bf16.conf index 15787ad3c3..66e325510a 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_bf16.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_bf16.conf @@ -1,264 +1,329 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Default, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Default, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<256, 256, 128, 32, Default, 32, 32, 4, 2, 2, 2, 2, 1, 1> -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v5> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 64, 64, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 64, 64, Default, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v5> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 64, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v5> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 64, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 64, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Default, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Default, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v5> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Filter1x1Pad0, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Filter1x1Pad0, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v5> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Filter1x1Stride1Pad0, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Filter1x1Stride1Pad0, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v5> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,8,4,4,32,32,2,1,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,64,8,4,4,32,32,4,1,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,8,4,4,32,32,2,2,Seq(2,16,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,16,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,64,64,8,4,4,32,32,1,2,Seq(2,16,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,16,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,8),2,bf16,bf16,Default> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,32,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,Interwave,v1,bf16,bf16,false,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,256,128,32,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,Interwave,v1,bf16,bf16,false,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_fp16.conf b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_fp16.conf index 99457320b0..8448c32117 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_fp16.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_fp16.conf @@ -1,255 +1,317 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Default, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Default, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 32, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<256, 256, 128, 32, Default, 32, 32, 4, 2, 2, 2, 2, 1, 1> -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Default, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Default, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Default, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Filter1x1Pad0, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Filter1x1Pad0, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Filter1x1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 32, 32, 4, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 224, 256, 64, Filter1x1Stride1Pad0, 16, 16, 7, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 224, 64, Filter1x1Stride1Pad0, 16, 16, 8, 7, 8, 8, 8, 2, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 128, 32, Filter1x1Stride1Pad0, 32, 32, 4, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Default, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 16, 256, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 32, 256, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 64, 64, Default, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 128, 64, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> \ No newline at end of file +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,8,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,2,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,8,4,4,32,32,2,1,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,64,8,4,4,32,32,4,1,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,8,4,4,32,32,2,2,Seq(2,16,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,16,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,64,64,8,4,4,32,32,1,2,Seq(2,16,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,16,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,8),2,fp16,fp16,Default> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,32,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,Interwave,v1,fp16,fp16,false,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,256,128,32,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,Interwave,v1,fp16,fp16,false,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_fp32.conf b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_fp32.conf index 4bf7a0b238..f667df52a8 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_fp32.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/profiler/nhwgc_fp32.conf @@ -1,177 +1,229 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Default, 32, 32, 2, 1, 4, 4, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Default, 32, 32, 2, 2, 1, 1, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 16, Default, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 16, Default, 32, 32, 2, 4, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 16, Default, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Default, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 16, Default, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 16, Default, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Default, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 16, Default, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 16, Default, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 16, Default, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 16, Default, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Default, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 16, Default, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 192, 16, Default, 32, 32, 2, 3, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 16, Filter1x1Pad0, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 16, Filter1x1Pad0, 32, 32, 2, 4, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 16, Filter1x1Pad0, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 16, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 16, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 16, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 16, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 16, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 16, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 16, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 192, 16, Filter1x1Pad0, 32, 32, 2, 3, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 256, 128, 16, Filter1x1Stride1Pad0, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 16, Filter1x1Stride1Pad0, 32, 32, 2, 4, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 128, 16, Filter1x1Stride1Pad0, 32, 32, 4, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 128, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 64, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 64, 128, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 16, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 64, 16, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 128, 16, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 128, 32, 16, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<128, 32, 128, 16, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 32, 16, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 16, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 192, 16, Filter1x1Stride1Pad0, 32, 32, 2, 3, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 1, 2, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 4, 4, 4, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 16, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<256, 256, 128, 16, Default, 32, 32, 4, 2, 4, 4, 4, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Default, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Default, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Default, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Default, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Default, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Default, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Default, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Default, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Default, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Pad0, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Pad0, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Pad0, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Pad0, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Pad0, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Pad0, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 128, 16, 64, Filter1x1Stride1Pad0, 16, 16, 4, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 32, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 64, 16, 64, Filter1x1Stride1Pad0, 16, 16, 2, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 64, 64, Filter1x1Stride1Pad0, 16, 16, 1, 2, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 64, 64, Filter1x1Stride1Pad0, 32, 32, 1, 1, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 128, 64, Filter1x1Stride1Pad0, 16, 16, 1, 4, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 64, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> \ No newline at end of file +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,16,4,4,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,16,4,4,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,128,16,4,4,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,64,128,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,128,16,4,4,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,32,16,4,4,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,32,128,16,4,4,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,32,64,16,4,4,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,192,16,4,4,32,32,2,3,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,256,128,16,4,4,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,256,16,4,4,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,128,16,4,4,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,64,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,64,128,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,128,16,4,4,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,32,16,4,4,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,32,128,16,4,4,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,32,64,16,4,4,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,192,16,4,4,32,32,2,3,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,256,128,16,4,4,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,256,16,4,4,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,128,16,4,4,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,64,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,64,128,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,128,16,4,4,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,32,16,4,4,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,32,128,16,4,4,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,32,64,16,4,4,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,192,16,4,4,32,32,2,3,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,256,128,16,4,4,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,256,16,4,4,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,128,128,16,4,4,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,128,64,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,64,128,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,64,128,16,4,4,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,128,32,16,4,4,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,32,128,16,4,4,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,64,32,64,16,4,4,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,128,192,16,4,4,32,32,2,3,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,256,128,16,4,4,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_bf16.conf b/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_bf16.conf index 73a3bd0357..633c7e2f85 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_bf16.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_bf16.conf @@ -1,50 +1,47 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> \ No newline at end of file +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_fp16.conf b/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_fp16.conf index 73a3bd0357..37bf2e8611 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_fp16.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_fp16.conf @@ -1,50 +1,46 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> \ No newline at end of file +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,16> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_fp32.conf b/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_fp32.conf index 0f66f1f48f..8c081de10a 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_fp32.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/tests/ndhwgc_fp32.conf @@ -1,43 +1,35 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 32, Default, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 32, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> \ No newline at end of file +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,16,4,4,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,64,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,32,16,4,4,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,128,16,4,4,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,32,64,16,4,4,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_bf16.conf b/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_bf16.conf index 73a3bd0357..ea2a00e86c 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_bf16.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_bf16.conf @@ -1,50 +1,65 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> \ No newline at end of file +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,8,4,4,32,32,2,1,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,bf16,bf16,Default,1> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,bf16,bf16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,256,128,32,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,Interwave,v1,bf16,bf16,false,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_fp16.conf b/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_fp16.conf index 06ff7d20a4..cebad8227e 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_fp16.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_fp16.conf @@ -1,50 +1,63 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Default, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 32, 64, Filter1x1Stride1Pad0, 32, 32, 2, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Default, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -# DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<64, 16, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3_DirectLoad<256, 64, 64, 64, Filter1x1Stride1Pad0, 16, 16, 2, 2, 2, 2, 4, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,64,32,8,8,32,32,2,2,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,128,128,32,32,8,8,32,32,2,1,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(4,4,8),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,64,8,4,4,32,32,2,1,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,Seq(2,32,4),Seq(1,0,2),Seq(1,0,2),2,1,1,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,128,32,8,8,32,32,1,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,128,32,8,8,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,64,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,64,32,64,32,8,8,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,32,32,8,8,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,64,128,32,8,8,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,32,128,32,8,8,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,16,1,8),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,true,1,1,Seq(1,32,1,4),8,fp16,fp16,Default,1> +DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,32,8,8,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp16,fp16,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,256,128,32,32,8,8,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,8,true,1,1,Seq(1,32,1,4),4,Interwave,v1,fp16,fp16,false,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16,false,1> diff --git a/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_fp32.conf b/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_fp32.conf index 0f66f1f48f..76e0e3ebe7 100644 --- a/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_fp32.conf +++ b/experimental/grouped_convolution_tile_instances/configs/forward/tests/nhwgc_fp32.conf @@ -1,43 +1,45 @@ -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Default, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 64, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 1, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 32, 64, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 8, 8, 8, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Default, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 64, 64, 32, Filter1x1Stride1Pad0, 16, 16, 2, 2, 1, 2, 1, 1, 1, 1> -# LargeTensor is temporary disable due to failures -# DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor<64, 64, 64, 32, Default, 32, 32, 2, 2, 1, 1, 1, 1, 1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Default, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 8> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 16> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<64, 64, 16, 16, Filter3x3, 16, 16, 4, 1, 4, 1, 1, 1, 1, 32> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 32, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v4> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Default, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 256, 256, 32, Filter1x1Stride1Pad0, 16, 16, 8, 8, 8, 8, 8, 1, 2, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v3> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Default, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 128, 64, Filter1x1Stride1Pad0, 32, 32, 2, 2, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v6> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Default, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<256, 128, 256, 32, Filter1x1Stride1Pad0, 32, 32, 2, 4, 8, 8, 8, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Default, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 16, 32, 64, Default, 16, 16, 1, 1, 8, 8, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 16, 64, Filter1x1Stride1Pad0, 16, 16, 1, 1, 8, 8, 2, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v1> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Default, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<64, 16, 16, 128, Filter1x1Stride1Pad0, 16, 16, 1, 1, 4, 4, 4, 1, 1, BlkGemmPipelineScheduler: Interwave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 32, Default, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 32, Filter1x1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> -DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<128, 32, 128, 32, Filter1x1Stride1Pad0, 32, 32, 1, 2, 4, 4, 8, 1, 1, BlkGemmPipelineScheduler: Intrawave, BlkGemmPipelineVersion: v2> \ No newline at end of file +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,128,256,16,4,4,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,32,16,4,4,32,32,2,1,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,256,128,128,16,4,4,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,64,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,1,128,128,32,16,4,4,32,32,2,1,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,64,64,16,4,4,32,32,2,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,8,1,8),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,128,128,128,16,4,4,32,32,4,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,128,64,16,4,4,32,32,2,1,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,64,32,64,16,4,4,32,32,1,2,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,8),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,256,256,128,16,4,4,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,16,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,64,128,16,4,4,32,32,2,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,1,128,32,128,16,4,4,32,32,1,2,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,Seq(4,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,true,1,1,Seq(1,8,1,16),4,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,1,1,Seq(1,32,1,4),2,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,1,256,64,64,32,8,8,16,16,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,1,8,true,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,2,8,true,1,1,Seq(1,32,1,4),1,fp32,fp32,Default,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,8> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter3x3,MNKPadding,1,64,64,16,16,4,4,16,16,4,1,Seq(4,16,1),Seq(0,2,1),Seq(0,2,1),1,4,4,true,Seq(4,16,1),Seq(1,0,2),Seq(1,0,2),2,1,4,true,1,1,Seq(1,16,1,4),1,fp32,fp32,Default,32> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> +DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,PassThrough,OddC,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,false,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32,false,1> diff --git a/experimental/grouped_convolution_tile_instances/generate_instances.py b/experimental/grouped_convolution_tile_instances/generate_instances.py index 4bbe12f8c5..925c9fe700 100755 --- a/experimental/grouped_convolution_tile_instances/generate_instances.py +++ b/experimental/grouped_convolution_tile_instances/generate_instances.py @@ -108,6 +108,33 @@ def check_vectors(a_scalar_per_vector, b_scalar_per_vector, c_scalar_per_vector) return False return True +def parse_instance_string(instance_string): + """Parse instance string, treating Seq(...) as a single parameter.""" + params = [] + current_param = "" + paren_depth = 0 + + for char in instance_string: + if char == '(': + paren_depth += 1 + current_param += char + elif char == ')': + paren_depth -= 1 + current_param += char + elif char == ',' and paren_depth == 0: + # Only split on comma if we're not inside parentheses + params.append(current_param.strip()) + current_param = "" + else: + current_param += char + + # Add the last parameter + if current_param.strip(): + params.append(current_param.strip()) + + return params + + def generate_calls_inc(instances, problem_name, direction, filter_pattern): generate_dir = Path(__file__).resolve().parent output_dir = Path(f"{generate_dir}/instances/{direction}") @@ -168,69 +195,80 @@ def parse_fwd_instances(instances, problem_name): for instance_id, instance in enumerate(instances): if instance.find("#") != -1 or instance.find(";") != -1: continue - instance_args_list = instance[instance.find("<") + 1 : instance.find(">")] - args = instance_args_list.split(", ") + start = instance.index('<') + 1 + end = instance.rindex('>') + params_str = instance[start:end] + args = parse_instance_string(params_str) - block_size = int(args[0]) - m_per_block = int(args[1]) - n_per_block = int(args[2]) - k_per_block = int(args[3]) - spec = args[4] - m_per_xdl = int(args[5]) - n_per_xdl = int(args[6]) - m_xdl_per_wave = int(args[7]) - n_xdl_per_wave = int(args[8]) - a_scalar_per_vector = int(args[9]) - b_scalar_per_vector = int(args[10]) - c_scalar_per_vector = int(args[11]) - if len(args) == 15: - num_groups_to_merge = int(args[14]) - elif len(args) != 16 and len(args) != 14: - raise RuntimeError("wrong number of parameters") + is_v3_instance = instance.find("Xdl_CShuffle_V3") != -1 + split_image = instance.find("Large_Tensor") != -1 + + if is_v3_instance: + spec = args[14] + block_size = int(args[16]) + m_per_block = int(args[17]) + n_per_block = int(args[18]) + k_per_block = int(args[19]) + k1 = int(args[20]) + m_per_xdl = int(args[22]) + n_per_xdl = int(args[23]) + m_xdl_per_wave = int(args[24]) + n_xdl_per_wave = int(args[25]) + a_scalar_per_vector = int(args[30]) + b_scalar_per_vector = int(args[37]) + c_scalar_per_vector = int(args[43]) + scheduler = args[44] + pipeline_version = args[45] + direct_load = args[48] == "true" + num_groups_to_merge = int(args[49]) else: - num_groups_to_merge = 1 - split_image = instance.find("Large") != -1 - double_smem_buffer = instance.find("BlkGemmPipelineVersion: v4") != -1 + spec = args[14] + block_size = int(args[17]) + m_per_block = int(args[18]) + n_per_block = int(args[19]) + k_per_block = int(args[20]) + k1 = int(args[21]) + m_per_xdl = int(args[23]) + n_per_xdl = int(args[24]) + m_xdl_per_wave = int(args[25]) + n_xdl_per_wave = int(args[26]) + a_scalar_per_vector = int(args[31]) + b_scalar_per_vector = int(args[38]) + c_scalar_per_vector = int(args[44]) + scheduler = "Intrawave" + pipeline_version = "v1" + direct_load = 0 + num_groups_to_merge = 0 if split_image else int(args[48]) + + double_smem_buffer = pipeline_version == "v4" num_wave_groups = 1 - scheduler = ( - "Intrawave" if instance.find("BlkGemmPipelineScheduler") == -1 else args[14] - ) - pipeline_version = ( - "v1" if instance.find("BlkGemmPipelineVersion") == -1 else args[15] - ) # Replace pipeline if Direct Load - if instance.find("DirectLoad") != -1: - if instance.find("BlkGemmPipelineVersion: v1") != -1: + if direct_load: + if pipeline_version == "v1": pipeline_version = "ASYNC_V1" - elif instance.find("BlkGemmPipelineVersion: v4") != -1: + elif pipeline_version == "v4": pipeline_version = "ASYNC_V4" else: - raise RuntimeError("not supported pipeline for direct load") + raise RuntimeError(f"{pipeline_version} not supported pipeline for direct load") else: - pipeline_version = f"""V{pipeline_version[-1:]}""" + pipeline_version = pipeline_version.upper() m_warp = int(m_per_block / (m_per_xdl * m_xdl_per_wave)) n_warp = int(n_per_block / (n_per_xdl * n_xdl_per_wave)) warp_size = 64 k_warp = int(block_size / (warp_size * m_warp * n_warp)) dtype = get_dtype(problem_name) - # TODO: Make it more flexible - # k_per_xdl = f"ck_tile::get_k_warp_tile<{dtype}, {m_per_xdl}>()" - if dtype == "float": - if m_per_xdl == 32: - if instance.find("BlkGemmPipelineVersion") == -1: - k_per_xdl = 4 - else: - # Increase for universal gemm - k_per_xdl = 8 - else: - k_per_xdl = 8 - else: - if m_per_xdl == 32: - k_per_xdl = 16 - else: - k_per_xdl = 32 - k_per_xdl = min(k_per_xdl, k_per_block) + k_per_xdl = max(k1, get_k_mfma(dtype, m_per_xdl, n_per_xdl)) + + if split_image: + print(f"Skipping instance {instance_id} with split_image since it's not supported yet.") + continue + if pipeline_version == "V5": + print(f"Skipping instance {instance_id} with V5 since it's not supported yet.") + continue + if pipeline_version == "ASYNC_V4": + print(f"Skipping instance {instance_id} with ASYNC_V4 since it's not supported yet.") + continue conv = ConvInstanceTemplateParams( spec, @@ -250,32 +288,6 @@ def parse_fwd_instances(instances, problem_name): convs.append(conv) return convs -def parse_instance_string(instance_string): - """Parse instance string, treating Seq(...) as a single parameter.""" - params = [] - current_param = "" - paren_depth = 0 - - for char in instance_string: - if char == '(': - paren_depth += 1 - current_param += char - elif char == ')': - paren_depth -= 1 - current_param += char - elif char == ',' and paren_depth == 0: - # Only split on comma if we're not inside parentheses - params.append(current_param.strip()) - current_param = "" - else: - current_param += char - - # Add the last parameter - if current_param.strip(): - params.append(current_param.strip()) - - return params - def parse_bwd_weight_instances(instances, problem_name): convs = [] diff --git a/include/ck/host_utility/flush_cache.hpp b/include/ck/host_utility/flush_cache.hpp index 382601de05..2d051233e4 100644 --- a/include/ck/host_utility/flush_cache.hpp +++ b/include/ck/host_utility/flush_cache.hpp @@ -10,6 +10,7 @@ #include "ck/ck.hpp" #include "ck/utility/env.hpp" +#include "ck/utility/tuple.hpp" #include "ck/stream_config.hpp" #include "ck/host_utility/hip_check_error.hpp" #include "ck/utility/flush_icache.hpp" diff --git a/include/ck/host_utility/kernel_launch.hpp b/include/ck/host_utility/kernel_launch.hpp index c923ee0cbe..1da4f16ca3 100644 --- a/include/ck/host_utility/kernel_launch.hpp +++ b/include/ck/host_utility/kernel_launch.hpp @@ -9,6 +9,7 @@ #include "ck/utility/env.hpp" #include "ck/stream_config.hpp" #include "ck/host_utility/hip_check_error.hpp" +#include "ck/host_utility/flush_cache.hpp" namespace ck { @@ -170,6 +171,130 @@ float launch_and_time_kernel_with_preprocess(const StreamConfig& stream_config, #endif } +template +float launch_and_time_kernel_with_preprocess_flush_cache(const StreamConfig& stream_config, + PreProcessFunc preprocess, + F kernel, + dim3 grid_dim, + dim3 block_dim, + std::size_t lds_byte, + Args... args) +{ +#if CK_TIME_KERNEL + if(stream_config.time_kernel_) + { + auto run_flush_cache = [&]() { ck::utility::flush_icache(); }; + if(ck::EnvIsEnabled(CK_ENV(CK_LOGGING))) + { + printf("%s: grid_dim {%u, %u, %u}, block_dim {%u, %u, %u} \n", + __func__, + grid_dim.x, + grid_dim.y, + grid_dim.z, + block_dim.x, + block_dim.y, + block_dim.z); + + printf("Warm up %d times\n", stream_config.cold_niters_); + } + // Warm up + preprocess(); + for(int i = 0; i < stream_config.cold_niters_; ++i) + { + kernel<<>>(args...); + hip_check_error(hipGetLastError()); + } + float total_time = 0, flush_cache_total_time = 0; + const int nrepeat = stream_config.nrepeat_; + // Main timing loop + { + if(ck::EnvIsEnabled(CK_ENV(CK_LOGGING))) + { + printf("Start running %d times...\n", nrepeat); + } + hipEvent_t start, stop; + + hip_check_error(hipEventCreate(&start)); + hip_check_error(hipEventCreate(&stop)); + + hip_check_error(hipDeviceSynchronize()); + hip_check_error(hipEventRecord(start, stream_config.stream_id_)); + + for(int i = 0; i < nrepeat; ++i) + { + run_flush_cache(); + preprocess(); + kernel<<>>(args...); + hip_check_error(hipGetLastError()); + } + + hip_check_error(hipEventRecord(stop, stream_config.stream_id_)); + hip_check_error(hipEventSynchronize(stop)); + + hip_check_error(hipEventElapsedTime(&total_time, start, stop)); + + hip_check_error(hipEventDestroy(start)); + hip_check_error(hipEventDestroy(stop)); + } + // Flush cache timing loop + { + if(ck::EnvIsEnabled(CK_ENV(CK_LOGGING))) + { + printf("Profile flush cache %d times...\n", nrepeat); + } + hipEvent_t start, stop; + + hip_check_error(hipEventCreate(&start)); + hip_check_error(hipEventCreate(&stop)); + + hip_check_error(hipDeviceSynchronize()); + hip_check_error(hipEventRecord(start, stream_config.stream_id_)); + + for(int i = 0; i < nrepeat; ++i) + { + run_flush_cache(); + } + + hip_check_error(hipEventRecord(stop, stream_config.stream_id_)); + hip_check_error(hipEventSynchronize(stop)); + + hip_check_error(hipEventElapsedTime(&flush_cache_total_time, start, stop)); + + hip_check_error(hipEventDestroy(start)); + hip_check_error(hipEventDestroy(stop)); + } + // Exclude flush cache from result + return (total_time - flush_cache_total_time) / nrepeat; + } + else + { + preprocess(); + kernel<<>>(args...); + hip_check_error(hipGetLastError()); + + return 0; + } +#else + kernel<<>>(args...); + hip_check_error(hipGetLastError()); + + return 0; +#endif +} + +template +float launch_and_time_kernel_flush_cache(const StreamConfig& stream_config, + F kernel, + dim3 grid_dim, + dim3 block_dim, + std::size_t lds_byte, + Args... args) +{ + auto preprocess = [&]() {}; + return launch_and_time_kernel_with_preprocess_flush_cache( + stream_config, preprocess, kernel, grid_dim, block_dim, lds_byte, args...); +} + } // namespace ck #endif diff --git a/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp b/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp index ce9ef26724..452b0af6c4 100644 --- a/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp +++ b/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp @@ -1158,26 +1158,52 @@ struct DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle isMultiB, CTranspose>; - return launch_and_time_kernel( - stream_config, - kernel, - dim3(gdx, gdy, gdz), - dim3(BlockSize), - 0, - arg.p_as_grid_, - arg.p_bs_grid_, - arg.p_ds_grid_, - arg.p_e_grid_, - arg.a_element_op_, - arg.b_element_op_, - arg.cde_element_op_, - as_grid_desc_ak0_m_ak1, - bs_grid_desc_bk0_n_bk1, - arg.ds_grid_desc_mblock_mperblock_nblock_nperblock_, - arg.e_grid_desc_mblock_mperblock_nblock_nperblock_, - arg.block_2_etile_map_, - arg.compute_ptr_offset_of_groups_, - arg.compute_ptr_offset_of_n_); + if(stream_config.flush_cache) + { + return launch_and_time_kernel_flush_cache( + stream_config, + kernel, + dim3(gdx, gdy, gdz), + dim3(BlockSize), + 0, + arg.p_as_grid_, + arg.p_bs_grid_, + arg.p_ds_grid_, + arg.p_e_grid_, + arg.a_element_op_, + arg.b_element_op_, + arg.cde_element_op_, + as_grid_desc_ak0_m_ak1, + bs_grid_desc_bk0_n_bk1, + arg.ds_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.e_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.block_2_etile_map_, + arg.compute_ptr_offset_of_groups_, + arg.compute_ptr_offset_of_n_); + } + else + { + return launch_and_time_kernel( + stream_config, + kernel, + dim3(gdx, gdy, gdz), + dim3(BlockSize), + 0, + arg.p_as_grid_, + arg.p_bs_grid_, + arg.p_ds_grid_, + arg.p_e_grid_, + arg.a_element_op_, + arg.b_element_op_, + arg.cde_element_op_, + as_grid_desc_ak0_m_ak1, + bs_grid_desc_bk0_n_bk1, + arg.ds_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.e_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.block_2_etile_map_, + arg.compute_ptr_offset_of_groups_, + arg.compute_ptr_offset_of_n_); + } } else { @@ -1230,26 +1256,53 @@ struct DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle isMultiA, isMultiB, CTranspose>; - return launch_and_time_kernel( - stream_config, - kernel, - dim3(gdx, gdy, gdz), - dim3(BlockSize), - 0, - p_b_grid, - p_a_grid, - arg.p_ds_grid_, - p_e_grid, - arg.b_element_op_, - arg.a_element_op_, - arg.cde_element_op_, - arg.b_grid_desc_bk0_n_bk1_, - arg.a_grid_desc_ak0_m_ak1_, - arg.ds_grid_desc_mblock_mperblock_nblock_nperblock_, - arg.e_grid_desc_mblock_mperblock_nblock_nperblock_, - arg.block_2_etile_map_, - arg.compute_ptr_offset_of_groups_, - arg.compute_ptr_offset_of_n_); + + if(stream_config.flush_cache) + { + return launch_and_time_kernel_flush_cache( + stream_config, + kernel, + dim3(gdx, gdy, gdz), + dim3(BlockSize), + 0, + p_b_grid, + p_a_grid, + arg.p_ds_grid_, + p_e_grid, + arg.b_element_op_, + arg.a_element_op_, + arg.cde_element_op_, + arg.b_grid_desc_bk0_n_bk1_, + arg.a_grid_desc_ak0_m_ak1_, + arg.ds_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.e_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.block_2_etile_map_, + arg.compute_ptr_offset_of_groups_, + arg.compute_ptr_offset_of_n_); + } + else + { + return launch_and_time_kernel( + stream_config, + kernel, + dim3(gdx, gdy, gdz), + dim3(BlockSize), + 0, + p_b_grid, + p_a_grid, + arg.p_ds_grid_, + p_e_grid, + arg.b_element_op_, + arg.a_element_op_, + arg.cde_element_op_, + arg.b_grid_desc_bk0_n_bk1_, + arg.a_grid_desc_ak0_m_ak1_, + arg.ds_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.e_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.block_2_etile_map_, + arg.compute_ptr_offset_of_groups_, + arg.compute_ptr_offset_of_n_); + } } else { @@ -1274,26 +1327,52 @@ struct DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle isMultiB, CTranspose>; - return launch_and_time_kernel( - stream_config, - kernel, - dim3(gdx, gdy, gdz), - dim3(BlockSize), - 0, - p_a_grid, - p_b_grid, - arg.p_ds_grid_, - p_e_grid, - arg.a_element_op_, - arg.b_element_op_, - arg.cde_element_op_, - arg.a_grid_desc_ak0_m_ak1_, - arg.b_grid_desc_bk0_n_bk1_, - arg.ds_grid_desc_mblock_mperblock_nblock_nperblock_, - arg.e_grid_desc_mblock_mperblock_nblock_nperblock_, - arg.block_2_etile_map_, - arg.compute_ptr_offset_of_groups_, - arg.compute_ptr_offset_of_n_); + if(stream_config.flush_cache) + { + return launch_and_time_kernel_flush_cache( + stream_config, + kernel, + dim3(gdx, gdy, gdz), + dim3(BlockSize), + 0, + p_a_grid, + p_b_grid, + arg.p_ds_grid_, + p_e_grid, + arg.a_element_op_, + arg.b_element_op_, + arg.cde_element_op_, + arg.a_grid_desc_ak0_m_ak1_, + arg.b_grid_desc_bk0_n_bk1_, + arg.ds_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.e_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.block_2_etile_map_, + arg.compute_ptr_offset_of_groups_, + arg.compute_ptr_offset_of_n_); + } + else + { + return launch_and_time_kernel( + stream_config, + kernel, + dim3(gdx, gdy, gdz), + dim3(BlockSize), + 0, + p_a_grid, + p_b_grid, + arg.p_ds_grid_, + p_e_grid, + arg.a_element_op_, + arg.b_element_op_, + arg.cde_element_op_, + arg.a_grid_desc_ak0_m_ak1_, + arg.b_grid_desc_bk0_n_bk1_, + arg.ds_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.e_grid_desc_mblock_mperblock_nblock_nperblock_, + arg.block_2_etile_map_, + arg.compute_ptr_offset_of_groups_, + arg.compute_ptr_offset_of_n_); + } } } }; diff --git a/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle_v3.hpp b/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle_v3.hpp index ae02f5ee77..5cff4cf7f2 100644 --- a/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle_v3.hpp +++ b/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle_v3.hpp @@ -26,7 +26,6 @@ #include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_utils.hpp" #include "ck/host_utility/device_prop.hpp" #include "ck/host_utility/kernel_launch.hpp" -#include "ck/host_utility/flush_cache.hpp" #include "ck/host_utility/io.hpp" #ifdef CK_EXPERIMENTAL_BUILDER #include "ck_tile/builder/reflect/conv_describe.hpp" @@ -1049,35 +1048,19 @@ struct DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3 const auto Run = [&](const auto& kernel) { if(stream_config.flush_cache) { - typename GridwiseGemm::Argument gemm_arg_ = gemm_arg; - ck::utility::RotatingMemWrapper rotating_mem( - gemm_arg_, - stream_config.rotating_count, - gemm_arg_.M * gemm_arg_.K * sizeof(ADataType), - gemm_arg_.K * gemm_arg_.N * sizeof(BDataType)); - rotating_mem.Print(); - - auto run_flush_cache = [&]() { - // flush icache - ck::utility::flush_icache(); - // rotating mem - rotating_mem.Next(); - }; - - ave_time += ck::utility::launch_and_time_kernel_with_preprocess( - stream_config, - run_flush_cache, - kernel, - dim3(gdx, gdy, gdz), - dim3(BlockSize), - 0, - gemm_arg_, - arg.a_grid_desc_ak0_m_ak1_, - arg.b_grid_desc_bk0_n_bk1_, - arg.ds_grid_desc_m_n_, - arg.e_grid_desc_m_n_, - arg.compute_ptr_offset_of_groups_, - arg.compute_ptr_offset_of_n_); + ave_time += + launch_and_time_kernel_flush_cache(stream_config, + kernel, + dim3(gdx, gdy, gdz), + dim3(BlockSize), + 0, + gemm_arg, + arg.a_grid_desc_ak0_m_ak1_, + arg.b_grid_desc_bk0_n_bk1_, + arg.ds_grid_desc_m_n_, + arg.e_grid_desc_m_n_, + arg.compute_ptr_offset_of_groups_, + arg.compute_ptr_offset_of_n_); } else { diff --git a/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_large_tensor_cshuffle.hpp b/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_large_tensor_cshuffle.hpp index 5e8df5e04e..d9e63e3ada 100644 --- a/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_large_tensor_cshuffle.hpp +++ b/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_large_tensor_cshuffle.hpp @@ -759,19 +759,36 @@ struct DeviceGroupedConvFwdMultipleD_Xdl_CShuffle_Large_Tensor CDEElementwiseOperation, ComputePtrOffsetOfStridedBatch, has_main_loop>; - - return launch_and_time_kernel(stream_config, - kernel, - dim3(gdx, gdy, gdz), - dim3(BlockSize), - 0, - arg.gemm_desc_kernel_args_, - arg.gemms_count_, - arg.a_element_op_, - arg.b_element_op_, - arg.cde_element_op_, - arg.compute_ptr_offset_of_groups_, - arg.compute_ptr_offset_of_n_); + if(stream_config.flush_cache) + { + return launch_and_time_kernel_flush_cache(stream_config, + kernel, + dim3(gdx, gdy, gdz), + dim3(BlockSize), + 0, + arg.gemm_desc_kernel_args_, + arg.gemms_count_, + arg.a_element_op_, + arg.b_element_op_, + arg.cde_element_op_, + arg.compute_ptr_offset_of_groups_, + arg.compute_ptr_offset_of_n_); + } + else + { + return launch_and_time_kernel(stream_config, + kernel, + dim3(gdx, gdy, gdz), + dim3(BlockSize), + 0, + arg.gemm_desc_kernel_args_, + arg.gemms_count_, + arg.a_element_op_, + arg.b_element_op_, + arg.cde_element_op_, + arg.compute_ptr_offset_of_groups_, + arg.compute_ptr_offset_of_n_); + } }; if(GridwiseGemm::CalculateHasMainKBlockLoop(K)) diff --git a/include/ck_tile/host/kernel_launch.hpp b/include/ck_tile/host/kernel_launch.hpp index 9b3fb88c16..6b7bf1b653 100644 --- a/include/ck_tile/host/kernel_launch.hpp +++ b/include/ck_tile/host/kernel_launch.hpp @@ -10,6 +10,8 @@ #include "ck_tile/host/hip_check_error.hpp" #include "ck_tile/host/stream_config.hpp" #include "ck_tile/host/timer.hpp" +#include "ck_tile/host/flush_icache.hpp" +#include "ck_tile/host/rotating_buffers.hpp" #include #include @@ -124,6 +126,47 @@ preprocess_profiling_impl(TimerType timer, const stream_config& s, PreprocessFun return timer.duration() / s.nrepeat_; } +template +CK_TILE_HOST double timing_loop_flush_cache_impl(TimerType timer, + const stream_config& s, + CallablesFunc&& callables_func, + PreprocessFunc preprocess = nullptr) +{ + auto run_flush_cache = [&]() { ck_tile::flush_icache(); }; + // Warm up + for(int i = 0; i < s.cold_niters_; i++) + { + if constexpr(!std::is_same_v) + { + preprocess(); + } + callables_func(); + } + // Main timing loop + int i = 0; + timer.start(s.stream_id_); + while(i < s.nrepeat_) + { + run_flush_cache(); + if constexpr(!std::is_same_v) + { + preprocess(); + } + + callables_func(); + i++; + } + timer.stop(s.stream_id_); + // Flush cache timing loop + auto flush_cache_time = preprocess_profiling_impl(gpu_timer{}, s, run_flush_cache); + if(i == 0) + { + return 0.; + } + // Exclude flush cache from result + return (timer.duration() / s.nrepeat_) - flush_cache_time; +} + template CK_TILE_HOST double timing_loop_impl(TimerType timer, const stream_config& s, @@ -138,12 +181,6 @@ CK_TILE_HOST double timing_loop_impl(TimerType timer, } callables_func(); } - // Only profile preprocess if it's provided - auto preprocess_time = 0.0; - if constexpr(!std::is_same_v) - { - preprocess_time = preprocess_profiling_impl(gpu_timer{}, s, preprocess); - } int i = 0; timer.start(s.stream_id_); @@ -159,9 +196,9 @@ CK_TILE_HOST double timing_loop_impl(TimerType timer, } timer.stop(s.stream_id_); - if(!i) + if(i == 0) return 0.; - return (timer.duration() / s.nrepeat_) - preprocess_time; + return timer.duration() / s.nrepeat_; } // clang-format off @@ -238,4 +275,31 @@ launch_kernel_time_mask(const stream_config& s, PreprocessFunc preprocess, Calla return timing_loop_impl(cpu_timer{}, s, callables_func, preprocess); } } + +template +CK_TILE_HOST float launch_kernel_time_mask_flush_cache(const stream_config& s, + PreprocessFunc preprocess, + Callables&&... callables) +{ + static_assert(sizeof...(callables) > 0, "At least one callable is required!"); + + if(!s.time_kernel_) + { + preprocess(); + launch_and_check(s, std::forward(callables)...); + return 0; + } + + auto callables_func = [&]() { launch_and_check(s, std::forward(callables)...); }; + + if(s.is_gpu_timer_) + { + return timing_loop_flush_cache_impl(gpu_timer{}, s, callables_func, preprocess); + } + else + { + return timing_loop_flush_cache_impl(cpu_timer{}, s, callables_func, preprocess); + } +} + } // namespace ck_tile diff --git a/profiler/include/profiler/grouped_convolution_forward_tile_algs.hpp b/profiler/include/profiler/grouped_convolution_forward_tile_algs.hpp index a9331d6456..054da8057a 100644 --- a/profiler/include/profiler/grouped_convolution_forward_tile_algs.hpp +++ b/profiler/include/profiler/grouped_convolution_forward_tile_algs.hpp @@ -65,7 +65,9 @@ run_grouped_conv_forward_tile_algs(const ckt::Args& args, const ckt::Outputs& outputs, const ck_tile::stream_config& s_conf) { - float best_avg_time = std::numeric_limits::max(); + // Run first instance as dummy to get proper time from the first instance + bool dummy_run_executed = false; + float best_avg_time = std::numeric_limits::max(); std::string best_op_name, op_name; bool is_supported; float avg_time; @@ -84,6 +86,12 @@ run_grouped_conv_forward_tile_algs(const ckt::Args& args, auto ref_conv = ReferenceInstance{}; auto ref_result = ckt::run(ref_conv, args, inputs, reference.get()); auto run_alg = [&](auto&& run_alg_func) { + if(!dummy_run_executed) + { + // Run first instance twice + std::tie(is_supported, avg_time, op_name) = run_alg_func(args, inputs, outputs, s_conf); + dummy_run_executed = true; + } std::tie(is_supported, avg_time, op_name) = run_alg_func(args, inputs, outputs, s_conf); if(is_supported) { diff --git a/profiler/include/profiler/profile_grouped_conv_fwd_impl.hpp b/profiler/include/profiler/profile_grouped_conv_fwd_impl.hpp index 3f14df086a..bf8da55f3c 100644 --- a/profiler/include/profiler/profile_grouped_conv_fwd_impl.hpp +++ b/profiler/include/profiler/profile_grouped_conv_fwd_impl.hpp @@ -65,7 +65,19 @@ void print_instances() for(const auto& op_ptr : op_ptrs) { +#ifdef CK_EXPERIMENTAL_BUILDER + const auto& instance_str = op_ptr->GetInstanceString(); + if(!instance_str.empty()) + { + std::cout << instance_str << std::endl; + } + else + { + std::cout << op_ptr->GetTypeString() << std::endl; + } +#else std::cout << op_ptr->GetTypeString() << std::endl; +#endif } } } // namespace fwd @@ -298,8 +310,13 @@ bool profile_grouped_conv_fwd_impl(int do_verification, auto invoker_ptr = op_ptr->MakeInvokerPointer(); - float avg_time = - invoker_ptr->Run(argument_ptr.get(), StreamConfig{nullptr, time_kernel}); + float avg_time = invoker_ptr->Run(argument_ptr.get(), + StreamConfig{nullptr, + time_kernel, + 0 /*log_level*/, + 5 /*cold_iters*/, + 50 /*nrepeat_*/, + time_kernel /*flush_cache*/}); std::size_t flop = conv_param.GetFlops(); std::size_t num_btype = conv_param.GetByte(); @@ -420,6 +437,30 @@ bool profile_grouped_conv_fwd_impl(int do_verification, std::cout << "\nValid instances for this problem:" << std::endl; } + // Run first instance twice to get proper time + { + auto argument_ptr = op_ptrs[0]->MakeArgumentPointer(in_device_buf.GetDeviceBuffer(), + wei_device_buf.GetDeviceBuffer(), + {}, + out_device_buf.GetDeviceBuffer(), + a_g_n_c_wis_lengths, + a_g_n_c_wis_strides, + b_g_k_c_xs_lengths, + b_g_k_c_xs_strides, + {}, + {}, + e_g_n_k_wos_lengths, + e_g_n_k_wos_strides, + conv_filter_strides, + conv_filter_dilations, + input_left_pads, + input_right_pads, + in_element_op, + wei_element_op, + out_element_op); + + run_impl(op_ptrs[0], argument_ptr); + } for(auto& op_ptr : op_ptrs) { auto argument_ptr = op_ptr->MakeArgumentPointer(in_device_buf.GetDeviceBuffer(), diff --git a/profiler/src/profile_grouped_conv_fwd_tile.cpp b/profiler/src/profile_grouped_conv_fwd_tile.cpp index 6980e8ae10..413842f46f 100644 --- a/profiler/src/profile_grouped_conv_fwd_tile.cpp +++ b/profiler/src/profile_grouped_conv_fwd_tile.cpp @@ -97,7 +97,16 @@ int call_profiler(const ckt::Args& args, bool time_kernel) std::string op_name; bool valid; std::tie(valid, avg_time, op_name) = ckp::run_grouped_conv_forward_tile_algs( - args, inputs.get(), outputs.get(), ck_tile::stream_config{nullptr, time_kernel, 0, 5, 50}); + args, + inputs.get(), + outputs.get(), + ck_tile::stream_config{nullptr, + time_kernel, + 0 /*log_level*/, + 5 /*cold_iters*/, + 50 /*nrepeat_*/, + true /*is_gpu_timer_*/, + time_kernel /*flush_cache*/}); if(time_kernel) { std::cout << "Best configuration parameters:" << "\nname: " << op_name